WO2017080425A1 - Health assessing method and expert system based on human body temperature modeling - Google Patents

Health assessing method and expert system based on human body temperature modeling Download PDF

Info

Publication number
WO2017080425A1
WO2017080425A1 PCT/CN2016/104943 CN2016104943W WO2017080425A1 WO 2017080425 A1 WO2017080425 A1 WO 2017080425A1 CN 2016104943 W CN2016104943 W CN 2016104943W WO 2017080425 A1 WO2017080425 A1 WO 2017080425A1
Authority
WO
WIPO (PCT)
Prior art keywords
temperature
vector
preset
human body
sample
Prior art date
Application number
PCT/CN2016/104943
Other languages
French (fr)
Inventor
Hong KANG
Original Assignee
Shanghai Well Diagnostics Technology Co., Ltd.
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 Shanghai Well Diagnostics Technology Co., Ltd. filed Critical Shanghai Well Diagnostics Technology Co., Ltd.
Publication of WO2017080425A1 publication Critical patent/WO2017080425A1/en

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/01—Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • A61B5/7282—Event detection, e.g. detecting unique waveforms indicative of a medical condition
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
    • G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders

Definitions

  • the present invention relates to the field of human body temperature measurement, and particularly to a health assessing method and expert system based on human body temperature modeling.
  • Body temperature generally refers to a temperature inside a human body, which is an average temperature in depth of an organism of the human body. Biologically, the body temperature refers to a temperature of extracellular fluid of the human body, which usually is 37 °C. Axillary temperatures of normal people are 36 ⁇ 37°C. Methods for measuring the body temperature include oral thermometry, axillary thermometry, and anal thermometry. Oral temperature is 0.2 ⁇ 0.4 °C higher than the axillary temperature, while anal temperature is 0.3 ⁇ 0.5°C higher than the oral temperature.
  • the body temperature of the human body is relatively constant. Usually, the body temperature of a normal human body slightly fluctuates in a range of not more than 1°C within 24 hours. Under physiological conditions, the body temperature of the normal human body is slightly lower in the morning than in the afternoon. The body temperature of the normal human body is slightly higher after he/she takes exercises or eats something, before a menstrual period or during a gestation period of a female; while the body temperature of the aged is slightly lower.
  • a situation that the body temperature is higher than a normal temperature is referred to as a fever, wherein 37.3 ⁇ 38°C is a low-grade fever, 38.1 ⁇ 39°C is a middle-grade fever, 39.1 ⁇ 41°Cis a high-grade fever, and over 41°C is an over-high fever.
  • a relatively constant body temperature is one of important conditions to maintain normal living activities of the human body.
  • a body temperature higher than 41°C or lower than 25°C will seriously affect functional activities of various systems, particularly a neurological system, or even endanger life of the human body. Heat generation and dissipation of the human body is adjusted by a nerve center. Many discomforts may cause dysfunction to normal body temperature regulation of the human body, thereby causing changes of the body temperature.
  • a relatively higher body temperature may make enzymes inside the human body more active, enhance the function of white blood cells, improve immunity, facilitate blood circulation, increase basal metabolic rate, and thus improve physical fitness.
  • the present invention provides a health assessing method and expert system based on human body temperature modeling and an electronic device, so as to determine whether an abnormality occurs in a human body by means of detection of temperature abnormality.
  • a health assessing method based on human body temperature modeling comprising:
  • determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
  • the preset body temperature reference model is built by acquiring temperature values of preset positions of a plurality of normal human bodies.
  • the acquiring temperature values of preset positions of a plurality of normal human bodies comprises:
  • a number of the plurality of normal human bodies is more than 5000.
  • the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
  • the temperature sample values of the preset positions are basal body temperatures of the human body
  • the preset body temperature reference model is a preset human body basal body temperature reference model.
  • the difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector comprises:
  • the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
  • the method further comprises:
  • the method further comprises:
  • abnormal body surface temperature reference models associated with all vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  • the method further comprises:
  • abnormal body surface temperature reference models associated with all minimum value vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  • a health assessing expert system based on human body temperature modeling comprising:
  • a sampling module configured to acquire temperature sample values of at least two preset positions of a human body to form a sample temperature vector
  • a matching module configured to difference match the sample temperature vector with a preset body temperature reference model to obtain a difference vector
  • a determining module configured to determine whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
  • sampling module is further configured to:
  • the acquiring temperature values of preset positions of a plurality of normal human bodies comprises:
  • a number of the plurality of normal human bodies is more than 5000.
  • the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
  • the temperature sample values of the preset positions are basal body temperature sample values of the human body
  • the preset body temperature reference model is a preset human body basal body temperature reference model.
  • the difference matching the sample temperature vector and a preset body temperature reference model to obtain a difference vector comprises:
  • the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
  • system further comprises:
  • a calculating module configured to calculate Euclidean distances between the sample temperature vector and a plurality of abnormal body surface temperature reference models respectively to form a first Euclidean distance vector
  • a selecting module configure to acquire an abnormal body surface temperature reference model corresponding to a vector element smaller than a preset Euclidean distance value in the first Euclidean distance vector associated with the human body.
  • the selecting module is further configured to:
  • abnormal body surface temperature reference models associated with all vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  • the selecting model is further configured to:
  • abnormal body surface temperature reference models associated with all minimum value vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  • an electronic device comprising a processor, a memory, a communication interface, and a bus, wherein:
  • the processor, the memory, and the communication interface are connected and communicate with each other via the bus;
  • the memory stores executable program codes
  • the processor executes a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to:
  • the temperature sample values of the at least two preset positions of the human body are acquired, the temperature sample values are difference matched with the preset body temperature reference model so that the difference vector is obtained, and it is determined whether the body surface temperature of the human body is abnormal based on the difference value and the preset temperature drift threshold. In this way, fast determination is enabled in a simple and effective manner, and the determination flow is simplified.
  • Fig. 1 is a flow diagram of a health assessing method based on human body temperature modeling according to an embodiment of the present invention
  • Fig. 2 is a schematic diagram of determining discomfort types based on an Euclidean space in another embodiment of the present invention.
  • Fig. 3 is a structure diagram of an apparatus for acquiring temperature signals in another embodiment of the present invention.
  • Fig. 4 is a structural diagram of a health assessing expert system based on human body temperature modeling according to another embodiment of the present invention.
  • Fig. 1 illustrates a health assessing method based on human body temperature modeling, comprising the following steps:
