CN116584939A

CN116584939A - Delirium risk monitoring device and system

Info

Publication number: CN116584939A
Application number: CN202310232528.0A
Authority: CN
Inventors: 张山; 吴瑛; 杨芳宇; 范环; 杨雪; 韩媛; 李宁; 李超群
Original assignee: Capital Medical University
Current assignee: Capital Medical University
Priority date: 2020-05-15
Filing date: 2020-05-15
Publication date: 2023-08-15
Also published as: CN111568445A; CN111568445B

Abstract

The invention relates to a delirium risk monitoring device and a delirium risk monitoring system. The apparatus includes an identification module. The identification module calculates based on third-party judgment information by using the delirium evaluation model to obtain an evaluation value about delirium characteristics obtained by rapidly evaluating delirium consciousness blur of the object to be evaluated, so as to determine delirium labels of the object to be evaluated or delirium labels of the object to be evaluated do not occur based on the evaluation value of the delirium characteristics. According to the scheme, the third party judging information can be corresponding to the answer input by the object to be evaluated, so that the third party judging information can be used as an auxiliary evidence to verify the input answer again, the evaluation of delirium characteristics of the object to be evaluated is realized, the evaluation result is generated according to the actual display reaction and state of the object to be evaluated in the evaluation process, the answer with strong subjectivity and sensory deviation input by auxiliary personnel can be effectively corrected and prompted, and the improvement of delirium characteristic evaluation accuracy and reliability is facilitated.

Description

Delirium risk monitoring device and system

The original basis of the divisional application is a patent application with the application number of CN202010417659.2, the application date of 2020, and the application date of 05, 15, and the invention is named as a spectrum delusion risk monitoring method and system based on a deliberate dynamic prediction model.

Technical Field

The present invention relates to the field of delirium care, and more particularly, to a delirium risk monitoring device and system.

Background

Delirium (delirium) is a group of acute cognitive impairment syndromes with attention deficit, confusion, altered states of consciousness, which is an acute or subacute illness that typically varies over hours to days, identification requiring brief cognitive screening and acute clinical observation, primary diagnostic features include altered states of consciousness, inattention, impaired levels of consciousness, cognitive impairment (e.g., disorientation, dysmnesia) of acute onset and fluctuation. Delirium occurring in patients in intensive care units (Intensive Care Unit, ICU) is often referred to clinically as ICU delirium, and it is reported in literature that 14% -24% of hospitalized patients develop delirium during hospitalization, with the incidence of delirium in elderly patients up to 30% -50% and ICU delirium up to 35% -80%.

Although the incidence of delirium is not so high, ICU patients have been mistaken for "ICU psychosis" after delirium occurrence, which is not regarded as major, nor important, and this neglect may cause a number of adverse consequences. Delirium is a complex neurological syndrome and is associated with a variety of adverse outcomes, such as: increased medical costs, prolonged hospital stay, cognitive impairment, reduced independence, increased complications, reduced cumulative survival, prolonged post-operative recovery time, increased post-operative mortality, and the like. Studies by troghlic Z, etc., have found that delirium occurs more complicated and more difficult patient treatment, and also may lead to permanent irreversible brain damage. If the patient is in delirium state for a long time, the potential organ dysfunction can be caused, the incidence rate is about 70% -92%, the risks of aspiration and iatrogenic pneumonia are increased by 10 times, the incidence rate of complications such as pulmonary embolism and pressure sore is greatly increased, the mechanical ventilation patient is difficult to take off line, and the patient is subjected to tracheal intubation again after accidental tube drawing or tube drawing, finally the ICU hospitalization time is prolonged, and the death rate is increased. Studies have shown that an increase in the average hospitalization time of delirium-free patients by 8 days, once delirium occurs, increases the patient's time to support survival with a ventilator, residence time in the ICU, and patient hospitalization time; according to related studies, 75.7% of patients with delirium are discharged from the hospital and cognitive disorder still exists, and the aged often predicts poor prognosis once delirium occurs, including impaired overall function, inability to live alone and extreme need to be cared for; according to related researches, compared with a patient without delirium, the occurrence rate of pulmonary complications of the patient with postoperative delirium also tends to be increased, the probability of further recuperation after operation is increased, the occurrence of ICU delirium can promote the risk of iatrogenic pneumonia to be increased by 10 times, and simultaneously, the patient with mechanical ventilation can be caused to have the conditions of accidental tube drawing, secondary tracheal intubation, off-line difficulty and the like.

However, delirium is prone to misdiagnosis or missed diagnosis due to atypical early symptoms, hidden onset of the disease, and often lack of knowledge and knowledge of the disease by nonprofessional psychiatrists, and often fails to give immediate attention and treatment.

The prior art systems and methods for predicting, screening and monitoring brain diseases/delirium, as disclosed in the patent document publication number CN109069081a, detect the presence of diffuse slowing (signs of brain attacks) in a patient's brain waves. The system and method can detect diffuse slowdown by spectral density analysis of brain waves recorded at a small number of discrete locations on the patient's head, thereby making bedside assessment easier, for example, using a hand-held device. That is, the system and method are capable of recording brain waves through two or more wires placed on the head of a patient, performing an algorithm to evaluate the ratio of recorded low frequency waves to high frequency waves, and comparing the ratio to a determined threshold to identify the onset of brain disease. In further embodiments, the systems and methods utilize machine learning and additional data, such as from medical records, to improve evaluation accuracy.

From the diagnostic criteria of DSM-IV-TR, an active version of the diagnostic and statistical manual for mental disorders, usually as a gold standard for diagnosis of delirium, established by the american psychiatric society, it is known that the diagnostic needs for delirium are met: a conscious disturbance with reduced attention, persistence or transfer capacity; b cognitive function changes (including memory recommendations, disorientation, language disorders), or the presence of sensory disorders that cannot be explained by dementia; the disease condition occurs in a short period (usually several hours to days), and the disease condition varies throughout the course of a day.

The above patent documents continuously monitor physiological information of a patient including at least brain signals with ten or more physiological sensors or brain sensors placed on the patient, and output an indication of the presence, absence, or subsequent likelihood of suffering from delirium of the patient. In practical operation, on the one hand, the electroencephalogram is mainly aimed at early monitoring of abnormal cerebral discharge which is atypical and is not easy to detect, and can not provide patient information required by DSM-IV-TR diagnostic standard, namely, the assessment of delirium is difficult to determine by only continuous monitoring of the electroencephalogram, and the time for taking measures for preventing delirium is easy to miss in time; on the other hand, one of the main purposes of examining delirium after operation is to take preventive measures in time and avoid high delirium treatment cost, and the current adoption of an evaluation scheme of continuous monitoring of high brain electrical accompanying increases the treatment cost of patients.

Except the electroencephalogram is adopted as an auxiliary examination means for delirium assessment and identification, the delirium scale derived on the basis of DSM-IV-TR diagnosis standard is commonly applied in clinic to assess the severity of ICU delirium, provide prognosis assessment for delirium patients and serve as a basis for treatment. The currently mainly used scales are as follows: consciousness blur rating scale (confusion assessment method, CAM), memory Delirium rating scale (memorial Delirium assessment scale, MDAS), delirium rating scale (Delirium rating scale, DRS), delirium rating scale-98 revision (DRS-R-98), delirium cognitive function measuring scale (cognitive test for Delirium, CTD), delirium consciousness blur rapid assessment method (3-Minute Diagnostic Interview for CAM-Defined Delirium, 3D-CAM). The above patent documents also mention solutions for continuing the evaluation of patients using one of the usual scales when delirium cannot be evaluated by electroencephalography. However, it is not considered that the delirium scale itself, although combining objective cognitive test assessment into delirium assessment, has a high requirement for delirium assessment ability of the caregivers, and it is difficult for the caregivers to achieve reliable and effective delirium assessment based on the solutions proposed in the above patent documents, i.e. based on the caregivers' own understanding assessment of electroencephalogram and scale.

Furthermore, there are differences in one aspect due to understanding to those skilled in the art; on the other hand, since the applicant has studied a lot of documents and patents while making the present invention, the text is not limited to details and contents of all but it is by no means the present invention does not have these prior art features, but the present invention has all the prior art features, and the applicant remains in the background art to which the right of the related prior art is added.

Disclosure of Invention

There are currently delirium care areas such as: the delirium scale requires a high delirium assessment capability for the caretaker, and it is difficult to realize reliable and effective delirium risk prediction based on the evaluation of the delirium scale by the caretaker. In this regard, in the prior art, a solution of performing cluster grabbing on similar illness state information in medical big data and performing risk prediction based on the grabbed data is proposed, but because the solutions perform cluster grabbing on a plurality of medical data related to the illness itself, the important influence of individual differences of the current object to be evaluated on delirium risk is not considered, i.e. a single medical data cannot reflect the state and the reaction condition of the current object to be evaluated; furthermore, the medical data objects captured by such solutions are all patients for which it has been determined that a disease already exists, resulting in a very low reliability of the risk prediction result determined based on the medical data of the diagnosed patients, and thus the solutions proposed by the prior art are not suitable for delirium risk prediction, in particular for delirium risk prediction of the subject to be evaluated for which delirium has not yet occurred.

In the solution provided by the application, on the one hand, the characteristic of objective cognitive test evaluation of the delirium table is utilized, the evaluation data which are closely related to the individual difference of the patient and are obtained in the evaluation process are used as hidden factors for clustering and grabbing the big data, the high matching degree of data grabbing is realized on the basis of fully meeting the individual difference of the patient, and on the other hand, the problem of low accuracy of risk prediction results caused by mutual superposition and offset among a plurality of medical data is considered.

