CN109124702B - Intelligent endoscope system with pneumoperitoneum control and central control module - Google Patents

Intelligent endoscope system with pneumoperitoneum control and central control module Download PDF

Info

Publication number
CN109124702B
CN109124702B CN201810998268.7A CN201810998268A CN109124702B CN 109124702 B CN109124702 B CN 109124702B CN 201810998268 A CN201810998268 A CN 201810998268A CN 109124702 B CN109124702 B CN 109124702B
Authority
CN
China
Prior art keywords
air pressure
air
video
central control
value
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810998268.7A
Other languages
Chinese (zh)
Other versions
CN109124702A (en
Inventor
丁帅
杨善林
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University of Technology
Original Assignee
Hefei University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hefei University of Technology filed Critical Hefei University of Technology
Priority to CN201810998268.7A priority Critical patent/CN109124702B/en
Publication of CN109124702A publication Critical patent/CN109124702A/en
Application granted granted Critical
Publication of CN109124702B publication Critical patent/CN109124702B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B17/00—Surgical instruments, devices or methods, e.g. tourniquets
    • A61B17/00234—Surgical instruments, devices or methods, e.g. tourniquets for minimally invasive surgery
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02M—APPARATUS FOR CONVERSION BETWEEN AC AND AC, BETWEEN AC AND DC, OR BETWEEN DC AND DC, AND FOR USE WITH MAINS OR SIMILAR POWER SUPPLY SYSTEMS; CONVERSION OF DC OR AC INPUT POWER INTO SURGE OUTPUT POWER; CONTROL OR REGULATION THEREOF
    • H02M3/00—Conversion of dc power input into dc power output
    • H02M3/22—Conversion of dc power input into dc power output with intermediate conversion into ac
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/50—Constructional details
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/50—Constructional details
    • H04N23/555—Constructional details for picking-up images in sites, inaccessible due to their dimensions or hazardous conditions, e.g. endoscopes or borescopes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/56—Cameras or camera modules comprising electronic image sensors; Control thereof provided with illuminating means
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10068—Endoscopic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Surgery (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Heart & Thoracic Surgery (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Probability & Statistics with Applications (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Power Engineering (AREA)
  • Endoscopes (AREA)

Abstract

The invention provides an intelligent endoscope system configured with pneumoperitoneum control and central control modules, which comprises an integrated endoscope system and a processing system, wherein the integrated endoscope system comprises a plurality of functional modules, a central control unit for controlling the plurality of functional modules to work and a power supply device for supplying power to the plurality of functional modules; the multiple functional modules comprise a pneumoperitoneum instrument for inflating the cavity, a man-machine interaction screen, a cold light source and a camera, the pneumoperitoneum instrument comprises a proportional control valve, a switching electromagnetic valve, an air escape valve, an air flow sensor and an air pressure sensor, and the power supply device comprises a switching power supply and multiple electromagnetic interference suppression circuits; the processing system is used for dynamic target positioning. The invention can solve the problem of video jitter caused by the over-fast instantaneous change of air pressure caused by the opening and closing of the switch valve, avoid the mutual transmission of electromagnetic interference signals among the functional modules and effectively improve the problem of noise sensitivity.