  • temperature sample values of at least two preset positions of a human body are acquired to form a sample temperature vector.
  • the sample temperature vector is difference matched with a preset body temperature reference model to obtain a difference vector.
  • Overall temperature data of the human body may be obtained from the above sample sequence, and some physiological characteristics of a user may be obtained by processing the sample data.
  • the axillar temperature and the oral temperature that are acquired from the user constitute a sample sequence ⁇ 38°C, 38.2°C ⁇ ; in the preset body temperature reference model, the axillar temperature is [36.8°C-36.9°C] , and the oral temperature is [36.9°C -37.1°C] ; a difference vector ⁇ 1.1, 1.1 ⁇ is obtained by comparing the sample sequence constituted by the axillar temperature and the oral temperature that are acquired from the user with maximal values in the preset body temperature reference model.
  • corresponding thresholds e.g., an axillar temperature threshold of 1°C, and an oral cavity temperature threshold of 0.9°C are looked up, and then it may be determined that the body surface temperature of the user is abnormal.
  • dot product operation is performed on the difference vector by means of weighting.
  • a weight vector ⁇ 0.4, 0.6 ⁇ whose vector elements add up to 1, is set; after performing the dot product operation on the difference vector ⁇ 1.1, 1.1 ⁇ and the weight vector, a weighted difference vector ⁇ 0.44, 0.66 ⁇ is obtained; the vector elements obtained by the dot product operation are added up to obtain a difference value for one-dimensional data comparison.
  • a concentration measure also referred to as a central tendency measure, is a numerical value representing a general level of a group of data. Frequently used concentration measures include mean, median, and mode.
  • the mean is an arithmetic mean value of all measurement data; the median is a variable attribute value which divides the measurement data into two parts in order of magnitude, i.e., a value at the middle position in the arrangement order; the mode is a value having a highest appearance frequency in the measurement data.
  • axillar temperatures of 9 different human bodies are acquired: 36.6°C, 36.6°C, 36.7°C, 36.9°C, 37.0°C, 37.1°C, 37.2°C, 37.4°C, and 38.5°C.
  • the mean is 37.0°C
  • the median is 37.0°C
  • the mode is 36.6°C.
  • the mean, the median or the mode may be taken as the value of the preset body temperature reference model.
  • temperatures of different parts for measurement may be acquired within a continuous period of time.
  • a first part for measurement is the axillar of the human body
  • a second part for measurement is the forehead of the human body
  • a sample sequence 1 of the axillar of the human body is ⁇ X a1 , X a2 , X a3 , ... , X an ⁇
  • a sample sequence 2 of the forehead of the human body is ⁇ X b1 , X b2 , X b3 , ... , X bn ⁇ .
  • the mean of the temperature sequence of the second part for measurement is obtained by:
  • X bv (X b1 + X b2 + X b3 +, ... , + X bn ) /n
  • the preset body temperature reference model is built by taking the means as the sample values. In order to improve the universality of the preset body temperature reference model, a large number of samples are usually needed, e.g., a sampled population of more than 5000.
  • a temperature measurement sample set X has a mean of m and a standard deviation of s, then standardized variables of the temperature measurement sample set X is (X-m) /s.
  • the standardized value (the value before standardization –the mean of the components) /the standard deviation of the components.
  • two temperature sequences may be compared.
  • two measurement sequences may be matched by calculating a correlation coefficient of the two measurement sequences, wherein the correlation coefficient is expressed below:
  • E is the mathematic expectation or mean
  • D is the variance
  • the quotient of the covariance and the standard deviation of the two variables X and Y is taken as the correlation coefficient of the random variables X and Y) .
  • the correlation coefficient is a correlation degree metric between the random variables X and Y, and the range of the correlation coefficient is [-1, 1] .
  • the value of the correlation coefficient is 1 (positive linear correlation) or -1 (negative linear correlation) .
  • the meaning of the correlation coefficient of the two variables X and Y may be interpreted as the following:
  • the two variables X and Y have no relation with each other.
  • the two variables are positively correlated and the correlation coefficient of the two variables is between 0.00 and 1.00.
  • the two variables are negatively correlated and the correlation coefficient of the two variables is between -1.00 and 0.00.
  • the correlation degree of the two variables is determined based on the range of the correlation coefficient:
  • Fig. 2 which illustrates positions of respective temperature sample models in an Euclidean space, wherein four blocks in the Euclidean space represent positions of body temperature models corresponding to Type I-IV discomforts, respectively; the circle in the center of the Euclidean space represents a normal body temperature model; dotted-line external to the circle represents a first threshold, i.e., the temperature drift threshold. If a distance between a temperature model of a study object and the normal body temperature model exceeds the temperature drift threshold, the study object may be screened to be of some suspected discomfort.
  • some discomfort is manifested as such that the temperature of the head is higher than a mean of e.g., 38.5°C, and the temperature of the foot is lower than a mean of e.g., 36°C, i.e., hot head and cold feet; while some discomfort is manifested as such that the temperatures of the whole body are unanimously over high, i.e. between 38.6 –39.1, without any cold part throughout the body.
  • Quantized conclusive data may distinguish the above two types of discomfort and their manifestations.
  • the Euclidean distance between the human body temperature model of the study object and a certain discomfort model must be smaller than a second threshold, i.e. a preset Euclidean distance value, because accuracy shall be ensured in health assessment. Since a discomfort model library is not necessarily comprehensive, similarities between the temperature model of the study object and multiple discomfort models may be determined based on the second threshold.
  • a five-pointed star represents the temperature model of the study object whose body temperature is abnormal, i.e. larger than the first threshold; at this point, the Euclidean distances between the temperature model and the type I-IV abnormal body surface temperature reference models in Fig. 2 may be calculated to form a first Euclidean distance vector, for example, ⁇ 0.8, 0.2, 0.3, 0.9 ⁇ ; if the second threshold is set to 0.5, elements less than 0.5 in the first Euclidean distance vector are all valid elements.
  • valid elements 0.2 and 0.3 in the first Euclidean distance vector are all selected as valid Euclidean distance values; by analysis, it may be determined that 0.2 and 0.3 represent type II abnormal body surface temperature reference model and type III abnormal body surface temperature reference model, respectively; in this case, it may be assumed that the user has discomforts respectively associated with type II abnormal body surface temperature reference model and type III abnormal body surface temperature reference model.
  • an abnormal body surface temperature reference model associated with a valid element having a minimum value in the first Euclidean distance vector may be selected as the discomfort model of the user, i.e., the discomfort represented by type II abnormal body surface temperature reference model corresponding to 0.2 is selected.
  • the acquiring a temperature value sequence of each temperature measurement point within the preset period of time comprising: disposing a temperature sensor for measuring ear temperatures within an ear of the human body, and taking the ear temperatures acquired with a preset sampling interval within the preset period of time as the temperature value sequence.
  • an apparatus 30 for acquiring temperature signals in an embodiment of the present invention comprises a stickup temperature sensor 303, a data transmission link 302, and a data communication interface 301.
  • the number of the stickup temperature sensors 303 and their attachment manners may be freely set according to requirements of the user.
  • the stickup temperature sensors 303 When the user places the stickup temperature sensors 303 at parts such as the temple or the forehead, the stickup temperature sensors 303 will acquire temperature data of the user with a predetermined interval.