In view of the shortcomings of the prior art, the present application provides a delirium risk monitoring device based on a delirium dynamic prediction model, the delirium risk monitoring device at least comprises: the delirium factor processing module is used for calling dominant factors and implicit factors related to the object to be evaluated in the medical information management system after the object to be evaluated completes at least one rapid evaluation of delirium consciousness blurring, and generating labels required by the delirium risk monitoring module according to the attribute of the dominant factors and/or the attribute of the implicit factors; the delirium risk monitoring module is used for acquiring a plurality of case information sets matched with the object to be evaluated in the cloud platform based on the generated tag in a mode of information interaction with the cloud platform, wherein the delirium risk monitoring module calculates by utilizing a delirium dynamic prediction model according to the acquired case information sets to acquire delirium risk prediction of the object to be evaluated.

For the risk prediction of delirium occurrence or delirium deterioration, the research aspect of scholars at home and abroad mainly uses factors (such as patients, own diseases, treatment and environment, and the like (which can be regarded as dominant factors) as indexes to construct a risk system for delirium risk prediction, however, the factors are huge in number, and the degree of correlation among the factors is unclear, that is, although many factors have potential influence on delirium induction, factors which are irrelevant, weakly relevant or mutually redundant to delirium occurrence or delirium induction are included, and the factors not only have obvious contribution to prediction results, but also increase data calculation amount and reduce prediction efficiency.

In contrast, the delirium risk monitoring method provided by the application utilizes the characteristic of objective cognitive test evaluation of the delirium scale itself, takes the evaluation data which is closely related to the individual difference of patients and is obtained through the evaluation process as the hidden factors for clustering and grabbing big data, on one hand, the number of the hidden factors is far lower than that of the dominant factors, the calculated data amount is less, and the improvement of the data processing efficiency is facilitated, on the other hand, the delirium risk monitoring method provided by the application is based on the fact that partial case information sets are screened out on the basis of the hidden factors, and then the dominant factors which have potential influences on delirium induction are utilized, and the case samples which are only accurately used for risk monitoring are screened are continuously, so that the delirium risk monitoring can fully meet the individual difference of patients and realize high matching degree and high processing efficiency of data grabbing.

Further preferably, the delirium risk monitoring module calculates by using a delirium dynamic prediction model according to the obtained plurality of history information sets and the overall change trend and/or the local change trend corresponding to each history information set, so as to obtain delirium risk prediction of the object to be evaluated. In particular, for individual patients, if the currently prevailing big data analysis method is adopted, the big data is usually collected by factors, delirium and the association relationship between factors and delirium at the current time point, however delirium is a time series disorder, because the change of the related factors at a certain time point (or can be simply understood as implementation of therapeutic measures, etc.), the delirium at a subsequent time point is influenced, the time series disorder has a certain hysteresis and a relatively high randomness, and because of the influence of other factors with poor controllability (such as sudden pain caused by the disease of the patient, etc.), there is a possibility of severe fluctuation in a short period.

In contrast, the delirium risk monitoring method provided by the application is characterized in that big data are respectively analyzed and processed from two different layers of the overall change trend of the hidden factor and the local change trend of the hidden factor, and the method is particularly suitable for being used as the hysteresis characteristic of delirium of time series diseases, and the influence of the change of the related factors at a certain time point on the induction of delirium at a subsequent time point can be determined through the change trend. Meanwhile, the big data processing process under the overall change trend is relatively rough, and for the samples with delirium trend deterioration and repeated fluctuation caused by the poor controllability factor, the overall change trend of the samples cannot reflect the real delirium change trend of the samples. Therefore, the method combines the overall change trend with the local change trend, further improves the accuracy and reliability of risk prediction, and is particularly suitable for delirium risk prediction of an object to be evaluated, in which delirium does not occur, or delirium potential risk is low.

According to a preferred embodiment, after the label is acquired, the delirium risk monitoring module compares the label of the hidden factor of the current object to be evaluated with the label of the hidden factor of the plurality of case information in the cloud platform based on a preset similarity interval, so as to determine the plurality of case information which accords with the similarity interval and is used for forming the case information group in the cloud platform.

According to a preferred embodiment, the delirium risk monitoring module corrects the similarity interval in such a way that the interval range is selectively enlarged when the number of case information of at least one of the plurality of case information sets does not reach the sample number threshold, thereby satisfying the number of samples required for delirium risk prediction of the object to be evaluated while maximizing the degree of matching between the case information set and the object to be evaluated.

According to a preferred embodiment, the obtaining of the plurality of case information sets refers to that after determining the plurality of case information meeting the similarity interval in the cloud platform, the delirium risk monitoring module screens out a plurality of case information meeting the same label as the label of the object to be evaluated in the plurality of case information according to the labels of the plurality of case information and the label of the object to be evaluated, and forms the case information set.

According to a preferred embodiment, the dominant factor comprises at least one delirium primary risk factor and at least one delirium secondary risk factor, and the implicit factor comprises at least delirium assessment data determined during the completion of the rapid assessment of delirium consciousness blur by the subject to be assessed, which comprises at least one or several of delirium potential risk level change trend, delirium feature k.

According to a preferred embodiment, the delirium risk monitoring device comprises a delirium assessment module configured to: acquiring feedback information about an object to be evaluated and/or about auxiliary personnel, acquiring behavior information of the object to be evaluated when the delirium is blur-fast evaluated, and/or respectively generating parameters required by a delirium evaluation model according to the attribute of the behavior information and the attribute of the feedback information, and/or calculating by using the delirium evaluation model according to the generated parameters to obtain an evaluation value about at least one delirium feature, wherein the evaluation value is obtained when the object to be evaluated is blur-fast evaluated.

According to a preferred embodiment, said delirium risk monitoring device further comprises: the video processing module is used for acquiring behavior information and/or feedback information about the object to be evaluated in a video acquisition mode of the behavior of the object to be evaluated, which is diagnosed to have delirium or has delirium potential risk, when the delirium consciousness blur rapid evaluation is performed, and/or is externally connected with an input device, the video processing module is operated by the object to be evaluated and used for acquiring feedback information input by the object to be evaluated for evaluation content, and detecting the autonomous operation condition of the object to be evaluated in the process of being evaluated so as to acquire the behavior information and/or feedback information about the object to be evaluated.

The application also provides a delirium risk monitoring system based on a delirium dynamic prediction model, which at least comprises: a memory; the first computer processor is used for calling out dominant factors and implicit factors related to the first user in the database after the first user finishes at least one evaluation, and generating labels required by the second computer processor according to the attribute of the dominant factors and/or the attribute of the implicit factors; and the second computer processor is used for acquiring a plurality of history information groups matched with the first user based on the generated label in a mode of information interaction with the database, wherein the second computer processor calculates by utilizing a dynamic prediction model according to the acquired history information groups and the integral change trend and/or the local change trend corresponding to each history information group to acquire the risk prediction of the first user.

According to a preferred embodiment, the second computer processor is further configured to, after acquiring the tag, compare the tag of the implicit factor of the current first user with the tags of the implicit factors of the plurality of historical information in the database based on a preset similarity interval, so as to determine the plurality of historical information which accords with the similarity interval and is used for forming the historical information group in the database.

According to a preferred embodiment, the second computer processor is further configured to correct the similarity interval in such a way that the interval range is selectively enlarged when the number of history information of at least one of the plurality of history information sets does not reach the number of samples threshold, thereby satisfying the number of samples required for risk prediction of the first user while maximizing the degree of matching between the history information set and the first user.

In the context of the present invention, a processing module may be used "configured to" describe performing one or more functions. Generally, an element configured to perform or be configured to perform a function is capable of performing, or is adapted to perform, or is operable to perform, or otherwise perform the function. It should be understood that "at least one of X, Y, Z" and "one or more of X, Y, Z" can be understood as X alone, Y alone, Z alone, or any combination of two or more of X, Y, Z (e.g., XYZ, XY, YZ, XZ, etc.). Similar logic may also be applied to any two or more objects that appear in the statement "at least one … …" and "one or more … …". As used in this specification, the singular forms "a", "an" and "the" include plural referents unless the content and context clearly dictates otherwise. That is, for example, reference to "a device" includes a combination of two or more such devices. Unless otherwise indicated, an "or" connection is intended to be used in its proper sense as a boolean logic operator, including both alternative feature choices (a or B) and conjunctive feature choices (a or B). The intelligent electronic equipment comprises, but is not limited to, various terminal equipment such as computers, mobile phones, tablet computers and the like.

The proposed apparatus comprises at least one processing module, a system memory device and at least one computer readable storage medium. The at least one computer-readable storage medium has computer-executable instructions embodied thereon for causing a processor to implement aspects of the present invention. With the illustration of fig. 2 as an example, a plurality of processors, interfaces are interconnected by a communication bus (solid lines) such as a motherboard (system memory device not shown). The interfaces include at least a communication interface and an I/O interface. The modules are operatively coupled to a computer network by means of a communication interface, such as a network adapter. The computer network may be the internet, the internet and/or an extranet, or an intranet and/or an extranet in communication with the internet. The modules communicate with the intelligent electronic devices via a computer network or via a straight line (e.g., wired, wireless) connection.

At least one processing module, such as a first processing module, is configured to execute the computer-executable instructions. Such as the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. Where each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). Each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions. Various aspects of the present invention are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions. The processing module is a functional unit for interpreting and executing instructions, also called a central processing unit or CPU, as an operation and control core of the computer system, and is a final execution unit for information processing and program running.

The computer readable storage medium described above can be a tangible device that can hold and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. The computer-executable instructions described herein may be downloaded from a computer-readable storage medium to the individual computing/processing modules or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, gateway computers and/or edge servers. The network interface card or network interface in each computing/processing module receives computer-readable program instructions from the network and forwards the computer-executable instructions for storage in a computer-readable storage medium in the respective computing/processing module.