Description

Intelligent endoscope system with pneumoperitoneum control and central control module
Technical Field
The invention relates to the technical field of endoscope systems, in particular to an intelligent endoscope system with pneumoperitoneum control and central control modules.
Background
At present, surgical endoscopic surgery has become more and more attentive, and minimally invasive surgery has become a consensus of surgeons and patients. The endoscope system can provide a high-definition amplified operation picture, can clearly display a fine structure of an internal tissue, and is clearer in visual field compared with the traditional open operation, so that the operation is more accurate and more precise, the visceral organs except the operation part are effectively prevented from being unnecessarily interfered, bleeding in the operation is less, and the operation is safer.
However, the current cavity mirror systems have the following disadvantages:
(1) in the endoscopic surgery process, due to the noise characteristic of the minimally invasive surgery visual field caused by the shaking of the movable treatment endoscope, certain influence is generated on the positioning of the target and the excavation of the target characteristic;
(2) when the abdominal pressure is balanced, the shaking can be generated at the moment when the switch valve is opened and closed, so that the visual effect in the operation can be seriously influenced;
(3) the endoscope all-in-one machine adopts the shielding box to shield the radiation of each internal functional module to the space, thereby improving the anti-interference ability to the space radiation. However, electromagnetic interference signals generated by the functional modules during operation are transmitted to other functional modules through the working power lines, so that the functional modules interfere with each other, and the image quality of the endoscope is affected.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides an intelligent endoscope system with a pneumoperitoneum control and central control module.
(II) technical scheme
In order to achieve the purpose, the invention is realized by the following technical scheme:
the invention provides an intelligent endoscope system configured with a pneumoperitoneum control and central control module, which comprises an integrated endoscope system and a processing system, wherein:
the integrated endoscope system comprises a plurality of functional modules, a central control unit for controlling the plurality of functional modules to work and a power supply device for supplying power to the plurality of functional modules; the functional modules comprise a pneumoperitoneum instrument, a man-machine interaction screen, a cold light source and a camera for inflating a cavity, and the pneumoperitoneum instrument, the man-machine interaction screen, the cold light source and the camera are all connected to the central control unit; the cold light source and the camera are both connected with an optical endoscope; the cold light source provides a light source for the optical endoscope; the camera converts the optical signal collected by the optical endoscope into a video and sends the video to the man-machine interaction screen for endoscopic video display; wherein:
pneumoperitoneum appearance includes proportional control valve, switch solenoid valve, release valve, airflow sensor and pneumatic pressure sensor, wherein: the proportional control valve, the switch electromagnetic valve, the air release valve, the air flow sensor and the air pressure sensor are all arranged on an air supply pipeline and are all connected with the central control unit; the air flow sensor is used for detecting an air flow parameter of the air supply pipeline; the air pressure sensor is used for detecting air pressure parameters of the air supply pipeline; the gas supply pipeline is a pipeline for inputting gas into the body cavity of a patient by the pneumoperitoneum instrument; the central control unit is used for acquiring the gas flow parameter and the air pressure parameter, and outputting a PWM (pulse width modulation) signal according to the gas flow parameter and a preset standard gas flow parameter when the air pressure parameter is smaller than a preset first air pressure parameter, wherein the PWM signal is used for adjusting the on-off time proportion of the control valve so as to adjust the gas flow parameter in the gas supply pipeline; the proportional control valve and the switch electromagnetic valve are also used for closing when the air pressure parameter is greater than or equal to the first air pressure parameter; the air release valve is also used for opening the air release valve when the air pressure parameter is greater than or equal to a second air pressure parameter; wherein the second air pressure parameter is greater than the first air pressure parameter;
the power supply device comprises a switching power supply and a plurality of electromagnetic interference suppression circuits; the switching power supply comprises a first rectifying circuit, a transformer and a plurality of second rectifying circuits which are connected in sequence; the first rectifying circuit is used for converting the input alternating current into unidirectional voltage required by the transformer; the transformer comprises a primary winding and a plurality of secondary windings, the primary winding is connected with the output end of the first rectifying circuit, and the plurality of secondary windings are connected with the input ends of the plurality of second rectifying circuits in a one-to-one correspondence manner; each second rectifying circuit is used for converting the output voltage of the corresponding secondary winding into direct-current voltage required by the corresponding functional module; the input ends of the electromagnetic interference suppression circuits are connected with the output ends of the second rectification circuits in a one-to-one correspondence manner, and the output ends of the electromagnetic interference suppression circuits are used for being connected with the input ends of the functional modules in a one-to-one correspondence manner and are used for suppressing bidirectional electromagnetic interference between the functional modules and the switching power supply; the secondary windings, the second rectifying circuit, the electromagnetic interference suppression circuit and the functional modules are the same in number;
the processing system is configured to perform: s100, acquiring a video converted by the camera; s200, selecting key frame images from the video according to the time and the color of each frame image in the video; s300, inputting each key frame image into a preset trained YOLO target detection model to obtain a plurality of images with target positioning frames and target category identifications; s400, synthesizing the plurality of images with the target positioning frames and the target category identification to obtain a target positioning video, and displaying the target positioning video; wherein the training process of the YOLO target detection model at least comprises: and clustering the training sample data set by adopting a K-centers clustering method.
(III) advantageous effects
The embodiment of the invention provides an intelligent endoscope system with a pneumoperitoneum control and central control module, which has the following effects:
(1) because the proportional control valve is arranged in the pneumoperitoneum instrument, and the central control unit adopts a PWM algorithm to realize the adjustment of the on-off proportional time of the proportional control valve, the fine adjustment of the gas flow parameter in the gas supply pipeline is realized, and further the fine adjustment of the gas pressure is realized, namely, the gas flow in the gas supply pipeline is adjusted in a stepless manner, so that the precise and stable adjustment of the pneumoperitoneum pressure is realized, the problem of image jitter caused by the over-fast transient change of the gas pressure brought by the switching moment of the on-off valve is solved, and the image quality is improved;
(2) in the switching power supply, a secondary winding and a second rectifying circuit are provided for each functional module, thereby forming a power supply circuit for the functional module. Aiming at the N functional modules, N mutually independent power supply circuits are formed, so that interference signals cannot be transmitted among the functional modules through a shared power line. On the basis, an electromagnetic interference suppression circuit is arranged in each functional module and the corresponding second rectifying circuit and is used for suppressing the transmission of electromagnetic interference generated by the functional module to the switching power supply through the working power supply line and also suppressing the electromagnetic interference of the switching power supply to the functional modules through the working power supply line, namely, the bidirectional electromagnetic interference between the functional modules and the switching power supply is suppressed, and further electromagnetic interference signals are prevented from being transmitted among the functional modules;
(3) the processing system firstly extracts the key frame images in the video, then adopts a pre-trained YOLO target detection model to position targets in the key frame images and determine target types, and then synthesizes the images with target positioning frames and target category identifications to obtain a dynamic target positioning video. As the training process of the pre-trained YOLO target detection model comprises clustering the training sample data set by adopting a K-centers clustering method, the problem of noise sensitivity can be effectively improved by adopting the K-centers clustering method, and the picture quality of the target positioning video can be improved. Meanwhile, the invention adopts the target detection model to carry out target positioning and target type identification, has high processing efficiency and high processing speed, and can realize real-time target positioning and target type identification.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 shows a schematic diagram of a smart endoscope system configured with a pneumoperitoneum control and central control module in an embodiment of the invention;
FIG. 2 is a schematic diagram of a switch unit according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of an EMI suppression circuit according to an embodiment of the present invention;
1 first rectification circuit
2 Transformer device
3 Switch voltage-stabilizing control unit
4 Secondary feedback unit
5 Second rectification circuit
21 Primary winding
22 Secondary winding
10 Schottky diode
11 Differential mode inductor
12～14 First capacitor
15 Common mode inductor
16、17 Second capacitor
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides an intelligent endoscope system configured with a pneumoperitoneum control and central control module, as shown in figure 1, the intelligent endoscope system comprises an integrated endoscope system and a processing system, wherein:
the integrated endoscope system comprises a plurality of functional modules, a central control unit for controlling the plurality of functional modules to work and a power supply device for supplying power to the plurality of functional modules; the functional modules comprise a pneumoperitoneum instrument, a man-machine interaction screen, a cold light source and a camera for inflating a cavity, and the pneumoperitoneum instrument, the man-machine interaction screen, the cold light source and the camera are all connected to the central control unit; the cold light source and the camera are both connected with an optical endoscope; the cold light source provides a light source for the optical endoscope; the camera converts the optical signal collected by the optical endoscope into a video and sends the video to the man-machine interaction screen for endoscopic video display.
The processing system is used for carrying out target positioning to obtain a target positioning video.
The following describes each part of the intelligent cavity mirror system in detail:
(1) control of the functional modules for each functional module and for the central control unit
(1.1) for a central control unit
The central control unit can be called as CCU, can adopt a 32-bit single-chip microcomputer, is suitable for self-creative communication specifications, is used as a host, is used as a slave, can be communicated with the pneumoperitoneum instrument, the cold light source, the camera and the man-machine interaction screen through a 485 interface and a 485 concentrator, can specifically integrate and pack running state information of the pneumoperitoneum instrument, the cold light source, the endoscope and other functional modules, and sends the information to the man-machine interaction screen to be displayed on the man-machine interaction screen, and can also send related information (such as videos) to a PC (such as a classroom and a conference room) through a DVI communication interface.
Wherein, the communication specification is communication protocol, adopts RTU mode, and the data frame structure is: the header (1 byte), address (1 byte), data length (1 byte), command (1 byte), data (N bytes), and CRC check (2 bytes) (the commands are all sent in hexadecimal), which specify various command formats of the master, the response format of the slave, the address of the slave, and the broadcast format of the master.
The central control unit may specifically process each task by using a multi-task coordination method, specifically: the main types of tasks that the central control unit needs to handle include: external control information (e.g., control information of a cold light source, control information of the pneumoperitoneum apparatus, etc.) and alarm information, internal control information (e.g., automatic adjustment information and feedback of the pneumoperitoneum apparatus, etc.), information uploading tasks and display tasks. In order to process each task more reasonably, a priority scheduling algorithm is used for grading the tasks, the task processed first is determined according to the priority, and the priority is as follows: external control information, alarm information, internal self control information, an information uploading task and a display task; each type of task above includes some small tasks, so we adopt a time slice rotation algorithm to process the tasks in the same stage.
(1.2) the pneumoperitoneum instrument
Pneumoperitoneum appearance includes proportional control valve, switch solenoid valve, release valve, airflow sensor and pneumatic pressure sensor, wherein:
the proportional control valve, the switch electromagnetic valve, the air release valve, the air flow sensor and the air pressure sensor are all arranged on an air supply pipeline and are all connected with the central control unit; the air flow sensor is used for detecting an air flow parameter of the air supply pipeline; the air pressure sensor is used for detecting air pressure parameters of the air supply pipeline; the gas supply pipeline is a pipeline for inputting gas into the body cavity of a patient by the pneumoperitoneum instrument;
the central control unit is used for acquiring the gas flow parameter and the air pressure parameter, and outputting a PWM (pulse width modulation) signal according to the gas flow parameter and a preset standard gas flow parameter when the air pressure parameter is smaller than a preset first air pressure parameter, wherein the PWM signal is used for adjusting the on-off time proportion of the control valve so as to adjust the gas flow parameter in the gas supply pipeline; the proportional control valve and the switch electromagnetic valve are also used for closing when the air pressure parameter is greater than or equal to the first air pressure parameter; the air release valve is also used for opening the air release valve when the air pressure parameter is greater than or equal to a second air pressure parameter; wherein the second air pressure parameter is greater than the first air pressure parameter;
it will be appreciated that the gas supply line may also be referred to as an output line, i.e. a gas line that outputs gas from the pneumoperitoneum apparatus into the body cavity of the patient.
It will be appreciated that the gas here may be carbon dioxide gas.
It will be appreciated that the gas pressure parameter is pressure and the gas flow parameter is gas velocity.
It can be understood that the proportional control valve, the on-off solenoid valve and the air release valve mentioned later are all arranged on the air supply pipeline, are all connected with the central control unit, and can be connected with the central control unit through the MOS pipe.
It can be understood that the central control unit adopts a PWM algorithm to control the proportional control valve, specifically: the central control unit compares the airflow quantity parameter (such as airflow speed) detected by the airflow quantity sensor with a set standard airflow quantity parameter, and if the airflow quantity parameter is smaller than the standard airflow quantity parameter, the duty ratio of the output PWM signal is increased, so that the pulse width of the airflow at the output end is increased, and the output airflow is increased; and if the airflow parameter is larger than the standard airflow parameter, the duty ratio of the output PWM signal is reduced, so that the pulse width of the airflow at the output end is reduced, and the output airflow is reduced. Therefore, the PWM algorithm adopted by the central control unit can realize fine adjustment of the air flow at the output end of the air supply pipeline, so that the air flow is kept near the preset standard air flow parameter.
And outputting a corresponding PWM signal to the MOS tube according to the comparison result, controlling the on-off time proportion of the proportional control valve through the MOS tube, and further controlling the on-off time proportion of the proportional control valve, so that a series of high-frequency pulse airflows with equal amplitudes are obtained after the air in the air supply pipeline passes through the proportional control valve, namely 'breathing-stopping' type air supply is realized. The gas flow of the high-frequency pulse gas flows is equivalent to the gas flow of one switching turn in the traditional switching mode within a certain time, so that the precise control of the gas flow and the smooth gas transmission can be realized. Here, the adjustment of the on-off time ratio of the proportional control valve can realize the adjustment of the width of the pulse airflow, thereby changing the airflow size at the output end of the air supply pipeline.
It can be understood that the PWM algorithm is adopted to adjust the size of the air flow at the output end of the air supply pipeline, so that the pressure of the output air of the air supply pipeline is adjusted, and the PWM algorithm is used for finely adjusting the air flow, so that the adjustment of the pressure of the output air is also finely adjusted. In practical applications, in the case that the pressure parameter (e.g., pressure) detected by the pressure sensor is smaller than the preset first pressure parameter, the PWM algorithm may be used to indirectly adjust the gas pressure. However, there may be a situation where the pressure parameter (e.g., pressure) detected by the pressure sensor exceeds or reaches the preset first pressure parameter, which may be because the pressure of the output gas cannot be adjusted by the PWM algorithm due to a failure at a certain portion of the pneumoperitoneum apparatus, and at this time, the proportional control valve and the switching solenoid valve may be closed, i.e., the output of the gas supply line is stopped until the pressure parameter detected by the pressure sensor is again smaller than the first pressure parameter.
However, when the air pressure in the air supply pipeline cannot be reduced by closing the proportional control valve and the on-off solenoid valve, in order to avoid the occurrence of dangerous conditions caused by excessive air pressure, an air release valve can be arranged in the pneumoperitoneum instrument, the air release valve can be arranged on the air supply pipeline and connected with the central control unit, and when the central control unit determines that the air pressure parameter detected by the air pressure sensor is greater than or equal to the second air pressure parameter, the air release valve is opened to release air until the air pressure in the air supply pipeline is lower than the first air pressure parameter.
It is understood that the second pressure parameter is greater than the first pressure parameter. And when the air pressure parameter detected by the air pressure sensor is greater than or equal to the first air pressure parameter, closing the proportional control valve and the switch control valve. And when the air pressure parameter detected by the air pressure sensor is greater than or equal to the second air pressure parameter, the air escape valve is opened on the basis of closing the proportional control valve and the switch control valve.
Certainly, when air supply is not needed, the proportional control valve and the switch electromagnetic valve can be closed simultaneously, mechanical clamping inside the proportional control valve and leakage caused by untight closing are prevented, and double protection of the air supply pipeline is achieved.
In practical application, a gas pressure sensor for detecting the input gas pressure of the carbon dioxide tank can be further arranged to monitor the gas quantity in the carbon dioxide tank, and when the gas quantity is lower than a specific value, an alarm is given out to remind medical staff.
The proportional control valve is arranged in the pneumoperitoneum instrument, and the on-off proportional time of the proportional control valve is adjusted by adopting a PWM (pulse width modulation) algorithm through the central control unit, so that the fine adjustment of air flow parameters in the air supply pipeline is realized, and further the fine adjustment of air pressure is realized, namely, the air flow in the air supply pipeline is adjusted in a stepless manner, so that the precise and stable adjustment of the pneumoperitoneum pressure is realized, the problem of visual image jitter caused by the fact that the air pressure changes too fast instantly due to the instant opening and closing of the switch valve is solved, and the image quality is improved.
In some embodiments, the integrated cavity mirror system may further comprise a storage unit, wherein:
the storage unit is connected with the central control unit and is used for storing data sets corresponding to a plurality of age groups;