  • the data transmission link 302 is configured for transmitting the temperature data acquired by the stickup temperature sensors 303 to the data communication interface by means of wired or wireless communication.
  • the data communication interface 301 is configured for transmitting the acquired temperature data to a corresponding client (not shown in the figure) .
  • the data communication interface 301 is a USB communication interface, which transmits the temperature data in the format of USB data to a mobile phone of the user for viewing corresponding contents.
  • the user may acquire data of his/her basal brain temperature by wearing the apparatus 30.
  • a sample sequence constituted by these body temperature data may be processed online or offline, so that it may be determined whether the basal brain temperature of the user is abnormal, and if so, an abnormal mode and the manifestations of its corresponding potential discomfort information may also be determined.
  • Fig. 4 is a structural diagram of a temperature measuring apparatus according to an embodiment of the present invention.
  • the apparatus comprises a sampling module 401, a matching module 402, and a determining module 403, wherein:
  • the acquiring module 401 is configured for acquiring the temperature sample values of the at least two preset positions of the human body to form the sample temperature vector.
  • the matching module 402 is configured for difference matching the sample temperature vector with the preset body temperature reference model to obtain the difference vector.
  • Overall temperature data of the human body may be obtained from the above sample sequence, and some physiological characteristics of the user may be obtained by processing the sample data.
  • the axillar temperature and the oral temperature that are acquired from the user constitute a sample sequence ⁇ 38°C, 38.2°C ⁇ ; in the preset body temperature reference model, the axillar temperature is [36.8-36.9] °C, and the oral temperature is [36.9-37.1] °C; a difference vector ⁇ 1.1, 1.1 ⁇ is obtained by comparing the sample sequence constituted by the axillar temperature and the oral temperature that are acquired from the user with the maximal values in the preset body temperature reference model.
  • the determining module 403 is configured for determining whether the body surface temperature of the human body is abnormal based on the comparison between the difference vector and the preset temperature drift threshold.
  • corresponding thresholds e.g., the axillar temperature threshold of 1°C, and the oral cavity threshold of 0.9°C, are looked up, and then it can be determined that the body surface temperature of the user is abnormal.
  • dot product operation is performed on the difference vector by means of weighting.
  • a weight vector ⁇ 0.4, 0.6 ⁇ whose vector elements add up to 1, is set; after performing the dot product operation on the difference vector ⁇ 1.1, 1.1 ⁇ and the weight vector, the weighted difference vector ⁇ 0.44, 0.66 ⁇ is obtained; the vector elements obtained by the dot product operation are added up to obtain the difference value for one-dimensional data comparison.
  • a concentration measure also referred to as a central tendency measure, is a numerical value representing a general level of a group of data. Frequently used concentration measures include mean, median, and mode.
  • the mean is an arithmetic mean value of all measurement data; the median is a variable attribute value which divides the measurement data into two parts in order of magnitude, i.e., a value at the middle position in the arrangement order; the mode is a value having a highest appearance frequency in the measurement data.
  • axillar body temperatures of 9 different human bodies are acquired: 36.6°C, 36.6°C, 36.7°C, 36.9°C, 37.0°C, 37.1°C, 37.2°C, 37.4°C, and 38.5°C.
  • the mean is 37.0°C
  • the median is 37.0°C
  • the mode is 36.6°C.
  • the mean, the median or the mode may be taken as the value of the preset body temperature reference model.
  • temperatures of different parts for measurement may be acquired within a continuous period of time.
  • a first part for measurement is the axillar of the human body
  • a second part for measurement is the forehead of the human body
  • the sampling sequence 1 of the axillar of the human body is ⁇ X a1 , X a2 , X a3 , ... , X an ⁇
  • the sampling sequence 2 of the forehead of the human body is ⁇ X b1 , X b2 , X b3 , ... , X bn ⁇ .
  • the mean of the temperature sequence of the second part for measurement is obtained by:
  • X bv (X b1 + X b2 + X b3 +, ... , + X bn ) /n
  • the preset temperature reference model is built by taking the means as the sample values. In order to improve the universality of the preset body temperature reference model, a large number of samples are usually needed, e.g., a sampled population of more than 5000.
  • the temperature measurement sample set X has a mean of m and a standard deviation of s, then the standardized variable of the temperature measurement sample set X is (X-m) /s.
  • the standardized value (the value before standardization –the mean of the components) /the standard deviation of the components.
  • two temperature sequences may be compared.
  • two measurement sequences may be matched by calculating a correlation coefficient of the two measurement sequences, wherein the correlation coefficient is expressed below:
  • E is the mathematic expectation or mean
  • D is the variance
  • the quotient of the covariance and the standard deviation of the two variables X and Y is taken as the correlation coefficient of the random variables X and Y) .
  • the correlation coefficient is a correlation degree metric between the random variables X and Y, and the range of the correlation coefficient is [-1, 1] .
  • the value of the correlation coefficient is 1 (positive linear correlation) or -1 (negative linear correlation) .
  • the meaning of the correlation coefficient of the two variables X and Y may be interpreted as the following:
  • the two variables X and Y have no relation with each other.
  • the two variables are positively correlated and the correlation coefficient of the two variables is between 0.00 and 1.00.
  • the two variables are negatively correlated and the correlation coefficient of the two variables is between -1.00 and 0.00.
  • the correlation degree of the two variables is determined based on the range of the correlation coefficient:
  • an embodiment of the present invention also provides an electronic device comprising a processor, a memory, a communication interface, and a bus.
  • the processor, the memory, and the communication interface are connected and communicate with each other via the bus.
  • the memory stores executable program codes.
  • the processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory so as to:
  • the electronic device may be implemented in a plurality of forms including, but not limited to:
  • a mobile communication device having mobile communication functions and mainly aiming to provide voice and data communications.
  • This kind of terminals includes a smart phone (e.g., iPhone) , a multimedia mobile phone, a functional mobile phone, and a low-profile mobile phone, etc.
  • An ultra mobile personal computer device which is a kind of personal computer and has computation and processing functions as well as mobile Internetwork accessing functions.
  • This kind of terminals includes a PDA, a MID, and an UMPC device, etc. (e.g., iPad) .
  • a portable entertainment device capable of displaying and playing multimedia contents.
  • This kind of device includes an audio/video player (e.g., iPod) , a palm game player, an electronic book, a smart toy, and a portable vehicle navigation device.
  • a server which provides computation services and is constituted by a processor, a hard disk, an internal memory, a system bus and so on.
  • the server is similar to a general purpose computer in architecture. However, it is necessary to provide highly reliable services, so its requirements on processing capability, stability, reliability, security, scalability, manageability and so on are relatively high.
  • the program may be stored in a computer readable storage medium.
  • the program when being executed, may include the processing flow of respective method embodiments above.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM) , etc.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Public Health (AREA)
  • Medical Informatics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Pathology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Veterinary Medicine (AREA)
  • Physics & Mathematics (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Physiology (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