Computer program instructions for carrying out operations of the present invention may be assembly instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, c++ or the like and conventional procedural programming languages, such as the C language or similar programming languages. The computer readable program instructions may be executed entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, aspects of the present invention are implemented by personalizing electronic circuitry, such as programmable logic circuitry, field Programmable Gate Arrays (FPGAs), or Programmable Logic Arrays (PLAs), with state information for computer readable program instructions, which can execute the computer readable program instructions.

The invention also relates to a delirium risk monitoring device, said device comprising at least: the identification module is used for calculating based on at least third-party judging information by utilizing a delirium evaluation model, and obtaining an evaluation value about at least one delirium characteristic obtained by rapidly evaluating delirium consciousness blurring of an object to be evaluated, so that delirium labels of the object to be evaluated or delirium labels of the object to be evaluated are determined based on the evaluation values of a plurality of delirium characteristics.

According to a preferred embodiment, after the identification module obtains an evaluation value regarding at least one delirium feature obtained by a rapid evaluation of delirium confusion of the object to be evaluated, the evaluation processing module combines several evaluation values regarding the at least one delirium feature and determines whether the combination satisfies a predetermined evaluation condition.

According to a preferred embodiment, when a combination of several evaluation values regarding at least one delirium feature obtained by a rapid evaluation of delirium consciousness blur of an object to be evaluated meets a predetermined evaluation condition, outputting a delirium evaluation result that delirium exists in the object to be evaluated; when the combination among a plurality of evaluation values related to at least one delirium feature obtained by carrying out delirium consciousness fuzzy rapid evaluation on the object to be evaluated does not meet a preset evaluation condition, historical information related to the object to be evaluated in the medical information management system is called, and secondary evaluation is carried out by combining the medical information and/or the evaluation information in the historical information, so that the delirium potential risk level of the object to be evaluated is determined.

According to a preferred embodiment, the device further comprises: a preprocessing module configured to: based on the patient history of the current subject to be evaluated, first pre-judgment data, second pre-judgment data and third pre-judgment data respectively associated with at least three physiological states of the subject to be evaluated are generated, and the pre-processing module determines at least one feedback acquisition mode after comprehensively conditioning the plurality of pre-judgment data.

According to a preferred embodiment, in the feedback acquisition mode determined after the pre-judgment, when the auxiliary personnel perform delirium consciousness fuzzy rapid evaluation on the object to be evaluated, at least one acquisition module acquires feedback data about one or more of voice, video images, hand behaviors and point options on an evaluation interface of the object to be evaluated, and one or more of a reaction time analysis unit, an eye movement analysis unit, a relevance analysis unit of feedback information, a speech speed analysis unit and a hand behavior analysis unit in the acquisition module respectively processes the feedback data according to the evaluation items and the pre-judgment data and obtains at least one primary analysis information corresponding to each other, and a data processing module in the acquisition module performs secondary processing on a plurality of primary analysis information to obtain third party judgment information on patient behaviors in the evaluation process.

According to a preferred embodiment, the single analysis unit in the acquisition module corresponds to at least one or several of the first and second pre-determined data, and excludes part of the disturbance information affected by the habit of the object to be evaluated in the feedback situation based on the one or several of the first and second pre-determined data. The single analysis unit refers to one of a reaction time length analysis unit, an eye movement analysis unit, a correlation analysis unit of feedback information, a speech speed analysis unit and a hand behavior analysis unit.

According to a preferred embodiment, the reaction time length analysis unit, the eye movement analysis unit, the correlation analysis unit of the feedback information, the speech rate analysis unit respectively correspond to the reaction time length, the blink situation, the correlation of the feedback information, the speech rate, wherein the third party decision information on the patient behaviour during the evaluation obtained by the secondary processing by the data processing module comprises at least one parameter value determined by the primary analysis information.

According to a preferred embodiment, the device further comprises: delirium factor processing module. The delirium factor processing module is used for modulating dominant factors and implicit factors related to the object to be evaluated in the medical information management system, and generating labels required by the delirium factor processing module according to the attribute of the dominant factors and/or the attribute of the implicit factors; based on the generated label, the delirium factor processing module obtains a plurality of case information groups matched with the object to be evaluated in the cloud platform in a mode of information interaction between the delirium factor processing module and the cloud platform; the delirium factor processing module calculates by using a delirium dynamic prediction model according to the obtained multiple case information sets to obtain delirium risk prediction of the object to be evaluated.

The invention also relates to a delirium risk monitoring system, which at least comprises an identification module, wherein the identification module utilizes a delirium evaluation model to calculate at least based on third party judgment information, and obtains an evaluation value about at least one delirium characteristic obtained by carrying out delirium consciousness blurring rapid evaluation on an object to be evaluated, so that delirium labels of the object to be evaluated which are happened or delirium labels which are not happened are determined based on the evaluation values of a plurality of delirium characteristics.

Drawings

FIG. 1 is a flow chart of a preferred delirium risk monitoring method provided by the present invention; and

fig. 2 is a simplified schematic diagram of the modular connection of a preferred delirium risk monitoring system provided by the present invention.

List of reference numerals

101: a mobile electronic device; 102: delirium factor processing module; 103: delirium assessment module; 1011: an audio-visual processing module; 1012: externally connecting input equipment; 104: an evaluation processing module; 105: delirium risk monitoring module; 106: a cloud platform; 107: a medical information management system.

Detailed Description

The following detailed description refers to the accompanying drawings.

The present invention will be described in detail with reference to the accompanying drawings.

Aiming at the defects of the prior art, such as high requirement of delirium evaluation capability of a delirium scale on a nursing staff, the problem that reliable and effective delirium risk prediction is difficult to realize only based on the understanding evaluation of the scale by the nursing staff. In the prior art, solutions of clustering and grabbing similar illness state information in medical big data and predicting risks based on the grabbed data are proposed. However, since such solutions are all cluster-grabbing for a plurality of medical data related to the disease itself, the important influence of individual differences of the current subject to be evaluated itself on the risk of delirium is not considered. That is, the single medical data cannot reflect the state and the reaction condition of the current object to be evaluated. Furthermore, the medical data objects captured by such solutions are all patients for whom it has been determined that a disease already exists, resulting in extremely low reliability of risk prediction results determined based on the medical data of the diagnosed patients. The solutions proposed by the prior art are therefore not suitable for delirium risk prediction, in particular for delirium risk prediction of an object to be evaluated for which delirium has not yet occurred.

In the solution provided by the invention for carrying out risk prediction by using similar illness state information in medical big data, on one hand, the evaluation data which is closely related to the individual difference of patients and is obtained through the evaluation process is used as a hidden factor for clustering and grabbing the big data, the high matching degree of data grabbing is realized on the basis of fully meeting the individual difference of the patients, and on the other hand, the problem of low accuracy of risk prediction results caused by mutual superposition and offset among a plurality of medical data is considered.

As shown in fig. 1, the present invention provides a delirium risk monitoring method based on a delirium dynamic prediction model. The delirium risk monitoring method is especially aimed at the object to be evaluated, on which delirium or a delirium potential risk level is low, has not yet appeared. As shown in fig. 1, the delirium risk monitoring method mainly includes:

Pretreatment step (not shown in the figure): after the patient information of the current object to be evaluated is acquired, the system invokes the patient history of the current object to be evaluated and transmits the patient history to the preprocessing module through information interaction with the medical information management system 107, and the preprocessing module prejudges the language expression capability, the body movement capability and the face expression capability of the object to be evaluated.

S1: acquiring feedback information about the object to be evaluated and/or about auxiliary personnel by at least one mobile electronic device 101 and acquiring behavior information of the object to be evaluated when performing delirium consciousness blur rapid evaluation;

s2: generating parameters required by the delirium evaluation model according to the attribute of the behavior information and the attribute of the feedback information by the delirium factor processing module 102;

s3: calculating by the delirium evaluation module 103 according to the generated parameters by using the delirium evaluation model to obtain an evaluation value about at least one delirium feature obtained by rapidly evaluating delirium consciousness blur of the object to be evaluated;

s4: after the delirium evaluation module 103 obtains the evaluation value about the at least one delirium feature obtained by rapidly evaluating the delirium consciousness blur of the object to be evaluated, the evaluation processing module 104 combines the several evaluation values about the at least one delirium feature, and determines whether the combination satisfies the predetermined evaluation condition.

S5: the delirium factor processing module 102 invokes dominant factors and implicit factors related to the subject to be evaluated in the medical information management system 107, and generates labels required by the delirium risk monitoring module 105 according to the attribute of the dominant factors and/or the attribute of the implicit factors;

s6: based on the generated labels, the delirium risk monitoring module 105 obtains a plurality of case information sets matched with the object to be evaluated in the cloud platform 106 in a mode of information interaction with the cloud platform 106;

s7: the delirium risk monitoring module 105 calculates a delirium risk prediction of the subject to be evaluated according to the obtained plurality of case information sets by using the delirium dynamic prediction model.

The following is a detailed description of the first steps S5 to S7:

for step S5: the delirium factor processing module 102 retrieves dominant factors and implicit factors related to the subject to be evaluated in the medical information management system 107, and generates labels required by the delirium risk monitoring module 105 according to the attribute of the dominant factors and/or the attribute of the implicit factors.