correspondingly, a plurality of age bracket options are displayed on the human-computer interaction screen, and when any one of the age bracket options is detected to be selected, a trigger signal is sent to the central control unit;
and the central control unit is also used for determining an air pressure recommended value and an air flow recommended value for the age group according to the data set corresponding to the age group corresponding to the selected option when receiving the trigger signal, and displaying the air pressure recommended value and the air flow recommended value on the man-machine interaction screen.
In practical applications, each age group may include: neonates, children, adults, the elderly, etc.
In practical applications, each age group data set may include a plurality of records, and each record may include a barometric pressure, a flow rate, a number of uses at the barometric pressure and flow rate setting, an average temperature of the gas when the abdominal pressure is inflated to within a specified barometric pressure threshold, an average time taken for the abdominal pressure to inflate to within the specified barometric pressure threshold, and a timestamp of the record. For example, { p:8, s:20, ts:20, c:20, t:480, st:1528969758 }. Where p is pressure, s is gas flow, ts is number of uses, c is gas average temperature, t is average elapsed time, and st is time stamp.
It can be understood that the medical staff can select any one of a plurality of age bracket options on the human-computer interaction screen, the central control unit extracts the data set corresponding to the selected option from the storage unit, further calculates the air flow recommendation value and the air pressure recommendation value, and displays the two recommendation values on the human-computer interaction screen for the medical staff to refer to, for example, the first air pressure parameter and/or the second air pressure parameter is set according to the air pressure recommendation value, and the standard air flow parameter is set according to the air flow recommendation value.
Here, the corresponding recommended airflow value and recommended air pressure value are calculated for each age group, so that the recommended values are more valuable to reference in consideration of the difference between different age groups when calculating the recommended values.
The average gas temperature in the data set is determined according to the temperatures detected a plurality of times, so that a temperature sensor is required, and the temperature sensor can be arranged on the gas supply pipeline and connected with the central control unit for detecting the gas temperature in the gas supply pipeline.
Wherein the central control unit may determine the air pressure recommendation value and the air flow recommendation value by:
a1, calculating the suitable score value of each recorded content by adopting a first formula, wherein the first formula comprises:
Figure BDA0001782364450000111
wherein g is a suitable score value, a1、a2And a3Is a weight value, tsmaxIs the value of the maximum number of uses in the plurality of pieces of recorded content, tmaxThe value of the average time taken for the plurality of pieces of recorded content to be the maximum, c is the average temperature of the gas when the abdominal cavity pressure in the piece of recorded content is inflated to within a specified pressure threshold range, t is the average time taken for the abdominal cavity pressure in the piece of recorded content to be inflated to within the specified pressure threshold range, and ts is the number of uses in the piece of recorded content.
The central control unit may specifically calculate the weight value a in the first formula according to a third formula1The third formula bagComprises the following steps:
a1(t)＝a0e-25(t-1)
where t is the number of weeks in which the time stamp of the currently recorded content is located in units of time of weeks, and the number of weeks is smaller as the time stamp is closer.
For example, if the timestamp of the currently recorded content is within the last week, t is 1; if the timestamp of the currently recorded content is within the last two weeks, t is 2.
Wherein, three weight values a1、a2And a3Is 1, i.e. a1+a2+a3＝1。
And A2, taking the air pressure in the recorded content corresponding to the maximum value in the obtained plurality of suitable score values as the air pressure recommended value and taking the air flow in the recorded content corresponding to the maximum value as the air flow recommended value.
It can be understood that, in the recorded content, the more times of use, the better the combination of the air pressure and the air flow, the closer the average temperature is to the body temperature, the better the inflation time is, and the higher the g value, the more suitable the combination of the air pressure and the air flow is for the patient, so that the air pressure in the recorded content corresponding to the maximum g value is taken as the air pressure recommended value, and the air flow in the recorded content corresponding to the maximum g value is taken as the air flow recommended value.
In some embodiments, the central control unit may also update or delete the recorded content in the data set in the storage unit, thereby ensuring the data validity of the data set.
The process that the central control unit updates the data set in the storage unit comprises the following steps:
when the pneumoperitoneum instrument is used, recording the current air pressure, the current air flow, the current air temperature, the time for inflating the abdominal cavity air pressure to the specified air pressure threshold range and a timestamp when the difference value between the first abdominal cavity air pressure and the preset pressure is kept within the specified threshold for a preset time; if the current record content comprises the record of the current air pressure and the current air flow, adding 1 to the corresponding use times, updating the time stamp, and calculating the updated average used time and the updated average gas temperature by adopting a second formula, wherein the second formula comprises:
Figure BDA0001782364450000121
wherein n is the updated average time or the updated average temperature of the gas, o is the updated average time or the updated average temperature of the gas, ts is the number of times of use before the update, and r is the time taken for the inflation of the abdominal cavity pressure to the specified pressure threshold range or the current gas temperature.
For example, if the average gas temperature was 37.2 ℃ in the first 10 times and 37.1 ℃ in the present time, the new average gas temperature is n ═ 37.2 × 9+ 37.1)/10.
Of course, if there is no current recorded content in the storage unit, the current recorded content is added to the data set, and the number of uses is 1.
Here, the update of the number of times of use is realized by adding 1 to the number of times of use, the update of the time stamp is realized by replacing the old time stamp with the new time stamp, the update of the average elapsed time is realized by replacing the old average elapsed time with the new average elapsed time, and the update of the average gas temperature is realized by replacing the old average gas temperature with the new average gas temperature.
The process of deleting the recorded content in the data set in the storage unit by the central control unit may include: and if the number of the recorded contents in the data set corresponding to each age group stored in the storage unit is greater than the preset maximum number, deleting the recorded content with the least use frequency from the recorded contents with the minimum time stamp in the data set corresponding to the age group and in the preset number.
That is, when the number of pieces of recorded content in the data set corresponding to a certain age group is greater than a certain number, the recorded content with the smallest number of usage times among the certain number of recorded content with the smallest time stamp is deleted, so that the recorded content with the earliest time and the smallest number of usage times can be deleted.
In the traditional technology, in the use process of the pneumoperitoneum instrument, medical staff need to set the pressure and the air flow by combining self experience, two factors of temperature and inflation time are difficult to consider, and the medical staff are provided with a proper combination of the pressure and the air flow according to the combination of the air temperature, the inflation time, the pressure, the air flow and the use times of the patient in the inflation process in historical data.
The pneumoperitoneum instrument is provided with the proportional control valve, and the on-off proportional time of the proportional control valve is adjusted by adopting a PWM algorithm through the central control unit, so that the fine adjustment of gas flow parameters in the gas supply pipeline is realized, the fine adjustment of the gas pressure is realized, namely, the gas flow in the gas supply pipeline is adjusted in a stepless manner, the precise and stable adjustment of the pneumoperitoneum pressure is realized, the problem of visual image jitter caused by the fact that the gas pressure changes too fast instantly due to the instant opening and closing of the switch valve is solved, and the image quality is improved.
(1.3) for Cold light sources
The cold light source includes light source module, constant current board and radiator, wherein:
the light source module comprises an LED bulb, a light gathering tube and a light guide beam;
the constant current plate and the radiator are both connected to the central control unit, and the central control unit is used for controlling the constant current plate to perform deviation adjustment on the output current of the light source module by adopting a PID algorithm, monitoring the light source temperature of the light source module in real time, and driving the radiator to dissipate heat when the light source temperature rises to a certain value.
It can be understood that the light-gathering cylinder and the light-guiding beam can process the light generated by the LED bulb, so that the output light can meet the use requirement.
It can be understood that, because the central control unit adopts the PID algorithm to control the output current, the constant current board can be ensured to quickly adjust the deviation, and the stability of constant current output is maintained.
The central control unit can monitor the temperature of the light source in real time, and the heat radiator can be driven to reduce the heat of light when the temperature of the light source is increased to a certain value, so that the condition that the light is irradiated near a wound for a long time to cause internal tissue burns in a minimally invasive surgery is avoided.
(1.4) for video cameras
The camera includes CCD camera and mainboard of making a video recording, wherein:
the CCD camera is used for converting optical signals collected by the optical endoscope into electric signals, and the camera main board is used for converting the electric signals into videos and sending the videos to the man-machine interaction screen and the processing system.
(1.5) for human-computer interaction screens
The man-machine interaction screen can also be called as a monitor, and can also display the working conditions of functional modules such as an optical endoscope, a pneumoperitoneum instrument and a cold light source.