A health assessing method based on human body temperature modeling, an expert system, and an electronic device. The health assessing expert system based on human body temperature modeling comprises: an acquiring module (401), a matching module (402), and a determining module (403). The acquiring module (401) is configured for acquiring the temperature sample values of the at least two preset positions of the human body to form the sample temperature vector, the matching module (402) is configured for difference matching the sample temperature vector with the preset body temperature reference model to obtain the difference vector, the determining module (403) is configured for determining whether the body surface temperature of the human body is abnormal based on the comparison between the difference vector and the preset temperature drift threshold.

Description

HEALTH ASSESSING METHOD AND EXPERT SYSTEM BASED ON HUMAN BODY TEMPERATURE MODELING FIELD OF THE INVENTION
The present invention relates to the field of human body temperature measurement, and particularly to a health assessing method and expert system based on human body temperature modeling.
BACKGROUND OF THE INVENTION
Body temperature generally refers to a temperature inside a human body, which is an average temperature in depth of an organism of the human body. Biologically, the body temperature refers to a temperature of extracellular fluid of the human body, which usually is 37 ℃. Axillary temperatures of normal people are 36～37℃. Methods for measuring the body temperature include oral thermometry, axillary thermometry, and anal thermometry. Oral temperature is 0.2 ～ 0.4 ℃ higher than the axillary temperature, while anal temperature is 0.3 ～ 0.5℃ higher than the oral temperature.
The body temperature of the human body is relatively constant. Usually, the body temperature of a normal human body slightly fluctuates in a range of not more than 1℃ within 24 hours. Under physiological conditions, the body temperature of the normal human body is slightly lower in the morning than in the afternoon. The body temperature of the normal human body is slightly higher after he/she takes exercises or eats something, before a menstrual period or during a gestation period of a female； while the body temperature of the aged is slightly lower. A situation that the body temperature is higher than a normal temperature is referred to as a fever, wherein 37.3～38℃ is a low-grade fever, 38.1～39℃ is a middle-grade fever, 39.1～41℃is a high-grade fever, and over 41℃ is an over-high fever. A relatively constant body temperature is one of important conditions to maintain normal living activities of the human body. A body temperature higher than 41℃ or lower than 25℃ will seriously affect functional activities of various systems, particularly a neurological system, or even endanger life of the human body. Heat generation and dissipation of the human body is adjusted by a nerve center. Many discomforts may cause dysfunction to normal body temperature regulation of the human body, thereby causing changes of the body temperature.
If the human body maintains a higher temperature within a normal body temperature range, the following effects will be induced to the human body: a relatively higher body temperature may make enzymes inside the human body more active, enhance the function of white blood cells, improve immunity, facilitate blood circulation, increase basal metabolic rate, and thus improve physical fitness.
If the human body maintains a lower temperature within the normal body temperature range in long term, some negative impacts will often be induced to the human body and thus discomforts will easily occur in the human body, e.g., affecting autonomic nerve function and hormonal balance, declining immunity, decreasing cell metabolic rate, and thus causing shoulder stiffness, headache, dizziness, cold hands and feet, susceptibility to fatigue, as well as various other uncomfortable symptoms such as constipation, gaseous distention, hypourocrinia, worsened skin and so on. Therefore, observing body temperature changes is important for health assessment and heath care.
SUMMARY OF THE INVENTION
The present invention provides a health assessing method and expert system based on human body temperature modeling and an electronic device, so as to determine whether an abnormality occurs in a human body by means of detection of temperature abnormality.
In one aspect, according to an embodiment of the present invention, there is provided a health assessing method based on human body temperature modeling, the method comprising:
acquiring temperature sample values of at least two preset positions of a human body to form a sample temperature vector；
difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector； and
determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
Optionally, the preset body temperature reference model is built by acquiring temperature values of preset positions of a plurality of normal human bodies.
Optionally, the acquiring temperature values of preset positions of a plurality of normal human bodies comprises:
continuously acquiring the temperature values of the preset positions of the plurality of normal human bodies within a preset period of time.
Optionally, a number of the plurality of normal human bodies is more than 5000.
Optionally, the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
if the difference vector is smaller than the preset temperature drift threshold, determining that the body surface temperature of the human body is normal.
Optionally, the temperature sample values of the preset positions are basal body temperatures of the human body；
The preset body temperature reference model is a preset human body basal body temperature reference model.
Optionally, the difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector comprises:
setting a weight vector with a same dimension as the sample temperature vector；
performing dot product operation of a result of subtracting the sample temperature vector by the preset body temperature reference model and the weight vector to obtain the difference vector.
Optionally, the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
if the difference vector is greater than the preset temperature drift threshold, determining that the body surface temperature of the human body is abnormal.
Optionally, the method further comprises:
calculating Euclidean distances between the sample temperature vector and a plurality of abnormal body surface temperature reference models respectively, to form a first Euclidean distance vector；
acquiring an abnormal body surface temperature reference model corresponding to a vector  elements smaller than a preset Euclidean distance value in the first Euclidean distance vector associated with the human body.
Optionally, the method further comprises:
determining abnormal body surface temperature reference models associated with all vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
Optionally, the method further comprises:
determining abnormal body surface temperature reference models associated with all minimum value vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
In the second aspect, according to another embodiment of the present invention, there is provided a health assessing expert system based on human body temperature modeling, the system comprising:
a sampling module configured to acquire temperature sample values of at least two preset positions of a human body to form a sample temperature vector；
a matching module configured to difference match the sample temperature vector with a preset body temperature reference model to obtain a difference vector； and
a determining module configured to determine whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
Optionally, the sampling module is further configured to:
build the preset body temperature reference model by acquiring temperature values of preset positions of a plurality of normal human bodies.
Optionally, the acquiring temperature values of preset positions of a plurality of normal human bodies comprises:
continuously acquiring the temperature values of the preset positions of the plurality of normal human bodies within a preset period of time.
Optionally, a number of the plurality of normal human bodies is more than 5000.
Optionally, the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
if the difference vector is smaller than the preset temperature drift threshold, determining that the body surface temperature of the human body is normal.
Optionally, the temperature sample values of the preset positions are basal body temperature sample values of the human body；
The preset body temperature reference model is a preset human body basal body temperature reference model.
Optionally, the difference matching the sample temperature vector and a preset body temperature reference model to obtain a difference vector comprises:
setting a weight vector with a same dimension as the sample temperature vector；
performing dot product operation of a result of subtracting the sample temperature vector by the preset body temperature reference model and the weight vector to obtain the difference vector.
Optionally, the determining whether a body surface temperature of the human body is  abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
if the difference vector is greater than the preset temperature drift threshold, determining that the body surface temperature of the human body is abnormal.
Optionally, the system further comprises:
a calculating module configured to calculate Euclidean distances between the sample temperature vector and a plurality of abnormal body surface temperature reference models respectively to form a first Euclidean distance vector；
a selecting module configure to acquire an abnormal body surface temperature reference model corresponding to a vector element smaller than a preset Euclidean distance value in the first Euclidean distance vector associated with the human body.
Optionally, the selecting module is further configured to:
determine abnormal body surface temperature reference models associated with all vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
Optionally, the selecting model is further configured to:
determine abnormal body surface temperature reference models associated with all minimum value vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
In the third aspect, according to still another embodiment of the present invention, there is provided an electronic device comprising a processor, a memory, a communication interface, and a bus, wherein:
the processor, the memory, and the communication interface are connected and communicate with each other via the bus；
the memory stores executable program codes；
the processor executes a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to:
acquire temperature sample values of at least two preset positions of a human body to form a sample temperature vector；
difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector； and
determine whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
In the health assessing method and expert system based on human body temperature modeling and the electronic device in the above embodiments, the temperature sample values of the at least two preset positions of the human body are acquired, the temperature sample values are difference matched with the preset body temperature reference model so that the difference vector is obtained, and it is determined whether the body surface temperature of the human body is abnormal based on the difference value and the preset temperature drift threshold. In this way, fast determination is enabled in a simple and effective manner, and the determination flow is simplified.
BRIEF DESCRIPTION OF THE PREFERRED EMBODIMENTS
In order to illustrate the technical solutions in the embodiments of the present invention or  the prior art, the accompanying drawings to be used in describing the embodiments or the prior art will be described briefly. Apparently, the accompanying drawings herein are only some embodiments of the present invention. For those skilled in the art, other accompanying drawings may also be obtained based on these accompanying drawings without any creative work.
Fig. 1 is a flow diagram of a health assessing method based on human body temperature modeling according to an embodiment of the present invention；