"dominant factor" refers to the primary risk factor for delirium and the secondary risk factor for delirium. The major risk factors for delirium include at least APACHE-II score, history of chronic disease, sleep disorders, sedative or anesthetic use, infection, indwelling catheter, hearing impairment. Delirium secondary risk factors include those for which ABCDEF bundling measures are aimed: pain, mechanical ventilation, sedative or analgesic use, limited mobility, and no family accompanies. For example, information Dm such as age D1, sex D2, body mass index D3, education level D4, history of alcoholism or smoking D5, basic disease D6, disease type D7 of the present hospital, and use of analgesic D8 of the subject to be evaluated.

Dominant factors refer to information that can be determined without evaluation, such as personal information or medical history information or medication information of the subject to be evaluated. "attribute of an dominant factor" refers to the distinction from the actual situation of the subject to be evaluated, which may or may not have the dominant factor, or which falls within a defined interval. Based on this attribute, the corresponding tag Dmn can be determined. Since the values corresponding to the attributes are all non-numeric, the delirium factor processing module 102 in step S5 generates the labels required by the delirium risk monitoring module 105 according to the attribute of the dominant factor. Taking "age D1" and "whether there is alcoholism or smoking history D5" as an example, four age-divided regions are defined for the age, {10 to 30, 31 to 50, 51 to 70, 71 to 100}, the four age-divided regions respectively correspond to one numerical designation 1 to 4, and if the age of the subject to be evaluated is 52 years old, i.e., falls within the defined region {51 to 70}, the label of the subject to be evaluated includes at least D13. The label of the subject to be evaluated comprises at least D51, with or without a digital indicator 1 and 0, respectively, for the presence or absence of history of alcoholism or smoking.

"implicit factor" refers to delirium assessment data of an object to be assessed. Implicit factors refer to information determined by evaluation. Delirium assessment data may include delirium potential risk level change trend δq. Delirium assessment data may also include information of combinations Σ, k1, k2, k3, k4, etc. Wherein, "delirium potential risk level change trend δq" may refer to information of change of delirium potential risk level with time according to each evaluation of the object to be evaluated. The values corresponding to the attributes are all of a numerical type, so that the attribute of the hidden factor is the label required by the delirium risk monitoring module 105. For example, four evaluations of the object under evaluation result in delirium risk potential levels Q of 0, 1, 2, 1, respectively, in turn, i.e. the label of the object under evaluation comprises at least Ω {0, 1, 2, 1}.

For step S6: based on the generated labels, the delirium risk monitoring module 105 obtains a plurality of case information sets matched with the object to be evaluated in the cloud platform 106 in a manner that the delirium risk monitoring module interacts with the cloud platform 106.

The "cloud platform 106" may be a third party service database that stores a large amount of case information. Each case information contains a label Hmn for the dominant factor and a label Φ for the recessive factor. Preferably, the delirium risk monitoring module 105 obtains a plurality of case information sets matched with the object to be evaluated in the cloud platform 106 on the basis of meeting a preset similarity screening condition in a manner of information interaction with the cloud platform 106. The "preset similarity screening condition" is used to screen part of case information matching with the preset similarity in a large amount of case information of the cloud platform 106. More specifically, step S6 further includes one or more of the following steps:

S61: based on the similarity interval of 90% -100%, comparing the similarity between the label omega of the hidden factor of the current object to be evaluated and the label phi of the hidden factor of the plurality of case information in the cloud platform 106, so as to determine X case information conforming to the similarity interval in the cloud platform 106;

the label omega of the hidden factor for similarity comparison can be one or more of change trend information related to time in sigma, k1, k2, k3, k4 and delirium potential risk level Q obtained by each evaluation of the object to be evaluated. Since the "evaluation value" of the delirium features k1, k2, k3, k4 only comprises both negative and positive results, the negative result may be denoted by "-" and the positive result may be denoted by "+" whereby the evaluation value of its implicit factor is the label required for generating the delirium risk monitoring module 105. For example, four evaluations of the object under evaluation result in delirium features k1 being sequentially respectively-, +, -, i.e. the label of the object under evaluation comprises at least Ω k1{ -, +, - }. In the early stage of evaluation, since the subject to be evaluated may perform delirium consciousness ambiguity quick evaluation only once and twice, the label Ω will be compared to obtain a huge amount of multiple case information. Therefore, "the similarity comparison of the tag Ω of the hidden factor of the current subject to be evaluated and the tag Φ of the hidden factor of the plurality of case information in the cloud platform 106" is performed on the basis of satisfaction of the precondition. The precondition is that the current object to be evaluated is evaluated for λ times at least λ∈ {1,2,3,4,5}, that is, the tag Ω includes at least λ values. For the initial stage of evaluation, when the number of times of evaluation performed by the current object to be evaluated is at least less than lambda times, the referenceable data of the object to be evaluated is insufficient. Preferably, based on the similarity interval 90% -100%, the similarity between the tag Dmn of the dominant factor of the current object to be evaluated and the tag Hmn of the dominant factor of the multiple case information in the cloud platform 106 is compared, so as to determine X case information conforming to the similarity interval in the cloud platform 106.

S62: screening out a plurality of case information meeting the requirement that Hmn is the same as Dmn in the X case information according to the respective label Hmn of the X case information and the label Dmn of the object to be evaluated, and forming a case information group;

s63: judging whether the number of the case information of each of the plurality of case information sets meets a sample number threshold or not based on a preset sample number threshold;

s64: when the number of the case information of each of the plurality of case information sets respectively meets the sample number threshold value, the delirium risk monitoring module 105 invokes the plurality of case information sets;

s65: when the number of case information in at least one of the plurality of case information sets does not reach the sample number threshold, correcting the similarity interval in a manner of expanding the interval range, substituting the corrected similarity interval into the step S61, repeating S61-S63 until the step S64 is satisfied, and calling out the plurality of case information sets by the delirium risk monitoring module 105.

Where Dmn refers to the label of a certain dominant factor of the object to be evaluated. For example, D51 refers to one of the subjects being evaluated having a history of alcohol abuse or smoking. For a determined m value, its corresponding n value is unique. And thus can be divided into a plurality of case information groups based on a plurality of Dmn values. The single case information group corresponds to one label Dmn, and the labels Dmn corresponding to the plurality of case information groups are different from each other. The "plurality of case information groups" are a plurality of combinations respectively classified by different labels Dmn. All case information in a single case information group has a label Hmn with the same label as Dmn.

The "similarity comparison" mainly refers to comparison of two aspects, namely, overall variation trend and local variation trend. The overall change trend is the change trend of the delirium potential risk level Q finally determined after the evaluation, and the local change trend refers to the change trend of the evaluation values of the delirium characteristics k1, k2, k3, k4 determined in the evaluation process. And under the condition that the overall change trend accords with the similarity interval, comparing the local change trend, and comparing whether the local change trend of the overall change trend accords with the similarity interval or not. "correcting the similarity interval in such a manner as to enlarge the interval range" refers to a selective correction manner. Specifically, when the number of case information in at least one case information group in the plurality of case information groups does not reach the sample number threshold, the similarity interval requirement of the overall change trend is kept unchanged, and the similarity interval requirement of the local change trend is reduced. The multiple case information with high similarity is obtained as the risk prediction sample of the delirium risk monitoring module 105 to the greatest extent.

In the prior art, a data set meeting a specified screening range in a database is screened out by adopting a mode of the specified screening range, then each sample in the data set is compared with a sample to be tested one by one, the similarity between each sample and the sample to be tested is obtained, the unnecessary data processing amount is increased, the screening is not continued after a sufficient number of samples are obtained through screening, the confidence interval of the screened data set is lower under the setting, and the required number of samples are selected after all samples in the database are screened in the other part of the prior art, the CPU load rate is increased sharply and the data processing efficiency is reduced under the setting. In contrast, the delirium risk monitoring system provided by the application adopts a hierarchical screening structure to acquire the required number of case information groups, screens the case information group with the highest similarity in the database in a mode of meeting the specified screening range of the maximum similarity, counts the number of samples in the group after primary screening, and continuously screens the database in a mode of reducing the specified screening range when the required number is not reached. Under the delirium risk monitoring method adopting the layered screening structure, the screening process and the similarity comparison process are completed synchronously, so that unnecessary data processing amount is reduced, and the sample acquisition task with higher confidence interval can be completed without completely screening all data.

The screening process firstly screens case information which accords with delirium evaluation data of a current object to be evaluated, then counts the number of primarily screened cases, and properly enlarges the screening range and performs secondary screening when the required minimum sample number threshold is not reached, so that on one hand, the obtained sample number is determined to be enough to support delirium risk prediction, on the other hand, the validity of the obtained data is ensured, and the accuracy of delirium risk prediction is improved.

For step S7: the delirium risk monitoring module 105 calculates a delirium risk prediction of the subject to be evaluated according to the obtained plurality of case information sets by using the delirium dynamic prediction model.

For easy understanding, the labels Φ of the plurality of case information will be described below: as the similarity screening condition is satisfied between the labels phi and omega of the screened case information, that is, the similarity screening condition is satisfied by the change trend of the delirium potential risk level of part of the labels phi and the change trend of the delirium potential risk level of the labels omega. For delirium potential risk level change trend in the label Φ located after the partial change trend, the trend value is regarded as the trend value. That is, the plurality of case information that is screened out corresponds to a trend value respectively, and the trend value is used for providing calculation data for the delirium risk prediction model of the object to be evaluated. "the delirium potential risk level change trend" later refers to a partial delirium potential risk level change trend within a preset time period. The preset duration range may be one month or two months.

For example, if the change trend of the delirium potential risk level within one month in the label Φ is 1, 0, 1, the average change trend is equal and no higher risk level exists, the trend value is 0, which indicates that the delirium risk is smaller and the state is more stable. In addition, if at least one of the average trend is increased or a higher risk level is present, the trend value is 1, which indicates that delirium risk is increased. If the average trend is reduced and there is no higher risk level, the trend value is-1, indicating that delirium risk is reduced and the state is stable.