Certainly, the human-computer interaction screen can also be provided with an input interface of control information of the pneumoperitoneum instrument and the cold light source, so that medical personnel can input the control information of the pneumoperitoneum instrument and the cold light source.
It can be understood that all the functional modules are integrated into a whole, and are integrated into a whole, for example, integrated into a case, so that the portable intelligent minimally invasive endoscope equipment is formed and has an endoscope video processing function. Because the intelligent minimally invasive endoscope equipment adopts an integrated structure, the occupied area is small, the movement and the carrying are convenient, the compatibility is good, and the intelligent minimally invasive endoscope equipment can be used for minimally invasive surgery in various environments.
(2) For power supply device
The power supply device comprises a switching power supply and a multi-electromagnetic interference suppression circuit; wherein:
as shown in fig. 2, the switching power supply includes a first rectification circuit 1, a transformer 2, and a plurality of second rectification circuits 5 connected in sequence; the first rectifying circuit 1 is used for converting the input alternating current into a unidirectional voltage required by the transformer 2; the transformer 2 comprises a primary winding 21 and a plurality of secondary windings 22, the primary winding 21 is connected with the output end of the first rectifying circuit 1, and the plurality of secondary windings 22 are connected with the input ends of the plurality of second rectifying circuits 5 in a one-to-one correspondence manner; each second rectifying circuit 5 is used for converting the output voltage of the corresponding secondary winding 22 into the direct-current voltage required by the corresponding functional module;
the input ends of the electromagnetic interference suppression circuits are connected with the output ends of the second rectification circuits 5 in a one-to-one correspondence manner, and the output ends of the electromagnetic interference suppression circuits are connected with the input ends of the functional modules in a one-to-one correspondence manner and used for suppressing bidirectional electromagnetic interference between the functional modules and the switching power supply.
It can be understood that the first rectifying circuit 1 can convert the input AC power into the unidirectional high voltage required by the transformer 2, for example, to convert the power grid supply AC 220V. The first rectification circuit 1 may adopt an EMI and rectification circuit, that is, the first rectification circuit 1 includes an electromagnetic interference filter module and a rectification filter module. Of course, the first rectification circuit 1 may also adopt other circuit modules to realize the conversion of the alternating current.
In practice, the output voltage of the secondary winding 22 may be non-direct current, that is, the voltage output by the secondary winding 22 has a certain waveform variation, but each voltage value on the waveform is a positive value. Each second rectifying circuit 5 converts the voltage output by the corresponding secondary winding 22 into direct current for the corresponding functional module to use.
It can be understood that, when the power supply device provided by the present invention is applied to an endoscope all-in-one machine, N functional modules are provided in the endoscope all-in-one machine, N secondary windings 22, N second rectification circuits 5 and N electromagnetic interference suppression circuits may be provided in the power supply device, the N secondary windings 22 and the N second rectification circuits 5 are connected in a one-to-one correspondence manner, and the N second rectification circuits 5 and the N electromagnetic interference suppression circuits are connected in a one-to-one correspondence manner.
It will be appreciated that in the switching power supply shown in fig. 2, the right side is the input terminal of the switching power supply, and the left side is the output terminal of the switching power supply.
In the switching power supply, one secondary winding 22 and one second rectification circuit 5 are provided for each functional module, thereby forming one power supply circuit for the functional module. Aiming at the N functional modules, N mutually independent power supply circuits are formed, so that interference signals cannot be transmitted among the functional modules through a shared power line. On this basis, an electromagnetic interference suppression circuit is further arranged in each functional module and the corresponding second rectification circuit 5, and is used for suppressing electromagnetic interference generated by the functional module per se from being transmitted to the switching power supply through the working power supply line, and also suppressing electromagnetic interference generated by the switching power supply to the functional module through the working power supply line, that is, suppressing bidirectional electromagnetic interference between the functional module and the switching power supply, and further avoiding electromagnetic interference signals from being transmitted among the functional modules.
In some embodiments, the switching power supply may further include, in addition to the first rectification circuit 1, the transformer 2, and the second rectification circuit 5, a secondary feedback unit 4 provided on one secondary winding 22 of the plurality of secondary windings 22, and a switching regulator control unit 3 provided on the primary winding 21, wherein: the secondary feedback unit 4 is configured to feed back a voltage signal of the secondary winding 22 to the switching regulator control unit 3, and the switching regulator control unit 3 is configured to perform regulator control on the output voltage of the secondary winding 22 according to the voltage signal.
Here, the secondary feedback unit 4 feeds back the voltage signal on the secondary winding 22 to the switching regulator control unit 3, and then the switching regulator control unit 3 adjusts the output voltage of the secondary winding 22 according to the voltage signal, so as to implement the regulator control, and further ensure the working performance of the switching power supply.
In some embodiments, the emi suppression circuit may be implemented in a variety of configurations, as shown in fig. 3, wherein one alternative configuration is: each electromagnetic interference suppression circuit comprises a Schottky diode 10, a differential mode LC filtering unit and a common mode LC filtering unit which are connected in sequence; wherein:
the anode of the Schottky diode 10 is connected with the positive output end of the corresponding second rectifying circuit 5;
the differential mode LC filtering unit comprises a differential mode inductor 11 and at least two first capacitors connected in parallel; one end of the differential mode inductor 11 is connected with the cathode of the schottky diode 10, the at least two parallel first capacitors are arranged between the other end of the differential mode inductor 11 and the negative output end of the corresponding second rectifying circuit 5, and the negative output end of the corresponding second rectifying circuit 5 is grounded;
the common mode LC filtering unit comprises a common mode inductor 15 and at least two second capacitors connected in parallel, one end of one winding in the common mode inductor 15 is connected with the other end of the differential mode inductor 11, one end of the other winding in the common mode inductor 15 is connected with the negative output end of the corresponding second rectifying circuit 5, and the at least two second capacitors connected in parallel are arranged between the other end of one winding of the common mode inductor 15 and the other end of the other winding of the common mode inductor 15; the other end of one winding of the common mode inductor 15 is used as a positive output end of the electromagnetic interference suppression circuit, and the positive output end is used for connecting a positive input end of a corresponding functional module; the other end of the other winding of the common mode inductor 15 is used as a negative output end of the electromagnetic interference suppression circuit, and the negative output end is used for being connected with a negative input end of a corresponding functional module.
It can be understood that the schottky diode 10 has the characteristics of ultrafast current recovery and unidirectional cut-off conduction, and can only allow the switching power supply to transmit signals to the functional module, and reduce or even prevent the functional module from transmitting electromagnetic interference signals to the switching power supply through the working power supply line, so as to suppress electromagnetic interference caused by the served functional module to other functional modules through the working power supply line.
In practical application, the differential-mode inductor 11 in the differential-mode LC filter unit may be a 10uH magnetic loop inductor, and such an inductor has a characteristic of preventing current from suddenly changing. The number of at least two parallel first capacitors in the differential mode LC filtering unit can be set according to needs, for example, three parallel first capacitors are set, the three first capacitors can adopt a 10uF NPO material capacitor 12, a 100nF NPO material capacitor 13 and a 100pF NPO material capacitor 14, so that a lower equivalent resistance can be achieved in the frequency range of 1MHz to 300MHz, and a good absorption and suppression effect on electromagnetic interference signals is achieved.
It can be understood that the common mode inductor 15 in the common mode LC filtering unit can well suppress the common mode signal coupled on the working power line, and can reduce the coupling of the external electromagnetic interference signal to the working power line of the functional module, thereby suppressing the transmission through the working power line in the form of common mode interference. The number of the at least two parallel second capacitors in the common mode LC filter unit may be selected according to need, for example, two parallel second capacitors 16 and 17 are provided. The second capacitor may be a decoupling high frequency capacitor. The common mode inductor 15 and the second capacitors 16 and 17 constitute a filter circuit.
It is understood that in the emi suppression circuit shown in fig. 3, the left side is the input terminal of the emi suppression circuit, and the right side is the output terminal of the emi suppression circuit.
Because the differential mode interference signal and the common mode interference on the working power line coexist, the bidirectional electromagnetic interference suppression circuit formed by combining the Schottky diode 10, the differential mode LC filtering unit and the common mode LC filtering unit can effectively suppress the transmission of the electromagnetic interference signal to the functional module by the switching power supply through the working power line, can also effectively suppress the transmission of the electromagnetic interference generated by the functional module to the switching power supply through the working power line, reduces the transmission of the electromagnetic interference of the functional module to the outside through space radiation, and improves the quality of the working power supply of the functional module.