Fig. 2 is a schematic diagram of determining discomfort types based on an Euclidean space in another embodiment of the present invention；
Fig. 3 is a structure diagram of an apparatus for acquiring temperature signals in another embodiment of the present invention；
Fig. 4 is a structural diagram of a health assessing expert system based on human body temperature modeling according to another embodiment of the present invention.
In the accompanying drawings, a same or similar reference numeral represents a same or similar step feature or means (module) .
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and comprehensively in combination with the accompanying drawings. It is apparent that the embodiments as described are only part of embodiments of the present invention. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without any creative work all fall within the scopes as claimed by the present invention.
Fig. 1 illustrates a health assessing method based on human body temperature modeling, comprising the following steps:
S101, temperature sample values of at least two preset positions of a human body are acquired to form a sample temperature vector.
In order to measure a plurality of temperature values of the human body accurately, it is necessary to set as many measurement points as possible.
As temperatures of different positions of the human body are slightly different, it is usual to measure the temperatures of typical areas including inside the oral, axillary, anal, foot, forehead, and so on of the human body to form the following sample sequence:
Sample sequence {a1, a2, a3, … , an} .
S102, the sample temperature vector is difference matched with a preset body temperature reference model to obtain a difference vector.
Overall temperature data of the human body may be obtained from the above sample sequence, and some physiological characteristics of a user may be obtained by processing the sample data. For example, the axillar temperature and the oral temperature that are acquired from the user constitute a sample sequence {38℃, 38.2℃} ； in the preset body temperature reference model, the axillar temperature is [36.8℃-36.9℃] , and the oral temperature is [36.9℃ -37.1℃] ； a difference vector {1.1, 1.1} is obtained by comparing the sample sequence constituted by the axillar temperature and the oral temperature that are acquired from the user with maximal values in the preset body temperature reference model.
S103, it is determined whether a body surface temperature of the human body is abnormal  based on a comparison between the difference vector and a preset temperature drift threshold.
For the difference vector {1.1, 1.1} obtained in step S102, corresponding thresholds, e.g., an axillar temperature threshold of 1℃, and an oral cavity temperature threshold of 0.9℃ are looked up, and then it may be determined that the body surface temperature of the user is abnormal.
Because of the presence of errors in the temperatures acquired at different parts of the human body, dot product operation is performed on the difference vector by means of weighting. For example, a weight vector {0.4, 0.6} , whose vector elements add up to 1, is set； after performing the dot product operation on the difference vector {1.1, 1.1} and the weight vector, a weighted difference vector {0.44, 0.66} is obtained； the vector elements obtained by the dot product operation are added up to obtain a difference value for one-dimensional data comparison.
In order to build the preset body temperature reference model, a central tendency analysis is required. A concentration measure, also referred to as a central tendency measure, is a numerical value representing a general level of a group of data. Frequently used concentration measures include mean, median, and mode. The mean is an arithmetic mean value of all measurement data； the median is a variable attribute value which divides the measurement data into two parts in order of magnitude, i.e., a value at the middle position in the arrangement order； the mode is a value having a highest appearance frequency in the measurement data.
For example, the following axillar temperatures of 9 different human bodies are acquired: 36.6℃, 36.6℃, 36.7℃, 36.9℃, 37.0℃, 37.1℃, 37.2℃, 37.4℃, and 38.5℃.
In this case, the mean is 37.0℃, the median is 37.0℃, and the mode is 36.6℃.
Dependent on requirements of different models, the mean, the median or the mode may be taken as the value of the preset body temperature reference model.
As temperatures of a same human body differ slightly in different periods of time, temperatures of different parts for measurement may be acquired within a continuous period of time. For example, a first part for measurement is the axillar of the human body, and a second part for measurement is the forehead of the human body； a sample sequence 1 of the axillar of the human body is {Xa1, Xa2, Xa3, … , Xan} , and a sample sequence 2 of the forehead of the human body is {Xb1, Xb2, Xb3, … , Xbn} .
For example, the mean of the temperature sequence of the first part for measurement may be obtained by Xas ＝ (Xa1 +Xa2+Xa3+, … , +Xan) /n.
The mean of the temperature sequence of the second part for measurement is obtained by:
Xbv ＝ (Xb1 + Xb2 + Xb3 +, … , + Xbn) /n
The preset body temperature reference model is built by taking the means as the sample values. In order to improve the universality of the preset body temperature reference model, a large number of samples are usually needed, e.g., a sampled population of more than 5000.
In view that distributions of components of the acquired data in respective dimensions are different, it is assumed that a temperature measurement sample set X has a mean of m and a standard deviation of s, then standardized variables of the temperature measurement sample set X is (X-m) /s.
Moreover, a mathematic expectation of the standardized variables is 0, and a variance of the standardized variables is 1. Therefore, the standardization procedure of the sample set X is expressed by the following equation:
Figure PCTCN2016104943-appb-000001
That is, the standardized value ＝ (the value before standardization –the mean of the components) /the standard deviation of the components.
By simple derivation, an equation of a standardized Euclidean distance between two n-dimension vectors a (x11, x12, … , x1n) and b (x21, x22, … , x2n) may be obtained:
Figure PCTCN2016104943-appb-000002
By comparing standardized Euclidean distances, two temperature sequences may be compared.
Besides, two measurement sequences may be matched by calculating a correlation coefficient of the two measurement sequences, wherein the correlation coefficient is expressed below:
Figure PCTCN2016104943-appb-000003
(where E is the mathematic expectation or mean, D is the variance, 
Figure PCTCN2016104943-appb-000004
is the standard deviation, E { [X-E (X) ] [Y-E (U) ] } is the covariance of random variables X and Y and is denoted as Cov (X, Y) , i.e., Cov (X, Y) ＝ E { [X-E (X) ] [Y-E (U) ] } , and the quotient of the covariance and the standard deviation of the two variables X and Y is taken as the correlation coefficient of the random variables X and Y) .
The correlation coefficient is a correlation degree metric between the random variables X and Y, and the range of the correlation coefficient is [-1, 1] . The larger the absolute value of the correlation coefficient is, the higher the correlation degree between X and Y is. When X and Y are linearly correlated, the value of the correlation coefficient is 1 (positive linear correlation) or -1 (negative linear correlation) .
Specifically, the meaning of the correlation coefficient of the two variables X and Y may be interpreted as the following:
If the correlation coefficient is 0, the two variables X and Y have no relation with each other.
If the value of X increases (decreases) while the value of Y increases (decreases) , the two variables are positively correlated and the correlation coefficient of the two variables is between 0.00 and 1.00.
If the value of X increases (decreases) while the value of Y decreases (increases) , the two variables are negatively correlated and the correlation coefficient of the two variables is between -1.00 and 0.00.
Generally, the correlation degree of the two variables is determined based on the range of the correlation coefficient:
Correlation coefficient 0.8-1.0 extremely strong correlation
0.6–0.8          strong correlation
0.4-0.6          medium degree correlation
0.2-0.4          weak correlation
0.0-0.2          extremely weak correlation or no correlation
Referring to Fig. 2, which illustrates positions of respective temperature sample models in an Euclidean space, wherein four blocks in the Euclidean space represent positions of body temperature models corresponding to Type I-IV discomforts, respectively； the circle in the center of the Euclidean space represents a normal body temperature model； dotted-line external to the circle represents a first threshold, i.e., the temperature drift threshold. If a distance between a temperature model of a study object and the normal body temperature model exceeds the temperature drift threshold, the study object may be screened to be of some suspected discomfort. For example, some discomfort is manifested as such that the temperature of the head is higher than a mean of e.g., 38.5℃, and the temperature of the foot is lower than a mean of e.g., 36℃, i.e., hot head and cold feet； while some discomfort is manifested as such that the temperatures of the whole body are unanimously over high, i.e. between 38.6 –39.1, without any cold part throughout the body. Quantized conclusive data may distinguish the above two types of discomfort and their manifestations. Specifically, besides the threshold represented by the dotted line in Fig. 2, the Euclidean distance between the human body temperature model of the study object and a certain discomfort model must be smaller than a second threshold, i.e. a preset Euclidean distance value, because accuracy shall be ensured in health assessment. Since a discomfort model library is not necessarily comprehensive, similarities between the temperature model of the study object and multiple discomfort models may be determined based on the second threshold.
As shown in Fig. 2, a five-pointed star represents the temperature model of the study object whose body temperature is abnormal, i.e. larger than the first threshold； at this point, the Euclidean distances between the temperature model and the type I-IV abnormal body surface temperature reference models in Fig. 2 may be calculated to form a first Euclidean distance vector, for example, {0.8, 0.2, 0.3, 0.9} ； if the second threshold is set to 0.5, elements less than 0.5 in the first Euclidean distance vector are all valid elements.
In an embodiment, valid elements 0.2 and 0.3 in the first Euclidean distance vector are all selected as valid Euclidean distance values； by analysis, it may be determined that 0.2 and 0.3 represent type II abnormal body surface temperature reference model and type III abnormal body surface temperature reference model, respectively； in this case, it may be assumed that the user has discomforts respectively associated with type II abnormal body surface temperature reference model and type III abnormal body surface temperature reference model.
In another embodiment, an abnormal body surface temperature reference model associated with a valid element having a minimum value in the first Euclidean distance vector may be selected as the discomfort model of the user, i.e., the discomfort represented by type II abnormal body surface temperature reference model corresponding to 0.2 is selected.
In still another embodiment, the acquiring a temperature value sequence of each temperature measurement point within the preset period of time comprising: disposing a temperature sensor for measuring ear temperatures within an ear of the human body, and taking the ear temperatures acquired with a preset sampling interval within the preset period of time as the temperature value sequence.
As illustrated in Fig. 3, an apparatus 30 for acquiring temperature signals in an  embodiment of the present invention comprises a stickup temperature sensor 303, a data transmission link 302, and a data communication interface 301.