Similarly, since the similarity screening conditions are satisfied between the labels Φ and the labels Ω of the plurality of case information, namely, the change trend of the evaluation values of the delirium features k1, k2, k3, k4 in the label Φ and the change trend of the partial evaluation values of the delirium features of the label Ω satisfy the similarity screening conditions. The trend of the estimated value of the delirium feature located after the partial trend in the label Φ is regarded as the trend value. The trend values of several different labels can be determined from different delirium assessment data.

For the "delirium dynamic prediction model" in step S7, a plurality of calculation formulas are pre-stored in the delirium dynamic prediction model. For example, it contains delirium risk prediction calculation formula. Thus, more specifically, step S7 comprises at least one or several of the following steps:

S71: acquiring a plurality of case information groups, and determining the duty ratio of each case information group based on the number of case information of each case information group;

s72: determining trend values of the case information based on the label phi between the case information and the label omega, wherein the label phi meets the similarity screening condition;

s73: respectively counting the trend values of each case information group to generate trend values of each case information group;

s74: based on the duty ratio and trend value of each case information set, delirium risk prediction for the current object to be evaluated in a preset time length range is determined and output.

The "trend value" includes at least a trend value corresponding to the overall trend and a trend value corresponding to the local trend. The calculation is preferably performed based on trend values corresponding to the overall trend, i.e. on the trend of the delirium potential risk level Q. Because of the comparison of the duty ratios in the calculation process, a situation that the difference between the duty ratios is small may occur, and in this case, the real variation trend of the sample cannot be reflected, so that the calculation result is greatly biased to the delirium risk to be predicted as worsening or to be good is two extreme. Therefore, for step S74, it is more preferable that: if the calculated trend values are 1, 0 and-1, the difference between the respective corresponding duty ratios is smaller, the delirium risk prediction for the current object to be evaluated in the preset time range cannot be determined, and then calculation is performed based on the trend values corresponding to the local variation trend, namely the variation trend of the evaluation values of delirium features k1, k2, k3 and k4, so that at least one delirium potential risk level Q can be determined, and the delirium risk prediction for the current object to be evaluated in the recent period can be determined and output based on the delirium potential risk level Q. Since the trend of the local trend/delirium feature evaluation value only comprises negative and positive results, and the delirium feature evaluation value combination Σ can obtain at least one delirium potential risk level Q corresponding to the positive result, and compare the positive result with the current delirium potential risk level of the object to be evaluated, the delirium risk prediction-delirium potential risk level rising, leveling or falling of the object to be evaluated within the preset duration range can be determined.

Further description is made for "the duty ratio of each case information group: first, the specific gravity between the dominant factors is set in a predetermined manner before risk prediction. As before, dominant factors differ from major and minor risk factors, and accordingly the influence of different dominant factors on delirium risk is of different specific gravity. The auxiliary personnel can modify the specific gravity among a plurality of dominant factors which are already preset according to the actual situation. Since a single case information set corresponds to the unique label Dmn, a single case information set also corresponds to only the unique specific gravity value. The respective corresponding proportion of the plurality of case information groups can be obtained through statistics of the number of the case information in the single case information group, the proportion is further optimized based on the proportion value corresponding to the single case information group, and the optimization result is the proportion of each case information group. For example, the product result is the duty ratio of each case information group by multiplying the two.

The following is a detailed description of S1 to S4 in steps:

for step S1: feedback information about the object to be evaluated and/or about the auxiliary personnel is acquired by the at least one mobile electronic device 101 and behavior information of the object to be evaluated is acquired when a delirium consciousness blur rapid evaluation is performed.

The "feedback information about the object to be evaluated and/or about the auxiliary staff" in the above step S1 refers to the selection made by the object to be evaluated or the auxiliary staff on the entries of the delirium consciousness blur rapid evaluation scale, and the feedback information may include four information including, no, incorrect, and correct.

For "feedback information about the auxiliary personnel" in step S1, which is an answer to the relevant item by the auxiliary personnel, the answer acquisition means may be determined by the answer input by the auxiliary personnel to the mobile electronic device 101. For example, "during the assessment, the patient has sleepiness, comatose or coma? The assistant inputs feedback information with or without to the mobile electronic device 101 based on the judgment of the situation of the object to be evaluated itself.

For "feedback information about the object to be evaluated" in step S1, which is an answer of the object to be evaluated to the relevant item, the answer acquisition means may be determined by an answer input to the mobile electronic device 101 by an auxiliary person. Preferably, the "feedback information about the object to be evaluated" may also be an answer input by the object to be evaluated by itself. For example, "how much you feel confused the last day" is an auxiliary person asking one of the subject under evaluation for a scale entry? "an operable remote control that the subject to be evaluated can hold, with or without input to the mobile electronic device 101 by itself. The above preferred embodiments are mainly presented for patients who cannot speak, such as mechanical ventilation, central venous cannulas, etc., in which case delirium assessment cannot be achieved by patient utterance description, the intelligent assessment system provided by the invention provides an operable remote control to the subject to be assessed, and answers the entry on the display screen by sliding or pressing a key on the operable remote control.

It is further preferred here for "feedback information about the object to be evaluated" in step S1, which is an answer to the relevant item by the object to be evaluated, the answer acquisition means may be determined by the mobile electronic device 101. For example, "please ask which year is this year? By way of example, the mobile electronic device 101 may analyze the answer of the object to be evaluated based on the video acquired by the video processing module 1011 to determine the answer under the entry. By using the auxiliary evaluation of the video acquisition mode, the invention can acquire an accurate answer by a mode of analyzing and processing the video, mutually verify the answer with the answer input by an auxiliary person to the mobile electronic equipment 101, and eliminate the problem that the answer input by the auxiliary person is wrong due to the error of the auxiliary person.

For step S2: parameters required by the delirium evaluation model are generated by the delirium factor processing module 102 according to the attribute of the behavior information and the attribute of the feedback information.

In step S2, the "attribute of the feedback information" may include four information including no, incorrect, and correct information, and the values corresponding to the attribute of the feedback information are all non-numeric, so that the delirium factor processing module 102 needs to generate parameters required by the delirium evaluation model according to the attribute of the feedback information. By "parameters needed for the assessment of a model for delirium" is meant at least one of the responses given by the patient, the responses given by the auxiliary staff, the patient behaviour during the assessment, the patient history or a combination of several of the four parameters. For example, "have or incorrect" may be set to 1 and "not or correct" may be set to 0 for a parameter corresponding to an attribute of feedback information about an object to be evaluated and/or about an auxiliary person. The examples illustrated herein are: when the patient gives feedback information "yes" for item 08, the parameter value corresponding to the parameter of the answer given by the patient then comprises at least a081.A indicates the parameter of the answer given by the patient, 08 indicates feedback information for entry 08,1 that is "there". For example, when the auxiliary staff gives feedback information of "no" to the item 13, the parameter value corresponding to the parameter of the answer given by the auxiliary staff includes at least B130.B indicates the parameter of the answer given by the auxiliary personnel, 13 indicates the feedback information for entry 13,0 indicating "none". The required parameters are determined based on a delirium assessment model. Which parameters are needed for the delirium assessment model for realizing the delirium assessment, generating corresponding parameters according to the attribute of the feedback information and the attribute of the behavior information, and determining corresponding parameter values.

The term "behavior information of the object to be evaluated during the delirium consciousness blur rapid evaluation" refers to the collection and analysis of the external performance of the object to be evaluated during the evaluation based on the third party angle of the mobile electronic device 101. For example, "please ask which year is this year? By way of example, the subject may be evaluated as having answered the correct year if the auxiliary personnel have repeated the entry at least twice, in which case the patient appears to be unable to keep up with the topic in question or to be inappropriately distracted by environmental stimuli. The mobile electronic device 101 analyzes the answer process of the object to be evaluated based on the video collected by the mobile electronic device 101 to determine the behavior information about the object to be evaluated under the entry. The auxiliary evaluation of the video acquisition mode is different from the sense angle of auxiliary personnel, the video acquisition mode is used for analyzing and processing the actual response of the patient to be evaluated from the objective fact angle, and the auxiliary personnel with strong subjectivity and larger understanding deviation are prevented from being singly relied on, so that the evaluation accuracy and reliability of the intelligent evaluation system provided by the invention are ensured.

Further preferably, the "behavior information" may include several information such as a reaction time period a, a blink situation b, a correlation of feedback information C, a speed of speech d, etc., and a third party decision C related to the patient answer. For the related items which are needed to be completed by auxiliary personnel in the scale and are carried out after the object to be evaluated completes the corresponding items, the auxiliary personnel mainly rely on memory and sense to answer, and the intelligent evaluation system provided by the invention provides a third party judgment C which can mutually verify the answer of the auxiliary personnel by means of the video processing technology of the mobile electronic equipment 101. For example, when the patient gives a correct answer to item 08 "where here is asked", and the mobile electronic device 101 analyzes that the object to be evaluated answers the correct year, the parameter value corresponding to the parameter of the patient's behavior in the evaluation process includes at least C081.

Further preferably, each of the entries 12 to 20 answered by the auxiliary staff is respectively associated with at least one behavior information by a preset for the relevant entry in the scale requiring completion by the auxiliary staff. For example, entry B11 "whether the patient is asleep, comatose or comatose during the evaluation", the entry 11 being associated with the blink situation B in the behavior information by being set in advance. For another example, entry B18 is associated with blink conditions a, B in the behavior information by being preset. More preferably, each of the items 1 to 10 answered by the object to be evaluated is respectively corresponding to the third party judgment C in the behavior information by presetting with respect to the relevant items in the scale which require completion by the auxiliary personnel. For example, by being set in advance, the item 1 is associated with the third party determination C1 for the item 1 in the behavior information.