In the above specific structure of the electromagnetic interference suppression circuit, the negative output end of the second rectification circuit 5 is grounded, and the specific implementation manner may be that the negative output end of the second rectification circuit 5 is directly grounded, or the negative output end of the second rectification circuit 5 is indirectly grounded, for example, the negative output end of the second rectification circuit 5 is connected to a metal shell of the endoscope all-in-one machine, and the metal shell is connected to a ground wire of a power supply grid through a protection ground wire of a three-phase power line, so as to realize the grounding of the negative output end of the second rectification circuit 5.
(3) For a processing system
S100, acquiring a video converted by the camera;
s200, selecting key frame images from the video according to the time and the color of each frame image in the video;
it can be understood that the key frame is extracted according to the time and color of the image, and actually the key frame is extracted by adopting a method of changing the time and color of the image. The time change can fully reflect the global information of the image, and the color characteristic can reflect the local change information of the image.
For example, the key frame image may be extracted by:
s201, taking a first frame image in a video as a key frame image, and setting d to be 2;
it is understood that d represents the frame number of the picture in the video, for example, d-2 represents the frame number of the 2 nd frame picture in the video.
S202, calculating
Figure BDA0001782364450000191
Wherein S isiFor the ith frame image, s in the videoi＝s(ti，ci)，tiC is the time point of the ith frame image in the videoiIs the color matrix of the ith frame image.
For example, s'2＝s2-s1，s'3＝(s2-s1)+(s3-s2)。
S203, judging S'dWhether the current time is greater than a corresponding preset threshold value, wherein s'dThe corresponding preset threshold is m × β, m is the total frame number of the current key frame image, β is a constant:
if yes, taking the d frame image in the video as a key frame image, and entering the step S204;
otherwise, the process proceeds to step S204.
It can be understood that s'dFor measuringDifference, s ', between images based on temporal and color variation features'dThe larger the difference between the images is, so that the images with high similarity can be removed, and the images with obvious difference are reserved as the key frame images.
Here, through s'dAnd comparing the frame number with a preset threshold value to judge whether the frame number d image in the video is a key frame image.
S204, judging that d is less than the total frame number of the video:
if yes, increasing the value of d by 1, and returning to the step S202;
otherwise, ending the key frame image extraction process;
here, by comparing d with the total number of frames, the key frame image extraction process is ended only when d is equal to the total number of frames, thereby realizing traversal of each frame image in the video.
Of course, other ways may also be adopted to extract the key frames in the video, and the above steps S201 to S204 are only one specific way.
S300-, smoothing the edge black edge of the key frame image according to the vision field parameters of the endoscope lens, filtering and denoising the smoothed image by adopting a high-pass filter, and filtering and enhancing the filtered and denoised image by adopting a median filter.
Understandably, the black edge of the key frame image is subjected to smoothing treatment, so that the endoscope image with clear boundary can be obtained. And filtering by adopting a high-pass filter and a median filter to obtain a high-frequency part which removes the noise in the key frame image and keeps the key frame image.
Step S300-is provided here at step S300-which mainly implements an optimization process for the key frame image, which is not necessary to achieve the basic object of the present invention, and thus step S300-may not be included in some embodiments.
S300, inputting each key frame image into a preset trained YOLO target detection model to obtain a plurality of images with target positioning frames and target category identifications;
wherein the training process of the YOLO target detection model at least comprises: and clustering the training sample data set by adopting a K-centers clustering method.
The specific process of clustering the training sample data set by adopting the K-centers clustering method comprises the following steps: the mass point after each iteration is selected from the sample points of the cluster, and the standard is selected to select the object closest to the average value in the cluster as the cluster center, so that the problem of noise sensitivity can be effectively improved. The K-centers clustering method can be used for clustering the actual set (namely the group channel) in the training sample data set, so that the statistical rule of the group channel is found. And taking the number k of clusters as the number of candidate frames (namely anchor boxes), and taking the width and height dimensions of the frames at the centers of the k clusters as the dimensions of the candidate frames.
However, the K-means clustering method used by the original YOLO neural network in the prior art is very sensitive to "noise", so that the problem of "noise" exists in the image under the mobile endoscope. In contrast, the invention can effectively improve the problem of 'noise' sensitivity and improve the picture quality.
In addition, the YOLO target detection model provided by the invention has other differences: the network structure of the YOLO target detection model includes a pooling layer, which can sequentially order n activation function values from small to large, sequentially order n weight values from small to large, multiply the n weight values with the corresponding activation function values, respectively, calculate an average value of the n multiplication results, and take the average value as a final activation function value.
The pooling layer employed in the present invention may be referred to as sort-pooling, and specifically arranges the n activation functions in increasing order: { a1，a2，a3...an}(a1＜a2＜a3< …) instead of selecting the largest one. With n weights w1，w2，w3...wnMultiplying it to obtain n values, and averaging the n values, i.e.
Figure BDA0001782364450000211
In this way, the neural network can still learn pairsShould be given by { w1，w2，w3...wnGood, old maximum pooling of {0,0,0 … 1}, and subsequent layers can get more information, with the gradient flowing through all values in the previous layer when propagating backwards. The sort-firing can achieve faster and better convergence, optimize iteration time, retain more image information and highlight important image information, so that target positioning and identification are more accurate and processing efficiency is higher.
However, the pooling layer in the prior art is max-pooling, which means that the largest of the n activation functions is selected and the other activation functions are deleted. Max-firing therefore suffers from spatial information loss, inability to use information from multiple activation functions, and back-propagation that can only improve the maximum pooled activation function.
Here, optimization of the target detection model can be achieved by using the K-centers clustering method and sort-firing.
In this step, the process of determining the image with the target positioning frame and the target category identifier by using the YOLO target detection model may specifically include:
s201, dividing each key frame image into S-S grids, wherein S is an integer larger than 1;
s202, aiming at each grid, determining the position, confidence coefficient and object class probability of an object by adopting a plurality of candidate frames, and multiplying the confidence coefficient corresponding to each candidate frame by the object class probability to obtain a confidence score of the object in the candidate frame of the network belonging to each object class;
s203, filtering the candidate frames corresponding to the confidence scores lower than the preset threshold value, and reserving the candidate frames corresponding to the confidence scores higher than or equal to the preset threshold value;
s204, performing non-maximum suppression (namely NMS) processing on each candidate frame reserved in each key frame image to obtain an image with a target positioning frame and a target category identifier; and the target positioning frame corresponds to the target category identification one by one.
S400, synthesizing the plurality of images with the target positioning frames and the target category identification to obtain a target positioning video, and displaying the target positioning video.
In practical application, the object class identifier may be set beside the object positioning box to illustrate the class of the object.
The processing system firstly acquires the endoscope video, then extracts the key frame image, then positions the target in the key frame image by adopting a pre-trained YOLO target detection model and determines the target type, and then synthesizes the image with the target positioning frame and the target type identification to obtain the dynamic target positioning video. As the training process of the pre-trained YOLO target detection model comprises clustering the training sample data set by adopting a K-centers clustering method, the problem of noise sensitivity can be effectively improved by adopting the K-centers clustering method, and the picture quality of the target positioning video can be improved. Meanwhile, the invention adopts the target detection model to carry out target positioning and target type identification, has high processing efficiency and high processing speed, and can realize real-time target positioning and target type identification.
It will be appreciated that the above method performed by the processing system is implemented based on image processing technology, wherein the target can be set according to the requirement, for example, some abnormal conditions in the cavity, and the location of the abnormal conditions and the type of the abnormal conditions can be identified by the above target positioning method.
It will be appreciated that the processing system described above is a device independent of the integrated cavity mirror system, and that the specific hardware of the device may include a processor, a memory in which computer programs are stored which, when executed, implement the method described above, and a display for displaying the target positioning video resulting from the execution of the method described above. The device is connected with a camera in the integrated endoscope system, obtains a video converted by the camera, performs image processing based on the video to obtain a target positioning video, and then displays the target positioning video for medical staff to refer.
It is understood that the central control unit is the central control module mentioned in the title, and the pneumoperitoneum apparatus is the pneumoperitoneum control module mentioned in the title.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (8)