The number of the stickup temperature sensors 303 and their attachment manners may be freely set according to requirements of the user. When the user places the stickup temperature sensors 303 at parts such as the temple or the forehead, the stickup temperature sensors 303 will acquire temperature data of the user with a predetermined interval.
The data transmission link 302 is configured for transmitting the temperature data acquired by the stickup temperature sensors 303 to the data communication interface by means of wired or wireless communication.
The data communication interface 301 is configured for transmitting the acquired temperature data to a corresponding client (not shown in the figure) . As an example, the data communication interface 301 is a USB communication interface, which transmits the temperature data in the format of USB data to a mobile phone of the user for viewing corresponding contents.
The user may acquire data of his/her basal brain temperature by wearing the apparatus 30. A sample sequence constituted by these body temperature data may be processed online or offline, so that it may be determined whether the basal brain temperature of the user is abnormal, and if so, an abnormal mode and the manifestations of its corresponding potential discomfort information may also be determined.
Fig. 4 is a structural diagram of a temperature measuring apparatus according to an embodiment of the present invention. The apparatus comprises a sampling module 401, a matching module 402, and a determining module 403, wherein:
The acquiring module 401 is configured for acquiring the temperature sample values of the at least two preset positions of the human body to form the sample temperature vector.
In order to measure a plurality of temperature values of the human body accurately, it is necessary to set as many measurement points as possible.
As temperatures of different positions of the human body are slightly different, it is usual to measure the temperatures of typical areas including inside the oral, axillary, anal, foot, forehead and so on of the human body to form the following sample sequence:
Sampling sequence {a1, a2, a3, … , an} .
The matching module 402 is configured for difference matching the sample temperature vector with the preset body temperature reference model to obtain the difference vector.
Overall temperature data of the human body may be obtained from the above sample sequence, and some physiological characteristics of the user may be obtained by processing the sample data. For example, the axillar temperature and the oral temperature that are acquired from the user constitute a sample sequence {38℃, 38.2℃} ； in the preset body temperature reference model, the axillar temperature is [36.8-36.9] ℃, and the oral temperature is [36.9-37.1] ℃； a difference vector {1.1, 1.1} is obtained by comparing the sample sequence constituted by the axillar temperature and the oral temperature that are acquired from the user with the maximal values in the preset body temperature reference model.
The determining module 403 is configured for determining whether the body surface temperature of the human body is abnormal based on the comparison between the difference vector and the preset temperature drift threshold.
For the difference vector {1.1, 1.1} obtained by the matching module 402, corresponding thresholds, e.g., the axillar temperature threshold of 1℃, and the oral cavity threshold of 0.9℃, are looked up, and then it can be determined that the body surface temperature of the user is abnormal.
Because of the presence of errors in the temperatures acquired at different parts of the human body, dot product operation is performed on the difference vector by means of weighting. For example, a weight vector {0.4, 0.6} , whose vector elements add up to 1, is set； after performing the dot product operation on the difference vector {1.1, 1.1} and the weight vector, the weighted difference vector {0.44, 0.66} is obtained； the vector elements obtained by the dot product operation are added up to obtain the difference value for one-dimensional data comparison.
In order to build the preset body temperature reference model, a central tendency analysis is required. A concentration measure, also referred to as a central tendency measure, is a numerical value representing a general level of a group of data. Frequently used concentration measures include mean, median, and mode. The mean is an arithmetic mean value of all measurement data； the median is a variable attribute value which divides the measurement data into two parts in order of magnitude, i.e., a value at the middle position in the arrangement order； the mode is a value having a highest appearance frequency in the measurement data.
For example, the following axillar body temperatures of 9 different human bodies are acquired: 36.6℃, 36.6℃, 36.7℃, 36.9℃, 37.0℃, 37.1℃, 37.2℃, 37.4℃, and 38.5℃.
In this case, the mean is 37.0℃, the median is 37.0℃, and the mode is 36.6℃.
Dependent on requirements of different models, the mean, the median or the mode may be taken as the value of the preset body temperature reference model.
As temperatures of a same human body differ slightly in different periods of time, temperatures of different parts for measurement may be acquired within a continuous period of time. For example, a first part for measurement is the axillar of the human body, and a second part for measurement is the forehead of the human body； the sampling sequence 1 of the axillar of the human body is {Xa1, Xa2, Xa3, … , Xan} , and the sampling sequence 2 of the forehead of the human body is {Xb1, Xb2, Xb3, … , Xbn} .
For example, the mean of the temperature sequence of the first part for measurement may be obtained by Xav ＝ (Xa1 +Xa2+Xa3+, … , +Xan) /n.
The mean of the temperature sequence of the second part for measurement is obtained by:
Xbv ＝ (Xb1 + Xb2 + Xb3 +, … , + Xbn) /n
The preset temperature reference model is built by taking the means as the sample values. In order to improve the universality of the preset body temperature reference model, a large number of samples are usually needed, e.g., a sampled population of more than 5000.
In view that distributions of components of the acquired data in respective dimensions are different, it is assumed that the temperature measurement sample set X has a mean of m and a standard deviation of s, then the standardized variable of the temperature measurement sample set X is (X-m) /s.
Moreover, a mathematic expectation of the standardized variables is 0, and a variance of the standardized variables is 1. Therefore, the standardization procedure of the sample set X is expressed by the following equation:
Figure PCTCN2016104943-appb-000005
That is, the standardized value ＝ (the value before standardization –the mean of the components) /the standard deviation of the components.
By simple derivation, an equation of a standardized Euclidean distance between two n-dimension vectors a (x11, x12, … , x1n) and b (x21, x22, … , x2n) may be obtained:
Figure PCTCN2016104943-appb-000006
By comparing the standardized Euclidean distances, two temperature sequences may be compared.
Besides, two measurement sequences may be matched by calculating a correlation coefficient of the two measurement sequences, wherein the correlation coefficient is expressed below:
Figure PCTCN2016104943-appb-000007
(where E is the mathematic expectation or mean, D is the variance, 
Figure PCTCN2016104943-appb-000008
is the standard deviation, E { [X-E (X) ] [Y-E (U) ] } is the covariance of random variables X and Y and is denoted as Cov (X, Y) , i.e., Cov (X, Y) ＝ E { [X-E (X) ] [Y-E (U) ] } , and the quotient of the covariance and the standard deviation of the two variables X and Y is taken as the correlation coefficient of the random variables X and Y) .
The correlation coefficient is a correlation degree metric between the random variables X and Y, and the range of the correlation coefficient is [-1, 1] . The larger the absolute value of the correlation coefficient is, the higher the correlation degree between X and Y is. When X and Y are linearly correlated, the value of the correlation coefficient is 1 (positive linear correlation) or -1 (negative linear correlation) .
Specifically, the meaning of the correlation coefficient of the two variables X and Y may be interpreted as the following:
If the correlation coefficient is 0, the two variables X and Y have no relation with each other.
If the value of X increases (decreases) while the value of Y increases (decreases) , the two variables are positively correlated and the correlation coefficient of the two variables is between 0.00 and 1.00.
If the value of X increases (decreases) while the value of Y decreases (increases) , the two variables are negatively correlated and the correlation coefficient of the two variables is between -1.00 and 0.00.
Generally, the correlation degree of the two variables is determined based on the range of the correlation coefficient:
Correlation coefficient 0.8-1.0extremely strong correlation
0.6–0.8          strong correlation
0.4-0.6          medium degree correlation
0.2-0.4          weak correlation
0.0-0.2          extremely weal correlation or no correlation
Corresponding to the above method embodiment and apparatus embodiment, an embodiment of the present invention also provides an electronic device comprising a processor, a memory, a communication interface, and a bus.
The processor, the memory, and the communication interface are connected and communicate with each other via the bus.
The memory stores executable program codes.
The processor runs a program corresponding to the executable program codes by reading the executable program codes stored in the memory so as to:
acquire the temperature sample values of the at least two preset positions of the human body to form the sample temperature vector；
difference match the sample temperature vector with the preset body temperature reference model to obtain the difference vector； and
determine whether the body surface temperature of the human body is abnormal based on the comparison between the difference vector and the preset temperature drift threshold.
The electronic device may be implemented in a plurality of forms including, but not limited to:
(1) A mobile communication device having mobile communication functions and mainly aiming to provide voice and data communications. This kind of terminals includes a smart phone (e.g., iPhone) , a multimedia mobile phone, a functional mobile phone, and a low-profile mobile phone, etc.
(2) An ultra mobile personal computer device which is a kind of personal computer and has computation and processing functions as well as mobile Internetwork accessing functions. This kind of terminals includes a PDA, a MID, and an UMPC device, etc. (e.g., iPad) .
(3) A portable entertainment device capable of displaying and playing multimedia contents. This kind of device includes an audio/video player (e.g., iPod) , a palm game player, an electronic book, a smart toy, and a portable vehicle navigation device.
(4) A server which provides computation services and is constituted by a processor, a hard disk, an internal memory, a system bus and so on. The server is similar to a general purpose computer in architecture. However, it is necessary to provide highly reliable services, so its requirements on processing capability, stability, reliability, security, scalability, manageability and so on are relatively high.
(5) other electronic devices having data interaction functions.
A person skill in the art may understand that all or part of the processing flow in the above method embodiments may be implemented by hardware with computer program instructions. The program may be stored in a computer readable storage medium. The program, when being executed, may include the processing flow of respective method embodiments above. Particularly, the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM) , etc.
What have been described above are only specific embodiments of the present invention, and the protection scope of the present invention is not limited thereto. Changes or  substitutions within the technical disclosure of the present invention which may be easily envisaged by the person skilled in the art should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be limited by the protection scope (s) of the claims.