Here, the "behavior information of the reaction time period a" is illustrated by: the auxiliary personnel inquire the objects to be evaluated one by one according to a preset item sequence, and after the inquiry is completed, the response time length corresponding to each item according to the preset item sequence is determined based on the mobile electronic device 101 in the step 1. The mobile electronic device 101 obtains the attribute of the behavior information of the reaction time length a, i.e., a occurrence of fluctuation or a non-occurrence of fluctuation, based on its analysis of the trend of variation of the reaction time length corresponding to each of the items 1 to 10 answered by the object to be evaluated.

Preferably, "predetermined order of entries" with respect to the object to be evaluated refers to the order of entry 4, entry 5, entry 6, entry 7, entry 8, entry 9, entry 10, entry 1, entry 2, entry 3. Preferably, the "predetermined order of items" that require an auxiliary person to answer refers to the order of items 16, 17, 18, 19, 20, 13, 14, 15, 11, 12, 21, 22. Wherein the "preset entry order" of the query of the object to be evaluated is performed in preference to the "preset entry order" of the response of the auxiliary personnel.

Since the values corresponding to the attributes of the behavior information are non-numeric, the delirium factor processing module 102 in step S2 generates parameters required by the delirium evaluation model according to the attributes of the feedback information. As above, the "parameters required for the delirium assessment model" in step S2 refers to at least one of the four parameters or a combination of several of the four parameters of the answer given by the patient, the answer given by the auxiliary personnel, the patient behaviour during the assessment, the patient history. For the parameter corresponding to the attribute of the behavior information, "occurrence of fluctuation" may be set to 1, and "no occurrence of fluctuation" may be set to 0. For example, when the mobile electronic device 101 obtains the attribute of the behavior information of the reaction duration a, i.e., a fluctuation occurs, based on its analysis of the trend of the variation of the reaction duration corresponding to each item of the preset item sequence, the parameter value corresponding to the parameter of the patient behavior in the evaluation process includes at least a1. Correspondingly, for a plurality of information such as blink condition b, relevance c of feedback information, speech speed d and the like, the parameter value corresponding to the parameter of patient behavior in the evaluation process at least comprises one or more of b1, b0, c1, c0, d1 and d 0.

For step S3: the delirium evaluation module 103 calculates according to the generated parameters by using the delirium evaluation model to obtain an evaluation value related to at least one delirium feature obtained by rapidly evaluating delirium consciousness blur of the object to be evaluated.

Step S3 is more specific: acquiring parameter values of the plurality of parameters; matching and updating the parameter values of the parameters by using a delirium evaluation model; calculating by using a delirium evaluation model according to a plurality of parameter values obtained after matching and updating; obtaining an evaluation value related to at least one delirium feature obtained by rapidly evaluating delirium consciousness blurring of an object to be evaluated.

"delirium features" include delirium features k, k e {1,2,3,4}, i.e. delirium features 1,2,3, 4. The "evaluation" of delirium characteristics includes both negative and positive results. Negative results may be indicated by "-" and positive results may be indicated by "+". Specifically, based on the rapid assessment of delirium confusion, delirium feature 1 refers to an acute onset or fluctuation change, delirium feature 2 refers to inattention, delirium feature 3 refers to confusion, and delirium feature 4 refers to a change in consciousness level. Rapid assessment of delirium confusion determines that delirium must satisfy delirium features 1 and delirium features 2, and at least either delirium feature 3 or one or both of delirium features 4.

Wherein the evaluation value of delirium feature 1 is determined based on entries 8-10 and entries 18-20, and delirium feature 1 corresponds to Aiji e {08, 09, 10} and Biji e {18, 19, 20} by predetermined settings.

Wherein the delirium feature 2 is determined based on items 4 to 7 and items 16 to 17, and the delirium feature 2 corresponds to Aiji e {04, 05, 06, 07} and Biji e {16, 17} by the predetermined setting.

Wherein the delirium feature 3 is determined based on items 1-3 and items 13-15, and the delirium feature 3 corresponds to Aiji e {01, 02, 03} and Biji e {13, 14, 15} by the predetermined setting.

Wherein the delirium feature 4 is determined based on the entries 11-12, and the delirium feature 4 corresponds to the Biji e {11, 12} by the predetermined setting.

For delirium feature 1, it is further preferable that after each of Aij and/or Bij corresponding to entries 1 to 20 is called out of database 106 to calculate, when current delirium feature 1 is negative, delirium feature 2 is positive, delirium feature 3 is positive and/or delirium feature 4 is positive, delirium feature 1 is again determined based on Biji e {21, 22}, and the evaluation value of delirium feature 1 is updated according to the determination result.

For the "delirium assessment model" in step S3, a plurality of calculation formulas are pre-stored in the delirium assessment model. For example, the method comprises delirium evaluation estimation formula based on delirium consciousness blur rapid evaluation method.

Thus, more specifically for the "delirium assessment model" in step S3, step S3 comprises at least one or several of the following steps:

s31: acquiring a plurality of parameters and corresponding parameter values thereof determined by the delirium factor processing module 102, wherein the parameters at least comprise Aij, bij, ζj and Cij;

aij refers to the parameter of the answer given by the patient, i refers to the feedback information for entry i, j, which is "present, incorrect, none or correct";

bij refers to the parameter of the answer given by the auxiliary person, i refers to feedback information for item i, j indicating "there is, is incorrect, is not or is correct";

ζj is a parameter of patient behavior during the evaluation, ζ is at least one behavior information, j is feedback information "fluctuation occurs or no fluctuation occurs";

cij refers to the parameter of patient behavior during the assessment, C refers to a third party decision contained in at least one behavior information, i refers to feedback information for entry i, j indicating "present, incorrect, not or correct";

i∈{01，02....09，10....21，22}，j∈{1，0}，ζ∈{a，b，c，d....}；

S32: based on the association relation between at least one Aij and at least one Cij, matching the Aij with the Cij, outputting the Aij when the matching between the Aij and the corresponding Cij is successful, updating the j value in the Aij by the j value in the Cij when the matching between the Aij and the corresponding Cij is failed, and outputting the updated Aij;

s33: based on the association relation between at least one Bij and at least one ζj, matching the Bij with the at least one ζj, outputting the Bij when the matching between the Bij and any one of the at least one ζj corresponding to the Bij is successful, updating the j value in the Bij with the j value in the ζj when the matching between the Bij and all ζj corresponding to the Bij fails, and outputting the updated Bij;

the delirium evaluation module 103 stores the association between at least one Aij and at least one Cij and the association between at least one Bij and at least one ζj in advance;

the association relation between at least one Aij and at least one Cij is determined in a mode that the numerical value between the first digit and the last digit of a non-number can be regarded as i value corresponding, and the association relation between at least one Bij and at least one ζj is determined in a preset mode;

The matching mode between the Bij and at least one ζj means that the j value in the Bij is compared with the j value in the at least one ζj, if the two values are the same, the matching is successful, and if the two values are the same, the matching is failed;

the matching mode between the Aij and the Cij means that the j value in the Aij is compared with the j value in at least one Cij, if the two values are the same, the matching is successful, and if the two values are the opposite, the matching is failed;

s34: based on the preset delirium characteristics k, k epsilon {1,2,3,4}, one by one, the items ii epsilon {01, 02..19, 20} corresponding to at least one delirium characteristic k are fetched, and the evaluation value of the delirium characteristic k is determined to be negative or positive according to the preset evaluation value judgment condition of the delirium characteristic k:

the "evaluation value determination condition of the delirium feature k" refers to that all Aij and/or Bij corresponding to the delirium feature k are/is retrieved, if the j value, which is the parameter value of any Aij or any Bij, is 1, the evaluation value of the delirium feature k is positive, and if the j value, which is the parameter value of all Aij and/or Bij, is 0, the evaluation value of the delirium feature k is negative;

as described above, regarding the parameter corresponding to the attribute of the object to be evaluated and/or the feedback information regarding the auxiliary personnel, the parameter value set to "have or incorrect" is set to 1, and the parameter value set to "have or correct" is set to 0;

S35: based on the evaluation value of the currently determined delirium feature k, k epsilon { l,2,3,4}, judging whether the delirium feature k, k epsilon {1,2,3,4} accords with the preset selective evaluation condition, and outputting the evaluation value of the determined delirium feature k, k epsilon {1,2,3,4}, when the delirium feature k, k epsilon {1,2,3,4 }; otherwise, prompting auxiliary personnel to carry out selective evaluation, updating the evaluation value of the currently determined delirium characteristics k, k epsilon {1,2,3,4} based on the evaluation result of the selective evaluation, and outputting the updated evaluation value.

The "selectivity evaluation condition" refers to that after the Aij and/or Bij corresponding to each of the entries 1 to 20 are called out of the database 106 to calculate, if the delirium feature 1 is negative, the delirium feature 2 is positive, the delirium feature 3 is positive and/or the delirium feature 4 is positive, the auxiliary personnel is prompted to need to perform the selectivity evaluation, including the entries 21 to 22.

The process of "selective evaluation" refers to that delirium feature 1 is again determined based on Biji e {21, 22} corresponding to delirium feature 1, and the evaluation value of delirium feature 1 is updated according to the determination result.

Through the above preset operation, the analysis data of the mobile electronic device 101, i.e. the third party, can be corresponded to the answer input by the object to be evaluated or the auxiliary personnel, so that the answer input manually can be verified again by taking the analysis data of the third party as an auxiliary evidence, especially the answer input by the auxiliary personnel, which is difficult to avoid strong subjectivity and sensory deviation, can be evaluated for delirium characteristics of the object to be evaluated, and the evaluation result is generated according to the reaction and state actually displayed by the object to be evaluated in the evaluation process, so that the answer input by the auxiliary personnel, which has strong subjectivity and sensory deviation, can be effectively corrected and prompted, thereby being beneficial to improving the accuracy and reliability of delirium characteristics evaluation.

To clarify the manner of setting the "mobile electronic device 101" in the present intelligent evaluation system, the devices used in the intelligent evaluation system will be described herein: the intelligent evaluation system at least comprises a handheld intelligent mobile terminal, a display and an input device, wherein the handheld intelligent mobile terminal is operated by an auxiliary person, the display can be erected on a patient bed for the object to be evaluated to watch, and the input device is operated by the object to be evaluated. The handheld intelligent mobile terminal can be a smart phone, a smart watch, a smart bracelet, a tablet personal computer, a notebook computer and other devices, the display is an external device connected with the handheld intelligent mobile terminal, and an auxiliary person operates on the handheld intelligent mobile terminal and can control a display interface on the display. The input device may be an external input device like a projector controller or a mouse, which interfaces with the display, and the person to be evaluated may enter information into the display by holding the input device in his hand. Only two physical control keys are arranged on the input device, one physical control key is a mouse wheel, the mouse wheel is turned to browse upwards or downwards mainly aiming at a plurality of options which are vertically arranged, for example, when numbers are required to be input to an object to be evaluated in part of items, 1-9 numbers are vertically arranged on a display, and different numbers can be selected by turning the mouse wheel to the object to be evaluated; the other physical control key is a trigger key, and the object to be evaluated can input the options selected by the current mouse wheel into the display only by pressing the trigger key. The input device has simple and easily understood structure and operation, is beneficial to the use of the object to be evaluated, and especially can indirectly output the answer to the questions of auxiliary staff through the manual input device for patients such as mechanical ventilation, central venous cannula and the like, which cannot speak.

In connection with the above, the "mobile electronic device 101" is mainly distinguished from two types of contact and non-contact acquisition for acquiring the external appearance of the object to be evaluated. The non-contact acquisition mode includes an audio/video processing module 1011, and the audio/video processing module 101l is a camera disposed on the display. Including external input device 1012 by way of contact acquisition. External input device 1012 refers to an input device in the intelligent assessment system.

Another preferred embodiment is set forth below for steps S1 to S3, and this embodiment may be a further improvement and/or addition to the above embodiment, and the repeated description is omitted. In the case of no conflict or contradiction, the whole and/or partial content of other embodiments may be added to the present embodiment:

s1: and (3) a pretreatment step. The preprocessing module generates first pre-judgment data, second pre-judgment data and third pre-judgment data respectively associated with at least three physiological states of the object to be evaluated based on the patient history of the current object to be evaluated, and the preprocessing module determines at least one feedback acquisition mode after comprehensively conditioning the plurality of pre-judgment data.

Wherein the first pre-judgement data is related to the language expression capability of the object to be evaluated. The first prognosis data may be obtained based on patient history information, for example, for cases where a respiratory mask is worn, or where it is diagnosed that a post-operative language nerve is pressed, or for patients who can speak spontaneously, such as speaking fluency, clarity of speaking, logic of language, etc. The second pre-determined data is related to the physical activity capabilities of the subject to be evaluated. The second predictive data may be obtained based on a behavior sensor provided on the hand of the patient, the behavior sensor being capable of monitoring hand activity of the subject to be evaluated and generating an autonomously controllable degree of the hand of the subject to be evaluated, controlling blurriness, and the like. The third pre-judgment data is related to the face expression ability of the subject to be evaluated. The third prognosis data may be derived based on patient history information, for example for the case of an administered ventilator or oral intubation. The comprehensive condition processing refers to comprehensively analyzing the state of the object to be evaluated based on the three pre-judgment data, and screening out a feedback acquisition mode capable of effectively acquiring feedback of the object to be evaluated. The feedback acquisition mode may be, for example, a control by a remote controller alone, a video acquisition analysis, or the like.

S2: in the feedback acquisition mode determined after the prejudgment, when auxiliary personnel perform delirium consciousness fuzzy rapid assessment on an object to be assessed, at least one acquisition module acquires feedback data of one or more of voice, video images, hand behaviors and point options on an assessment interface of the object to be assessed, and one or more of a reaction duration analysis unit, an eye movement analysis unit and a relevance analysis unit, a speech speed analysis unit and a hand behavior analysis unit in the acquisition module respectively process the feedback data according to the assessment items and the prejudgment data and obtain at least one corresponding primary analysis information, and a data processing module in the acquisition module performs secondary processing on the primary analysis information to obtain third party judgment information on patient behaviors in the assessment process.

For the relevant evaluation items in the scale which need to be completed by the auxiliary personnel, each item in the items 12-20 which are answered by the auxiliary personnel is respectively corresponding to at least one analysis unit through the preset. The single acquisition module at least comprises one or more of a reaction time length analysis unit, an eye movement analysis unit, a relevance analysis unit of feedback information, a speech speed analysis unit and a hand behavior analysis unit. For example, the item B11 "whether the patient has sleepiness, comatose or coma during the evaluation" is associated with the above-described reaction time period analysis unit by a preset. For a plurality of analysis units, the single analysis unit at least corresponds to one or more of the first pre-judgment data and the second pre-judgment data, and based on the one or more of the first pre-judgment data and the second pre-judgment data, part of disturbance information influenced by habits of the object to be evaluated in the feedback situation can be eliminated.

Preferably, the reaction time length analysis unit, the eye movement analysis unit, the correlation analysis unit of the feedback information, the speech speed analysis unit and the like respectively correspond to a plurality of pieces of information such as the reaction time length a, the blink condition b, the correlation c of the feedback information, the speech speed d and the like.

Preferably, the reaction time length analysis unit is used for processing the feedback data according to the evaluation item and the prejudgment data and obtaining the reaction time length a. The auxiliary personnel inquire the objects to be evaluated one by one according to a preset item sequence, and the reaction time length analysis unit can determine the reaction time length corresponding to each item according to the preset item sequence based on a preset condition of the time length from the end of inquiry of the auxiliary personnel to the start of feedback response of the objects to be evaluated. The reaction time length analysis unit obtains the attribute of the behavior information of the reaction time length a, that is, a fluctuation or a no fluctuation, based on analysis of the change trend of the reaction time length corresponding to each of the items 1 to 10 answered by the object to be evaluated.

Preferably, the correlation c of the feedback information is used for processing the feedback data according to the evaluation item and the pre-judgment data and obtaining the correlation c of the feedback information. The correlation c of the feedback information means that the obtained attribute of the feedback information deviates from the attribute of the answer of the item. The attribute deviation referred to herein does not mean that the feedback information does not include a correct answer, but means that there is no correlation between the two. For example for item 7 "do you last month from 12 months? The attribute of the item answer is a logical continuous number, and the patient may answer the questions of the patient's own birthday, month and year or confusing repeated auxiliary personnel, and the attribute of the feedback information is personal information or understanding obstacle, and the feedback information is not a logical continuous number. In the above case, there is an attribute deviation, and the correlation c of the decision feedback information fluctuates, that is, it indicates that the subject to be evaluated has unclear thinking, no answer questions, and cannot follow the topic being discussed.

Preferably, the eye movement analysis unit is configured to process the feedback data based on the evaluation item and the pre-determined data and obtain the blink situation b. By means of monitoring and calculating the blink frequency of the patient in the evaluation process, whether the blink condition b fluctuates or not can be obtained, and if the blink condition b is judged to fluctuate, the condition indicates that the to-be-evaluated object has the change of response speed, the tendency to fall asleep and the low alertness in the evaluation process.

Preferably, the speech rate analysis unit is configured to process the feedback data according to the evaluation item and the pre-judgment data and obtain the speech rate d. The speed d here refers to the frequency of hand operations, for example for item 7 "do you last month from 12 months? And the numbers of 1-20 and a plurality of answer content items irrelevant to the stem are vertically distributed on the display, different numbers or items can be selected by an object to be evaluated by stirring the mouse wheel, and the options selected by the current mouse wheel can be input into the display by pressing the trigger key. In the process, the frequency of the patient for poking the mouse wheel and pressing the trigger button is recorded, and if the judgment speed d fluctuates, the speed change, the response speed change and the no-follow performance of the topic in question of the object to be evaluated are indicated in the evaluation process.

Preferably, the third party decision information on patient behavior during the evaluation of the secondary processing by the data processing module includes at least one parameter value determined by the primary analysis information. By presetting, the data processing module sets the occurrence fluctuation of the primary analysis information as a digital type 1 and sets the occurrence absence fluctuation of the primary analysis information as 0. For example, when the reaction duration analysis unit obtains the attribute of the behavior information of the reaction duration a, i.e., a fluctuation occurs, based on the analysis of the variation trend of the reaction duration corresponding to each item of the preset item sequence, the parameter value corresponding to the third party judgment information at least includes a1. Accordingly, for other primary analysis information such as blink condition b, relevance c of feedback information, speech speed d, etc., the parameter values corresponding to the third party judgment information at least include one or more of b1, b0, c1, c0, d1, d 0.