1. An intelligent endoscope system configured with pneumoperitoneum control and central control module, said intelligent endoscope system comprising an integrated endoscope system and processing system, wherein: the integrated endoscope system comprises a plurality of functional modules, a central control unit for controlling the plurality of functional modules to work and a power supply device for supplying power to the plurality of functional modules; the functional modules comprise a pneumoperitoneum instrument, a man-machine interaction screen, a cold light source and a camera for inflating a cavity, and the pneumoperitoneum instrument, the man-machine interaction screen, the cold light source and the camera are all connected to the central control unit; the cold light source and the camera are both connected with an optical endoscope; the cold light source provides a light source for the optical endoscope; the camera converts the optical signal collected by the optical endoscope into a video and sends the video to the man-machine interaction screen for endoscopic video display; wherein: pneumoperitoneum appearance includes proportional control valve, switch solenoid valve, release valve, airflow sensor and pneumatic pressure sensor, wherein:
the proportional control valve, the switch electromagnetic valve, the air release valve, the air flow sensor and the air pressure sensor are all arranged on an air supply pipeline and are all connected with the central control unit; the air flow sensor is used for detecting an air flow parameter of the air supply pipeline; the air pressure sensor is used for detecting air pressure parameters of the air supply pipeline; the gas supply pipeline is a pipeline for inputting gas into the body cavity of a patient by the pneumoperitoneum instrument; the central control unit is used for acquiring the gas flow parameter and the air pressure parameter, and outputting a PWM (pulse width modulation) signal according to the gas flow parameter and a preset standard gas flow parameter when the air pressure parameter is smaller than a preset first air pressure parameter, wherein the PWM signal is used for adjusting the on-off time proportion of the control valve so as to adjust the gas flow parameter in the gas supply pipeline; the proportional control valve and the switch electromagnetic valve are also used for closing when the air pressure parameter is greater than or equal to the first air pressure parameter; the air release valve is also used for opening the air release valve when the air pressure parameter is greater than or equal to a second air pressure parameter; wherein the second air pressure parameter is greater than the first air pressure parameter;
the power supply device comprises a switching power supply and a plurality of electromagnetic interference suppression circuits; the switching power supply comprises a first rectifying circuit, a transformer and a plurality of second rectifying circuits which are connected in sequence; the first rectifying circuit is used for converting the input alternating current into unidirectional voltage required by the transformer; the transformer comprises a primary winding and a plurality of secondary windings, the primary winding is connected with the output end of the first rectifying circuit, and the plurality of secondary windings are connected with the input ends of the plurality of second rectifying circuits in a one-to-one correspondence manner; each second rectifying circuit is used for converting the output voltage of the corresponding secondary winding into direct-current voltage required by the corresponding functional module; the input ends of the electromagnetic interference suppression circuits are connected with the output ends of the second rectification circuits in a one-to-one correspondence manner, and the output ends of the electromagnetic interference suppression circuits are used for being connected with the input ends of the functional modules in a one-to-one correspondence manner and are used for suppressing bidirectional electromagnetic interference between the functional modules and the switching power supply; the secondary windings, the second rectifying circuit, the electromagnetic interference suppression circuit and the functional modules are the same in number;
the processing system is configured to perform: s100, acquiring a video converted by the camera; s200, selecting key frame images from the video according to the time and the color of each frame image in the video; s300, inputting each key frame image into a preset trained YOLO target detection model to obtain a plurality of images with target positioning frames and target category identifications; s400, synthesizing the plurality of images with the target positioning frames and the target category identification to obtain a target positioning video, and displaying the target positioning video; wherein the training process of the YOLO target detection model at least comprises: clustering the training sample data set by adopting a K-centers clustering method;
the integrated cavity mirror system further comprises: the storage unit is connected with the central control unit and is used for storing data sets corresponding to a plurality of age groups;
correspondingly, a plurality of age bracket options are displayed on the human-computer interaction screen, and the human-computer interaction screen is further used for sending a trigger signal to the central control unit when any one of the age bracket options is detected to be selected;
the central control unit is further used for determining an air pressure recommended value and an air flow recommended value for the age group according to the data set corresponding to the age group corresponding to the selected option when the trigger signal is received, and displaying the air pressure recommended value and the air flow recommended value on the human-computer interaction screen;
the pneumoperitoneum instrument further comprises: the temperature sensor is arranged on the gas supply pipeline, is connected with the central control unit and is used for detecting the gas temperature in the gas supply pipeline;
the data set corresponding to each age group stored in the storage unit comprises a plurality of record contents, and each record content comprises air pressure, air flow, use times under the setting of the air pressure and the air flow, average air temperature when the abdominal cavity air pressure is inflated to a specified air pressure threshold range, average time for the abdominal cavity air pressure to be inflated to the specified air pressure threshold range and a timestamp of the record content;
correspondingly, the central control unit is configured to determine the recommended air pressure value and the recommended air flow value by:
calculating a suitable score value of each of the recorded contents using a first formula, the first formula including:
Figure FDA0002566481820000031
wherein g is a suitable score value, a1, a2 and a3 are weighted values, tsmax is a value at which the number of uses in the plurality of pieces of recorded content is the largest, tmax is a value at which the average time taken for the plurality of pieces of recorded content is the largest, c is an average temperature of gas at which the abdominal cavity pressure in the piece of recorded content is inflated to within a specified pressure threshold range, t is an average time taken for the abdominal cavity pressure in the piece of recorded content to be inflated to within the specified pressure threshold range, and ts is the number of uses in the piece of recorded content;
and taking the air pressure in the recorded content corresponding to the maximum value of the obtained plurality of suitable scoring values as the air pressure recommended value and taking the air flow in the recorded content corresponding to the maximum value as the air flow recommended value.
2. The intelligent endoscope system of claim 1,
the cold light source includes: light source module, constant current board and radiator, wherein:
the light source module comprises an LED bulb, a light gathering tube and a light guide beam;
the constant flow plate and the radiator are both connected to the central control unit, and the central control unit is used for controlling the constant flow plate to perform deviation adjustment on the output current of the light source module by adopting a PID algorithm, monitoring the light source temperature of the light source module in real time, and driving the radiator to dissipate heat when the light source temperature rises to a certain value; and/or the camera includes CCD camera and mainboard of making a video recording, wherein: the CCD camera is used for converting optical signals collected by the optical endoscope into electric signals, and the camera main board is used for converting the electric signals into videos and sending the videos to the human-computer interaction screen and the processing system.
3. The intelligent endoscope system of claim 1, wherein said central control unit is further configured to update the data set in said storage unit, specifically comprising: when the pneumoperitoneum instrument is used, recording the current air pressure, the current air flow, the current air temperature, the time for inflating the abdominal cavity air pressure to the specified air pressure threshold range and a timestamp when the difference value between the first abdominal cavity air pressure and the preset pressure is kept within the specified threshold for a preset time;
if the current record content comprises the record of the current air pressure and the current air flow, adding 1 to the corresponding use times, updating the time stamp, and calculating the updated average used time and the updated average gas temperature by adopting a second formula, wherein the second formula comprises:
Figure FDA0002566481820000041
wherein n is the updated average time or the updated average temperature of the gas, o is the updated average time or the updated average temperature of the gas, ts is the number of times of use before the update, and r is the time taken for the inflation of the abdominal cavity pressure to the specified pressure threshold range or the current gas temperature.
4. The intelligent endoscope system of claim 1, wherein each electromagnetic interference suppression circuit comprises a schottky diode, a differential mode LC filter unit and a common mode LC filter unit connected in sequence; wherein: the positive electrode of the Schottky diode is connected with the positive output end of the corresponding second rectifying circuit;
the differential mode LC filtering unit comprises a differential mode inductor and at least two first capacitors connected in parallel; one end of the differential mode inductor is connected with the cathode of the Schottky diode, the at least two first capacitors connected in parallel are arranged between the other end of the differential mode inductor and the negative output end of the corresponding second rectifying circuit, and the negative output end of the corresponding second rectifying circuit is grounded;
the common mode LC filtering unit comprises a common mode inductor and at least two second capacitors connected in parallel, one end of one winding in the common mode inductor is connected with the other end of the differential mode inductor, one end of the other winding in the common mode inductor is connected with the negative output end of the corresponding second rectifying circuit, and the at least two second capacitors connected in parallel are arranged between the other end of one winding in the common mode inductor and the other end of the other winding in the common mode inductor; the other end of one winding of the common mode inductor is used as a positive output end of the electromagnetic interference suppression circuit, and the positive output end is used for being connected with a positive input end of a corresponding functional module; and the other end of the other winding of the common mode inductor is used as a negative output end of the electromagnetic interference suppression circuit, and the negative output end is used for being connected with a negative input end of a corresponding functional module.
5. The intelligent endoscope system of claim 1, wherein said processing system selects key frame images from said video according to time and color of each frame image in said video, comprising: s201, taking a first frame image in the video as a key frame image, and setting d to be 2;
s202, calculating
Figure FDA0002566481820000051
S203, judging whether S'd is larger than a corresponding preset threshold, wherein the preset threshold corresponding to S'd is m beta, m is the total frame number of the current key frame image, and beta is a constant: if yes, taking the d frame image in the video as a key frame image, and entering the step S204;
otherwise, go to step S204;
s204, judging that d is less than the total frame number of the video:
if yes, increasing the value of d by 1, and returning to the step S202;
otherwise, ending the key frame image extraction process;
and Si is the ith frame image in the video, s (ti, ci) is the time point of the ith frame image in the video, and ci is the color matrix of the ith frame image.
6. The system as claimed in claim 1, wherein the YOLO target detection model includes a pooling layer in the network structure, the pooling layer being capable of sequentially sorting n activation function values from small to large, sequentially sorting n weight values from small to large, multiplying the n weight values by the corresponding activation function values, respectively, calculating an average value of the n multiplication results, and taking the average value as a final activation function value.
7. The intelligent endoscope system of claim 1, wherein the processing system is further configured to smooth the edge black edge of each key frame image according to the vision parameters of the endoscope lens before inputting the key frame image into a preset trained YOLO target detection model, filter and denoise the smoothed image with a high-pass filter, and filter and enhance the filtered and denoised image with a median filter.
8. The intelligent endoscope system of any one of claims 1-7, wherein the process of inputting each key frame image into a preset trained YOLO target detection model by the processing system to obtain a plurality of images with target positioning frames and target category identifiers comprises:
s301, dividing each key frame image into S-S grids, wherein S is an integer larger than 1;
s302, aiming at each grid, determining the position, confidence coefficient and target category probability of a target by adopting a plurality of candidate frames, and multiplying the confidence coefficient corresponding to each candidate frame by the target category probability to obtain the confidence score of the target in the candidate frame of the grid belonging to each target category;
s303, filtering the candidate frames corresponding to the confidence scores lower than the preset threshold value, and reserving the candidate frames corresponding to the confidence scores higher than or equal to the preset threshold value;
s304, carrying out non-maximum suppression processing on each candidate frame reserved in each key frame image to obtain an image with a target positioning frame and a target category identifier; and the target positioning frame corresponds to the target category identification one by one.
CN201810998268.7A 2018-08-29 2018-08-29 Intelligent endoscope system with pneumoperitoneum control and central control module Active CN109124702B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810998268.7A CN109124702B (en) 2018-08-29 2018-08-29 Intelligent endoscope system with pneumoperitoneum control and central control module