Claims (23)

  1. A health assessing method based on human body temperature modeling, comprising:
    acquiring temperature sample values of at least two preset positions of a human body to form a sample temperature vector；
    difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector； and
    determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
  2. The method according to claim 1, characterized in
    building the preset body temperature reference model by acquiring temperature values of preset positions of a plurality of normal human bodies.
  3. The method according to claim 2, characterized in that the acquiring temperature values of preset positions of a plurality of normal human bodies comprises:
    continuously acquiring the temperature values of the preset positions of the plurality of normal human bodies within a preset period of time.
  4. The method according to claim 2 or 3, characterized in that
    a number of the plurality of normal human bodies is more than 5000.
  5. The method according to claim 1, characterized in that the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
    if the difference vector is smaller than the preset temperature drift threshold, determining that the body surface temperature of the human body is normal.
  6. The method according to claim 1, characterized in that
    the temperature sample values of the preset positions are basal body temperature sample values of the human body；
    the preset body temperature reference model is a preset human body basal body temperature reference model.
  7. The method according to claim 1, characterized in that the difference matching the sample temperature vector and a preset body temperature reference model to obtain a difference vector comprises:
    setting a weight vector with a same dimension as the sample temperature vector；
    performing dot product operation of a result of subtracting the sample temperature vector by the preset body temperature reference model and the weight vector to obtain the difference vector.
  8. The method according to claim 1 or 7, characterized in that the determining whether a body surface temperature of the human body is abnormal based on a comparison between the  difference vector and a preset temperature drift threshold comprises:
    if the difference vector is greater than the preset temperature drift threshold, determining that the body surface temperature of the human body is abnormal.
  9. The method according to claim 8, characterized in that the method further comprises:
    calculating Euclidean distances between the sample temperature vector and a plurality of abnormal body surface temperature reference models respectively to form a first Euclidean distance vector；
    acquiring an abnormal body surface temperature reference model corresponding to a vector element smaller than a preset Euclidean distance value in the first Euclidean distance vector associated with the human body.
  10. The method according to claim 9, characterized in that the method further comprises:
    determining abnormal body surface temperature reference models associated with all vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  11. The method according to claim 9, characterized in that the method further comprises:
    determining abnormal body surface temperature reference models associated with all minimum value vector element smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  12. A health assessing expert system based on human body temperature modeling, comprising:
    a sampling module configured for acquiring temperature sample values of at least two preset positions of a human body to form a sample temperature vector；
    a matching module configured for difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector； and
    a determining module configured for determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
  13. The system according to claim 12, characterized in that the sampling module is further configured to:
    build the preset body temperature reference model by acquiring temperature values of preset positions of a plurality of normal human bodies.
  14. The system according to claim 13, characterized in that the acquiring temperature values of preset positions of a plurality of normal human bodies comprises:
    continuously acquiring the temperature values of the preset positions of the plurality of normal human bodies within a preset period of time.
  15. The system according to claim 13 or 14, characterized in that
    a number of the plurality of normal human bodies is more than 5000.
  16. The system according to claim 12, characterized in that the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
    if the difference vector is smaller than the preset temperature drift threshold, determining that the body surface temperature of the human body is normal.
  17. The system according to claim 12, characterized in that
    the temperature sample values of the preset positions are basal body temperature sample values of the human body；
    the preset body temperature reference model is a preset human body basal body temperature reference model.
  18. The system according to claim 12, characterized in that the difference matching the sample temperature vector and a preset body temperature reference model to obtain a difference vector comprises:
    setting a weight vector with a same dimension as the sample temperature vector；
    performing dot product operation of a result of subtracting the sample temperature vector by the preset body temperature reference model and the weight vector to obtain the difference vector.
  19. The system according to claim 12 or 18, characterized in that the determining whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold comprises:
    if the difference vector is greater than the preset temperature drift threshold, determining that the body surface temperature of the human body is abnormal.
  20. The system according to claim 19, characterized in that the system further comprises:
    a calculating module configured for calculating Euclidean distances between the sample temperature vector and a plurality of abnormal body surface temperature reference models respectively to form a first Euclidean distance vector；
    a selecting module configure for acquiring an abnormal body surface temperature reference model corresponding to a vector element smaller than a preset Euclidean distance value in the first Euclidean distance vector associated with the human body.
  21. The system according to claim 20, characterized in that the selecting module is further configured to:
    determine abnormal body surface temperature reference models associated with all vector elements smaller than the preset Euclidean distance value in the first Euclidean distance vector as abnormality models of the human body.
  22. The system according to claim 20, characterized in that the selecting model is further configured to:
    determine abnormal body surface temperature reference models associated with all minimum value vector elements smaller than the preset Euclidean distance value in the first  Euclidean distance vector as abnormality models of the human body.
  23. An electronic device comprising a processor, a memory, a communication interface, and a bus, wherein
    the processor, the memory, and the communication interface are connected and communicate with each other via the bus；
    the memory stores executable program codes；
    the processor executes a program corresponding to the executable program codes by reading the executable program codes stored in the memory, so as to:
    acquire temperature sample values of at least two preset positions of a human body to form a sample temperature vector；
    difference matching the sample temperature vector with a preset body temperature reference model to obtain a difference vector； and
    determine whether a body surface temperature of the human body is abnormal based on a comparison between the difference vector and a preset temperature drift threshold.
PCT/CN2016/104943 2015-11-09 2016-11-07 Health assessing method and expert system based on human body temperature modeling WO2017080425A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510757081.4A CN105260620B (en) 2015-11-09 2015-11-09 Health evaluating method and expert system based on body temperature modeling
CN201510757081.4 2015-11-09

Publications (1)

Publication Number Publication Date
WO2017080425A1 true WO2017080425A1 (en) 2017-05-18

Family

ID=55100309

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2016/104943 WO2017080425A1 (en) 2015-11-09 2016-11-07 Health assessing method and expert system based on human body temperature modeling

Country Status (2)

Country Link
CN (1) CN105260620B (en)
WO (1) WO2017080425A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113520306A (en) * 2020-04-17 2021-10-22 青岛海尔空调器有限总公司 Human sleep state monitoring method and intelligent household device

Families Citing this family (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105260620B (en) * 2015-11-09 2019-01-29 上海温尔信息科技有限公司 Health evaluating method and expert system based on body temperature modeling
CN105740614A (en) * 2016-01-28 2016-07-06 上海温尔信息科技有限公司 Comparison assessment method and device for humiture field model based on cancer cell abnormality
CN105760701A (en) * 2016-03-22 2016-07-13 上海温尔信息科技有限公司 Method and device for capturing abnormal temperature metabolic heat of tumor cells
WO2018133080A1 (en) * 2017-01-22 2018-07-26 上海温尔信息科技有限公司 Temperature analysis method and device
CN107890338A (en) * 2017-09-22 2018-04-10 杭州爱上伊文化创意有限公司 A kind of breast development data collecting system and its underwear
CN107997747A (en) * 2017-12-27 2018-05-08 上海传英信息技术有限公司 Body temperature detection method and terminal
CN108324250B (en) * 2018-02-12 2021-07-20 厚磊 Human body thermal metabolism state monitoring method based on infrared imaging temperature index
CN108703776A (en) * 2018-05-25 2018-10-26 上海夏先机电科技发展有限公司 A kind of method, apparatus and electronic equipment of determining user's physiological stage
CN109009017B (en) * 2018-07-03 2021-05-07 中南大学 Intelligent health monitoring system and data processing method thereof
CN109726326A (en) * 2018-11-28 2019-05-07 国云科技股份有限公司 A kind of books recommended method closest based on standardization Euclidean distance and K
CN109727667B (en) * 2019-02-27 2022-05-31 中国人民解放军第四军医大学 Metabolic state evaluation system based on postprandial deep body temperature time variability analysis
CN112102849A (en) * 2019-06-18 2020-12-18 成都医学院 Sound analysis method and device
CN111202502B (en) * 2020-01-19 2021-06-15 珠海格力电器股份有限公司 Health index detection method and system, storage medium and health monitoring equipment
CN112086202B (en) * 2020-08-03 2022-11-22 北京邮电大学 Body temperature sequence occupational imaging method
CN112315458B (en) * 2020-11-23 2021-08-20 山东大学 Movement dysfunction recognition device, system and method
CN113029392B (en) * 2021-04-15 2024-05-31 高创(苏州)电子有限公司 Body temperature measurement method and device and electronic equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102319060A (en) * 2011-09-19 2012-01-18 广州天绎智能科技有限公司 Abnormal body temperature detection method and detection system
CN102908130A (en) * 2005-11-29 2013-02-06 风险获利有限公司 Residual-based monitoring of human health
US20140350352A1 (en) * 2013-05-23 2014-11-27 Children's Medical Center Corporation System and method of assessing stability of patients
CN105260620A (en) * 2015-11-09 2016-01-20 上海温尔信息科技有限公司 Health assessment method and expert system based on human body temperature modeling

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN100353152C (en) * 2003-07-04 2007-12-05 北方工业大学 Method for monitoring temperature of rotary kiln barrel through infrared scanning
CN102349757B (en) * 2011-07-20 2013-03-06 上海波汇通信科技有限公司 Home intelligent chair capable of measuring body temperature of human beings
CN103371807A (en) * 2012-04-12 2013-10-30 夏普株式会社 Method and equipment for determining reference body temperature
CN104083152B (en) * 2014-07-17 2016-05-18 上海翰临电子科技有限公司 A kind of body temperature monitoring method and Thermometer System
CN104605939B (en) * 2015-02-05 2019-07-16 腾讯科技（深圳）有限公司 Physiologic information processing method and information processing unit

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102908130A (en) * 2005-11-29 2013-02-06 风险获利有限公司 Residual-based monitoring of human health
CN102319060A (en) * 2011-09-19 2012-01-18 广州天绎智能科技有限公司 Abnormal body temperature detection method and detection system
US20140350352A1 (en) * 2013-05-23 2014-11-27 Children's Medical Center Corporation System and method of assessing stability of patients
CN105260620A (en) * 2015-11-09 2016-01-20 上海温尔信息科技有限公司 Health assessment method and expert system based on human body temperature modeling

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113520306A (en) * 2020-04-17 2021-10-22 青岛海尔空调器有限总公司 Human sleep state monitoring method and intelligent household device
CN113520306B (en) * 2020-04-17 2024-03-22 青岛海尔空调器有限总公司 Human sleep state monitoring method and intelligent household device

Also Published As

Publication number Publication date
CN105260620B (en) 2019-01-29
CN105260620A (en) 2016-01-20

Similar Documents

Publication Publication Date Title
WO2017080425A1 (en) Health assessing method and expert system based on human body temperature modeling
US11058327B2 (en) Detecting medical status and cognitive impairment utilizing ambient data
JP6723952B2 (en) Non-transitory computer readable medium, device and method
US11154203B2 (en) Detecting fever from images and temperatures
Huang et al. Hidden Markov models for monitoring circadian rhythmicity in telemetric activity data
US20200221956A1 (en) Smartglasses for detecting congestive heart failure
US10365120B2 (en) Device, method and system for counting the number of cycles of a periodic movement of a subject
KR20170031757A (en) Data tagging
US10376207B2 (en) Calculating a current circadian rhythm of a person
Pustiek et al. Challenges in wearable devices based pervasive wellbeing monitoring
JP4830765B2 (en) Activity measurement system
US20210369143A1 (en) Method and system utilizing pattern recognition for detecting atypical movements during physical activity
Tlili et al. Design and architecture of smart belt for real time posture monitoring
WO2024055931A1 (en) Exercise recommendation method and apparatus, sleep recommendation method and apparatus, electronic device, and storage medium
US20200324074A1 (en) Electronic device and method for providing information for stress relief by same
Fotouhi-Ghazvini et al. Wearable wireless sensors for measuring calorie consumption
CN115802931A (en) Detecting temperature of a user and assessing physiological symptoms of a respiratory condition
JP2015141593A (en) Server device, server processing method, program, client device, and terminal processing method
KR20210060246A (en) The arraprus for obtaining biometiric data and method thereof
US20170039335A1 (en) Nearest neighbor predictions for providing health insights
Bogen et al. The association between daily walking behavior and self-reported physical function in community-dwelling older adults
CN114947773A (en) Body temperature monitoring method and device, terminal equipment and storage medium
US20210022663A1 (en) Hydration assessment using a binary multispectral sensor
PEDRAZA et al. A novel model to quantify balance alterations in older adults based on the center of pressure (CoP) measurements with a cross-sectional study
Marcello et al. Daily activities monitoring of users for well-being and stress correlation using wearable devices

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16863601

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC, EPO FORM 1205A DATED 23.08.18

122 Ep: pct application non-entry in european phase

Ref document number: 16863601

Country of ref document: EP

Kind code of ref document: A1