S3: the identification module utilizes the delirium evaluation model to calculate at least based on third party judgment information to obtain an evaluation value about at least one delirium characteristic obtained by rapidly evaluating delirium consciousness blur of the object to be evaluated, so that the delirium label of the object to be evaluated can be determined based on the evaluation values of the delirium characteristics.

s31: acquiring a plurality of parameters and corresponding parameter values thereof determined by the generation module 102, wherein the parameters at least comprise Aij, bij, ζj and Cij;

i∈{01，02....09，10....21，22}，j∈{1，0}，ζ∈{a，b，c，d....}；

s32: based on the association relation between at least one Aij and at least one Cij, matching the Aij with the Cij, outputting the Aij when the matching between the Aij and the corresponding Cij is successful, updating the j value in the Aij by the j value in the Cij when the matching between the Aij and the corresponding Cij is failed, and outputting the updated Aij:

the identification module 103 stores in advance an association relationship between at least one Aij and at least one Cij, and an association relationship between at least one Bij and at least one ζj;

S34: based on the preset delirium characteristics k, k epsilon {1,2,3,4}, one by one, taking the items ii epsilon {01, 02..19, 20} corresponding to at least one delirium characteristic k, and determining that the evaluation value of the delirium characteristic k is negative or positive according to the preset evaluation value judgment condition of the delirium characteristic k;

s35: based on the evaluation value of the currently determined delirium feature k, k epsilon {1,2,3,4}, judging whether the delirium feature k, k epsilon {1,2,3,4} accords with the preset selective evaluation condition, and outputting the evaluation value of the determined delirium feature k, k epsilon {1,2,3,4}, when the delirium feature k, k epsilon {1,2,3,4 }; otherwise, prompting auxiliary personnel to carry out selective evaluation, updating the evaluation value of the currently determined delirium characteristics k, k epsilon {1,2,3,4} based on the evaluation result of the selective evaluation, and outputting the updated evaluation value.

The "selectivity evaluation condition" refers to that after the Aij and/or Bij corresponding to each of the items 1 to 20 are respectively called out of the cloud platform 106 for calculation, if the delirium feature 1 is negative, the delirium feature 2 is positive, the delirium feature 3 is positive and/or the delirium feature 4 is positive, the auxiliary personnel is prompted to need to perform selectivity evaluation, including the items 21 to 22.

By combining the above-mentioned several evaluation values with respect to the at least one delirium feature is meant that the several evaluation values with respect to the at least one delirium feature are generalized to one combination Σ, e.g. k1+, k2+, k3+, k4+.

"predetermined evaluation condition" means three combinations ψ1{ k1+, k2+, k3+, k4+ }, ψ2{ k1+, k2+, k3+, k4- } determined in advance that delirium exists in the evaluation object, ψ3{ k1+, k2+, k3-, k4+ }.

The "determination whether or not the combination satisfies the predetermined evaluation condition" means that the combination Σ is compared with the combinations { ψ1, ψ2, ψ3} respectively.

S41: when the combination of a plurality of evaluation values related to at least one delirium characteristic obtained by rapidly evaluating delirium consciousness blurring of the object to be evaluated meets a predetermined evaluation condition, outputting a delirium evaluation result of the object to be evaluated for delirium.

Specifically, in the case that the combination Σ is matched with one of the combinations { ψ1, ψ2, ψ3}, that is, in the case that the combination between several evaluation values regarding at least one delirium feature obtained by the rapid evaluation of delirium consciousness blur by the object to be evaluated satisfies a predetermined evaluation condition, a delirium evaluation result that delirium exists in the object to be evaluated is output.

S42: when the combination of the several evaluation values about the at least one delirium feature obtained by the rapid evaluation of delirium consciousness blur of the object to be evaluated does not meet the predetermined evaluation condition, the history information related to the object to be evaluated in the medical information management system 107 is called, and the secondary evaluation is performed in combination with the medical information and/or the evaluation information in the history information, so as to determine the delirium potential risk level Q of the object to be evaluated.

Wherein, in case that the combination Σ does not match any of the combinations { ψ1, ψ2, ψ3}, i.e. the combination between the several evaluation values regarding the at least one delirium feature obtained by the above-mentioned "delirium consciousness blur rapid evaluation by the object to be evaluated" does not satisfy the predetermined evaluation condition. Wherein "evaluation information" refers to behavior information and feedback information.

Throughout this document, the word "preferably" is used in a generic sense to mean only one alternative, and not to be construed as necessarily required, so that the applicant reserves the right to forego or delete the relevant preferred feature at any time.

It should be noted that the above-described embodiments are exemplary, and that a person skilled in the art, in light of the present disclosure, may devise various solutions that fall within the scope of the present disclosure and fall within the scope of the present disclosure. It should be understood by those skilled in the art that the present description and drawings are illustrative and not limiting to the claims. The scope of the invention is defined by the claims and their equivalents. The description of the invention encompasses multiple inventive concepts, such as "preferably," "according to a preferred embodiment," or "optionally," all means that the corresponding paragraph discloses a separate concept, and that the applicant reserves the right to filed a divisional application according to each inventive concept.

Claims

1. Delirium risk monitoring device, characterized in that it comprises at least:

the identification module is used for calculating based on at least third-party judging information by utilizing a delirium evaluation model, and obtaining an evaluation value about at least one delirium characteristic obtained by rapidly evaluating delirium consciousness blurring of an object to be evaluated, so that delirium labels of the object to be evaluated or delirium labels of the object to be evaluated are determined based on the evaluation values of a plurality of delirium characteristics.

2. Delirium risk monitoring device according to claim 1, characterized in that after the identification module obtains an evaluation value regarding at least one delirium feature obtained by a rapid evaluation of delirium confusion of the object to be evaluated, several evaluation values regarding at least one delirium feature are combined by an evaluation processing module (104) and a determination is made as to whether the combination fulfils a predetermined evaluation condition.

3. Delirium risk monitoring device according to claim 1 or 2, characterized in that,

outputting delirium evaluation results of delirium existence of the object to be evaluated when a combination among a plurality of evaluation values of at least one delirium feature obtained by rapidly evaluating delirium consciousness blur of the object to be evaluated meets a predetermined evaluation condition;

When the combination among a plurality of evaluation values related to at least one delirium characteristic obtained by carrying out delirium consciousness fuzzy rapid evaluation on the object to be evaluated does not meet a preset evaluation condition, historical information related to the object to be evaluated in a medical information management system (107) is called, and secondary evaluation is carried out by combining the medical information and/or the evaluation information in the historical information, so that the delirium potential risk level of the object to be evaluated is determined.

4. A delirium risk monitoring device according to any one of claims 1-3, characterized in that the device further comprises: a preprocessing module configured to:

based on the patient history of the current subject to be evaluated, first pre-judgment data, second pre-judgment data and third pre-judgment data respectively associated with at least three physiological states of the subject to be evaluated are generated, and the pre-processing module determines at least one feedback acquisition mode after comprehensively conditioning the plurality of pre-judgment data.

5. Delirium risk monitoring device according to any one of claims 1-4, characterized in that,

in the feedback acquisition mode determined after the prejudgment, when auxiliary personnel perform delirium consciousness fuzzy rapid assessment on an object to be assessed, at least one acquisition module acquires feedback data of one or more of voice, video images, hand behaviors and point options on an assessment interface of the object to be assessed, and one or more of a reaction duration analysis unit, an eye movement analysis unit and a relevance analysis unit, a speech speed analysis unit and a hand behavior analysis unit in the acquisition module respectively process the feedback data according to the assessment items and the prejudgment data and obtain at least one primary analysis information corresponding to each, and a data processing module in the acquisition module performs secondary processing on the primary analysis information to obtain third party judgment information on patient behaviors in the assessment process.

6. Delirium risk monitoring device according to any one of claims 1-5, characterized in that a single analysis unit in the acquisition module corresponds to at least one or several of the first and second pre-judgement data, based on which one or several of the first and second pre-judgement data part of the disturbance information affected by the habit of the subject to be evaluated in the feedback situation is excluded.

7. Delirium risk monitoring device according to any one of claims 1-6, characterized in that,

the reaction time length analysis unit, the eye movement analysis unit, the relevance analysis unit of the feedback information and the speech speed analysis unit respectively correspond to the reaction time length, the blink condition, the relevance of the feedback information and the speech speed, wherein the third party judgment information of the patient behavior in the evaluation process obtained by the secondary processing of the data processing module comprises at least one parameter value determined by the primary analysis information.

8. Delirium risk monitoring device according to any one of claims 1-7, characterized in that the device further comprises: delirium factor processing module (102),

The delirium factor processing module (102) is used for modulating dominant factors and recessive factors related to the object to be evaluated in the medical information management system (107) and generating labels required by the delirium factor processing module (102) according to the attribute of the dominant factors and/or the attribute of the recessive factors;

based on the generated labels, the delirium factor processing module (102) obtains a plurality of case information groups matched with the object to be evaluated in the cloud platform (106) in a mode of information interaction between the delirium factor processing module and the cloud platform (106); the delirium factor processing module (102) calculates by using a delirium dynamic prediction model according to the obtained multiple case information sets to obtain delirium risk prediction of the object to be evaluated.

9. The delirium risk monitoring system is characterized by an identification module, wherein the identification module utilizes a delirium evaluation model to calculate based on at least third party judgment information, obtains an evaluation value about at least one delirium feature obtained by rapidly evaluating delirium consciousness blur of an object to be evaluated, and determines whether delirium labels or delirium labels of the object to be evaluated occur or not based on evaluation values of a plurality of delirium features.

10. Delirium risk monitoring system according to claim 9, characterized in that after the identification module obtains an evaluation value for at least one delirium feature obtained by a rapid evaluation of delirium confusion of the object to be evaluated, several evaluation values for at least one delirium feature are combined by an evaluation processing module (104) and a determination is made as to whether the combination fulfils a predetermined evaluation condition.