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810998268.7A CN109124702B (en) 2018-08-29 2018-08-29 Intelligent endoscope system with pneumoperitoneum control and central control module

Publications (2)

Publication Number Publication Date
CN109124702A CN109124702A (en) 2019-01-04
CN109124702B true CN109124702B (en) 2020-09-01

Family

ID=64829322

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810998268.7A Active CN109124702B (en) 2018-08-29 2018-08-29 Intelligent endoscope system with pneumoperitoneum control and central control module

Country Status (1)

Country Link
CN (1) CN109124702B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109363615A (en) * 2018-08-31 2019-02-22 合肥德铭电子有限公司 Inhibit the mobile integrated cavity mirror system of function with pneumoperitoneum control and electromagnetism
CN109363736A (en) * 2018-09-10 2019-02-22 安徽省胸科医院 Configure the integrated cavity mirror system of central control module
CN113171142B (en) * 2021-04-13 2023-09-26 上海澳华内镜股份有限公司 Pneumoperitoneum machine and control method

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN203352463U (en) * 2013-07-17 2013-12-18 浙江沃尔德电力电子有限公司 multi-channel isolated output and anti-interference intelligent power capacitor
CN204012697U (en) * 2014-09-01 2014-12-10 成都乐创自动化技术股份有限公司 A kind of safeguard structure that is applied to laser machine internal control part power supply
CN105431093A (en) * 2013-08-06 2016-03-23 奥林巴斯株式会社 Pneumoperitoneum apparatus
CN107730458A (en) * 2017-09-05 2018-02-23 北京飞搜科技有限公司 A kind of fuzzy facial reconstruction method and system based on production confrontation network
CN108009473A (en) * 2017-10-31 2018-05-08 深圳大学 Based on goal behavior attribute video structural processing method, system and storage device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11087229B2 (en) * 2017-02-03 2021-08-10 Adxcel Inc. Accelerated machine learning optimization strategy to determine high performance cluster with minimum resources

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN203352463U (en) * 2013-07-17 2013-12-18 浙江沃尔德电力电子有限公司 multi-channel isolated output and anti-interference intelligent power capacitor
CN105431093A (en) * 2013-08-06 2016-03-23 奥林巴斯株式会社 Pneumoperitoneum apparatus
CN204012697U (en) * 2014-09-01 2014-12-10 成都乐创自动化技术股份有限公司 A kind of safeguard structure that is applied to laser machine internal control part power supply
CN107730458A (en) * 2017-09-05 2018-02-23 北京飞搜科技有限公司 A kind of fuzzy facial reconstruction method and system based on production confrontation network
CN108009473A (en) * 2017-10-31 2018-05-08 深圳大学 Based on goal behavior attribute video structural processing method, system and storage device

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Localisation of Colorectal Polyps by Convolutional Neural Network Features Learnt from White Light and Narrow Band Endoscopic Images of Multiple Databases;Yali Zheng等;《2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)》;20180721;第4142-4145页 *
新型池化层sort_pool2d实现更快更好的收敛：表现优于最大池化层（附代码实现）;黄小天 等;《http://www.sohu.com/a/195961723_465975》;20171002;全文 *
面向公交车场景的视频关键帧提取算法研究;陈响;《中国优秀硕士学位论文全文数据库信息科技辑》;20180815;第18-49页 *

Also Published As

Publication number Publication date
CN109124702A (en) 2019-01-04

Similar Documents

Publication Publication Date Title
CN109124782B (en) Intelligent integrated endoscope system
CN109124702B (en) Intelligent endoscope system with pneumoperitoneum control and central control module
CN109171605B (en) Intelligent edge computing system with target positioning and endoscope video enhancement processing functions
CN101330862B (en) Automatic endoscope recognition and selection of image processing and display settings
CN106871365A (en) The progress control method of air-conditioner, device and air-conditioning system
CN109223178B (en) Cavity mirror intelligent edge computing system with target positioning function
CN104483851A (en) Context awareness control device, system and method
CN104398231B (en) System and method for adjusting stereoscopic endoscope light source brightness
CN109363613B (en) Mobile intelligent integrated endoscope system with electromagnetic suppression function
CN110890078A (en) Method for automatically adjusting brightness of display screen based on pupil size
WO2018163023A1 (en) Cooker hood and video surveillance method based on cooker hood
JP2007078283A (en) Air conditioner, and air conditioning method
WO2023040577A1 (en) Method and system for intelligent management and control of use of screen terminal by child
CN108926319B (en) Mobile intelligent integrated endoscope system with high-definition video processing module
CN107422703B (en) Equipment control method and device
CN109350239A (en) Intelligent integral robot cavity mirror system with target positioning function
CN109171606B (en) Intelligent integrated robot endoscope system
CN109363615A (en) Inhibit the mobile integrated cavity mirror system of function with pneumoperitoneum control and electromagnetism
CN109243586A (en) Configure the hysteroscope intelligence edge calculations system of central control module
CN108984140B (en) Display control method and system
CN109151275B (en) Intelligent edge computing system with real-time endoscope visual image enhancement processing function
CN109288487B (en) Intelligent cavity mirror system with high-definition video processing and electromagnetic suppression functions
CN110430369B (en) Cavity mirror light source cooperative control method and device based on central processing unit
CN206341308U (en) A kind of vision composite control apparatus
CN114430463B (en) Optimization method and system for self-adaptive medical-grade image effect

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant