Movatterモバイル変換


[0]ホーム

URL:


CN120167898A - Parkinson's disease quantitative early diagnosis system based on behavioral feature recognition - Google Patents

Parkinson's disease quantitative early diagnosis system based on behavioral feature recognition
Download PDF

Info

Publication number
CN120167898A
CN120167898ACN202510267356.XACN202510267356ACN120167898ACN 120167898 ACN120167898 ACN 120167898ACN 202510267356 ACN202510267356 ACN 202510267356ACN 120167898 ACN120167898 ACN 120167898A
Authority
CN
China
Prior art keywords
writing
patient
data
internet
things
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.)
Granted
Application number
CN202510267356.XA
Other languages
Chinese (zh)
Other versions
CN120167898B (en
Inventor
尹西
高中宝
王炜
王淼
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.)
Second Medical Center of PLA General Hospital
Original Assignee
Second Medical Center of PLA General Hospital
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 Second Medical Center of PLA General HospitalfiledCriticalSecond Medical Center of PLA General Hospital
Priority to CN202510267356.XApriorityCriticalpatent/CN120167898B/en
Priority claimed from CN202510267356.XAexternal-prioritypatent/CN120167898B/en
Publication of CN120167898ApublicationCriticalpatent/CN120167898A/en
Application grantedgrantedCritical
Publication of CN120167898BpublicationCriticalpatent/CN120167898B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

Translated fromChinese

本发明公开了基于行为特征识别的帕金森定量化早期诊断系统,属于帕金森诊断技术领域,包括:智能采集设备,用于采集患者运动数据、患者语音数据、患者书写数据及患者面部数据,确定基于物联网的患者行为实时数据,基于无线网络将基于物联网的患者行为实时数据传输给定量化诊断平台;定量化诊断平台,用于对基于物联网的患者行为实时数据进行预处理及分析,并对患者进行帕金森定量化早期诊断,根据患者诊断结果,对患者进行智能化管控治疗。本发明解决了现有的不能基于行为特征识别对患者进行帕金森定量化早期诊断,导致帕金森患者诊断效果差的问题。本发明可基于行为特征识别对患者进行帕金森定量化早期诊断,可提升帕金森患者诊断效果。

The present invention discloses a quantitative early diagnosis system for Parkinson's disease based on behavioral feature recognition, which belongs to the technical field of Parkinson's disease diagnosis, and includes: an intelligent acquisition device for collecting patient movement data, patient voice data, patient writing data and patient facial data, determining real-time patient behavior data based on the Internet of Things, and transmitting the real-time patient behavior data based on the Internet of Things to a given quantitative diagnosis platform based on a wireless network; a quantitative diagnosis platform for preprocessing and analyzing the real-time patient behavior data based on the Internet of Things, and performing quantitative early diagnosis of Parkinson's disease on patients, and performing intelligent management and treatment on patients according to the patient's diagnosis results. The present invention solves the problem that the existing quantitative early diagnosis of Parkinson's disease on patients cannot be performed based on behavioral feature recognition, resulting in poor diagnostic results for Parkinson's disease patients. The present invention can perform quantitative early diagnosis of Parkinson's disease on patients based on behavioral feature recognition, which can improve the diagnostic results for Parkinson's disease patients.

Description

Parkinson quantification early diagnosis system based on behavior feature recognition
Technical Field
The invention relates to the technical field of parkinsonism diagnosis, in particular to a parkinsonism quantitative early diagnosis system based on behavior feature recognition.
Background
Parkinson's disease is a complex syndrome involving multiple organs, multiple systems and multiple neurotransmitters, and is not a mere dyskinesia. The progress of the disease of the patients suffering from the middle and late-stage parkinsonism, side effects of medicines, movement complications, depression, sleep disorder, pain, fatigue and other numerous non-movement symptoms are interwoven together, so that the treatment difficulty is greatly increased. Most patients with advanced parkinson's disease are unable to self-care and require home treatment, so early diagnosis is critical to delay disease progression and improve quality of life for the patient.
The Chinese patent with publication number CN117253592A discloses a whole course management mode of Parkinson's disease, which comprises a management team consisting of parkinsonist specialists and parkinsonist sub-specialists in tertiary hospitals, wherein parkinsonist specialists in tertiary hospitals train parkinsonist sub-specialist community doctors to achieve homogenization of parkinsonist diagnosis and treatment, parkinsonist sub-specialist community doctors are responsible for executing and follow-up treatment schemes of all patients entering groups and for home treatment of advanced parkinsonist patients which cannot be self-managed, information sharing is carried out between tertiary hospitals and communities through a parkinsonist management platform, and micro-communication is used as medium between doctors and patients. The management mode takes parkinsonist doctors as the leading mode, takes parkinsonist sub-specialized community doctors as the center and patients as the main body, provides multi-disciplinary, whole-course, individual and online and offline integrated management for early-stage parkinsonism patients and middle-late-stage parkinsonism patients, but the patent has the following defects:
The prior art cannot conduct early diagnosis of parkinsonism quantification on patients based on behavior feature recognition, so that parkinsonism patients are poor in diagnosis effect.
Disclosure of Invention
The invention aims to provide a parkinsonism quantitative early diagnosis system based on behavior feature recognition, which can perform parkinsonism quantitative early diagnosis on a patient based on the behavior feature recognition, can improve the diagnosis effect of the parkinsonism patient and solves the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions:
The system comprises an intelligent acquisition device and a quantitative diagnosis platform, wherein the intelligent acquisition device is configured to acquire the behavior characteristics of a patient in the course of the behavior activities of the patient, determine the real-time data of the behavior of the patient, and transmit the acquired real-time data of the behavior of the patient to the quantitative diagnosis platform based on a wireless transmission technology;
the writing photographing device is used for performing writing and photographing according to the writing tremble times and the corresponding writing pressure values;
The quantitative diagnosis platform is configured to construct a model for early diagnosis of parkinsonism quantitative based on the behavior characteristics of the patient, and analyze real-time data of the behavior of the patient and early diagnosis of parkinsonism quantitative based on the constructed model.
Preferably, the intelligent acquisition equipment comprises a motion acquisition unit, a voice acquisition unit, a writing acquisition unit and a face acquisition unit;
The motion acquisition unit is configured to monitor and acquire gait data, balance data and finger knocking data of a patient in real time based on the internet of things technology to acquire motion data of the patient;
The voice acquisition unit is configured to monitor and acquire voice characteristics and voice contents of a patient in real time based on the internet of things technology to acquire voice data of the patient;
The writing acquisition unit is configured to monitor and acquire writing dynamics data and writing graphic data of a patient in real time based on the internet of things technology to acquire writing data of the patient;
The face acquisition unit is configured to monitor and acquire face action data and facial expression data of a patient in real time based on the internet of things technology, and acquire the face data of the patient;
and determining real-time data of patient behaviors based on the Internet of things according to the acquired patient motion data, the patient voice data, the patient writing data and the patient face data.
Preferably, the writing acquisition unit comprises an intelligent writing pen and a writing photographing device, wherein wireless communication connection is established between the intelligent writing pen and the writing photographing device:
The intelligent writing pen collects the writing tremble times and the writing pressure of a patient in real time in a unit time;
comparing the writing tremble times and the writing pressure with a preset writing tremble times threshold and a preset writing pressure threshold respectively;
When the writing tremble times exceeds a preset writing tremble times threshold, but the writing pressure does not exceed the writing pressure threshold, triggering a writing photographing device to photograph through an intelligent writing pen, and obtaining a first writing image;
When the writing tremble times do not exceed a preset writing tremble times threshold, but the writing pressure exceeds a writing pressure threshold, comprehensively judging whether to trigger a writing photographing device to photograph or not according to the writing tremble times and corresponding writing pressure values;
when the writing tremble times and the writing pressure do not exceed the corresponding writing tremble times threshold and writing pressure threshold, controlling a writing photographing device to photograph according to a preset image acquisition frequency;
And setting the evaluation weight value of the first writing image to be higher than the corresponding evaluation weight value of the second writing image.
Preferably, the method for comprehensively judging whether to trigger the writing photographing device to photograph by the corresponding values of the writing tremble times and the writing pressure comprises the following steps:
When the writing pressure exceeds a writing pressure threshold, extracting the writing tremble times corresponding to unit time when the writing tremble times do not exceed a preset writing tremble times threshold, and taking the writing tremble times as a target tremble times data set;
Generating a comprehensive writing state parameter by using the target tremble frequency data set and the writing pressure, wherein the comprehensive writing state parameter is used for evaluating whether the writing state of the current patient has the necessity of image acquisition or not;
Wherein, the comprehensive writing state parameter is obtained by the following formula:
The method comprises the steps of S representing comprehensive writing state parameters, P representing writing pressure corresponding to a writing pressure threshold, Py representing the writing pressure threshold, N representing the number of data contained in a target trembling frequency data set, wherein the number of data is consistent with the number of unit time when the number of the writing trembling frequency does not exceed a preset writing trembling frequency threshold, Ci representing the number of trembling frequencies corresponding to the ith trembling data in the target trembling frequency data set, N representing the number of unit time which a patient has written, N0 representing a preset unit time number reference value, and Cy representing a preset writing trembling frequency threshold;
Comparing the comprehensive writing state parameter with a preset parameter threshold;
when the comprehensive writing state parameter is lower than a preset parameter threshold, judging that the writing photographing device is not required to be triggered to photograph;
When the comprehensive writing state parameter is not lower than a preset parameter threshold, judging that the writing photographing device needs to be triggered to photograph;
Taking a writing image obtained when the writing tremble times do not exceed a preset writing tremble times threshold value but the writing pressure exceeds a writing pressure threshold value as a third writing image;
and setting the evaluation weight value of the third writing image to be higher than the evaluation weight value corresponding to the second writing image, but lower than the evaluation weight value corresponding to the first writing image.
Preferably, the gait data comprises step size, step width, step frequency, step speed, swing phase, support phase, gait symmetry, gait variability, torso swing, arm swing and cornering gait;
the balance data comprise standing balance, sitting balance, gravity center swing and posture stability;
the finger knocking data comprise knocking frequency, rhythm, force and finger coordination;
the voice features include fundamental frequency, sound intensity, sound length, speech speed, pause, intonation, pronunciation clarity and voice fluency;
The voice content comprises vocabulary, grammar structure, semantic expression and emotion expression;
writing dynamics data including writing speed, acceleration and pressure, stroke length, width and angle, writing pause and tremble;
The writing graphic data comprises font size, shape and spacing, handwriting integrity and consistency;
Facial motion data includes movements of eyebrows, eyes, nose, and mouth portions;
facial expression data includes expression change frequency, amplitude, and duration.
Preferably, the quantitative diagnosis platform comprises a behavior data preprocessing unit and an early diagnosis evaluation unit;
The behavior data preprocessing unit is configured to clean, transform, integrate and extract characteristics of real-time data of patient behaviors based on the Internet of things, and determine characteristic data of the patient behaviors based on the Internet of things;
the early diagnosis evaluation unit is configured to analyze the patient behavior characteristic data based on the Internet of things, conduct parkinsonism quantitative early diagnosis on the patient, and conduct intelligent management and control treatment on the patient according to the diagnosis result of the patient.
Preferably, the behavior data preprocessing unit comprises a behavior data cleaning module and a behavior data conversion module;
the behavior data cleaning module is configured to clean real-time data of patient behaviors based on the Internet of things;
the method comprises the steps of checking real-time data of patient behaviors based on the Internet of things based on a Python library, identifying repeated values, missing values and abnormal values in the real-time data of the patient behaviors based on the Internet of things, and deleting the repeated values, the missing values and the abnormal values in the real-time data of the patient behaviors based on the Internet of things;
the behavior data transformation module is configured to transform the real-time data of the patient behavior based on the Internet of things;
The method comprises the steps of normalizing real-time data of patient behaviors based on the Internet of things based on Z-Score standardization, converting the real-time data of the patient behaviors based on the Internet of things into standard normal distribution with a mean value of 0 and a standard deviation of 1, and removing dimension differences among the real-time data of the patient behaviors based on the Internet of things.
Preferably, the behavior data preprocessing unit further comprises a behavior data integrating module and a behavior data extracting module;
the behavior data integration module is configured to integrate real-time data of patient behaviors based on the Internet of things;
Integrating the real-time data of the patient behaviors based on the Internet of things based on the Python library, integrating the real-time data of the patient behaviors based on the Internet of things into a unified data view, verifying the integrated real-time data of the patient behaviors based on the Internet of things, and judging whether the integrated real-time data of the patient behaviors based on the Internet of things is absent;
The behavior data extraction module is configured to extract real-time data of patient behaviors based on the Internet of things;
The method comprises the steps of extracting characteristics related to Parkinson's disease from real-time data of patient behaviors based on the Internet of things, carrying out weighted fusion on the extracted characteristic vectors, and determining the patient behavior characteristic data based on the Internet of things.
Preferably, the early diagnosis evaluation unit comprises a diagnosis model training module, a diagnosis model analysis module and a diagnosis model analysis module, wherein the diagnosis model training module is configured to train a parkinsonism quantitative early diagnosis model;
Collecting patient behavior historical data based on the Internet of things according to parkinsonism quantification early diagnosis requirements based on behavior feature recognition, dividing the collected patient behavior historical data based on the Internet of things, and dividing the patient behavior historical data based on the Internet of things into a training set and a testing set;
based on a deep learning technology, training a deep learning model by adopting a training set, enabling the deep learning model to autonomously learn a parkinsonism quantitative early diagnosis process based on behavior feature recognition, and determining a parkinsonism quantitative early diagnosis model based on the behavior feature recognition;
Based on a test set, testing the parkinsonism quantitative early diagnosis model based on behavior feature recognition, evaluating the performance of the parkinsonism quantitative early diagnosis model based on the behavior feature recognition based on the accuracy, recall rate and F1 value index, and judging whether the parkinsonism quantitative early diagnosis model based on the behavior feature recognition can achieve the expected effect;
And adjusting parameters and structures of the parkinsonism quantitative early diagnosis model based on behavior feature recognition according to the test evaluation result, and determining an optimal parkinsonism quantitative early diagnosis model through continuous iterative optimization.
Preferably, the early diagnosis and evaluation unit further comprises a diagnosis and evaluation management and control module, wherein the diagnosis and evaluation management and control module is configured to conduct parkinsonism quantitative early diagnosis and evaluation management and control on a patient;
acquiring an optimal parkinsonism quantitative early diagnosis model, and disposing the optimal parkinsonism quantitative early diagnosis model in an actual parkinsonism quantitative early diagnosis environment based on behavior feature recognition;
Inputting the patient behavior characteristic data based on the Internet of things into an optimal parkinsonism quantitative early diagnosis model, analyzing the patient behavior characteristic data based on the Internet of things based on the optimal parkinsonism quantitative early diagnosis model, performing parkinsonism quantitative early diagnosis on a patient, and determining a patient diagnosis result;
the system comprises a patient diagnosis module, a patient monitoring module and a patient monitoring module, wherein the patient behavior characteristic data are analyzed based on the patient diagnosis result, a patient condition evaluation report is determined, intelligent management and control treatment is carried out on the patient based on the patient condition evaluation report, the treatment effect is monitored in real time, and the management and control treatment scheme of the patient is dynamically adjusted according to the monitoring condition.
Compared with the prior art, the invention has the beneficial effects that:
According to the invention, the real-time data of the patient behaviors based on the Internet of things are determined by collecting the patient motion data, the patient voice data, the patient writing data and the patient facial data, the real-time data of the patient behaviors based on the Internet of things are cleaned, transformed, integrated and extracted, the characteristic data of the patient behaviors based on the Internet of things are determined, the parkinsonism quantitative early diagnosis model is trained based on the deep learning technology, the characteristic data of the patient behaviors based on the Internet of things are analyzed based on the parkinsonism quantitative early diagnosis model, the parkinsonism quantitative early diagnosis is carried out on the patient, the patient diagnosis result is determined, the characteristic data of the patient behaviors are analyzed based on the patient diagnosis result, the condition evaluation report of the patient is determined, the intelligent management and control treatment is carried out on the patient based on the condition evaluation report of the patient, the real-time monitoring is carried out on the treatment effect, the management and control treatment scheme of the patient is dynamically adjusted according to the monitoring condition, the parkinsonism quantitative early diagnosis is carried out on the patient based on the behavior characteristic identification, and the parkinsonism patient diagnosis effect can be improved.
Drawings
FIG. 1 is a block diagram of a parkinsonism quantitative early diagnosis system based on behavioral feature recognition of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In order to solve the problem that the existing method cannot perform parkinsonian quantitative early diagnosis on a patient based on behavior feature recognition, resulting in poor diagnosis effect on parkinsonian patients, referring to fig. 1, the present embodiment provides the following technical scheme:
The parkinsonism quantitative early diagnosis system based on behavior feature recognition comprises intelligent acquisition equipment and a quantitative diagnosis platform.
The intelligent acquisition equipment is used for acquiring patient motion data, patient voice data, patient writing data and patient face data, determining patient behavior real-time data based on the Internet of things, and transmitting the acquired patient behavior real-time data based on the Internet of things to the quantitative diagnosis platform based on a wireless transmission technology;
in this embodiment, the intelligent acquisition device includes:
The motion acquisition unit is used for monitoring and acquiring gait data, balance data and finger knocking data of a patient in real time based on the internet of things technology to acquire motion data of the patient;
The gait data include step size, step width, step frequency, step speed, swing phase, support phase, gait symmetry, gait variability, torso swing, arm swing and cornering gait.
The balance data includes standing balance, sitting balance, center of gravity swing, and posture stability.
It should be noted that the finger knocking data includes knocking frequency, rhythm, dynamics and finger coordination.
The voice acquisition unit is used for monitoring and acquiring voice characteristics and voice contents of a patient in real time based on the internet of things technology to acquire voice data of the patient;
It should be noted that the speech features include fundamental frequency, intensity, duration, speech speed, pause, intonation, articulation clarity, and speech fluency.
It should be noted that, the voice content includes vocabulary, grammar structure, semantic expression and emotion expression;
the writing acquisition unit is used for monitoring and acquiring writing dynamics data and writing graphic data of a patient in real time based on the internet of things technology to acquire writing data of the patient;
the writing dynamics data includes writing speed, acceleration and pressure, stroke length, width and angle, writing pause and tremble.
The writing graphic data comprises font size, shape and interval, handwriting integrity and consistency;
The facial acquisition unit is used for monitoring and acquiring facial action data and facial expression data of a patient in real time based on the internet of things technology to acquire the facial data of the patient;
The facial motion data includes movements of eyebrows, eyes, nose, and mouth.
It should be noted that the facial expression data includes expression change frequency, amplitude and duration.
And determining real-time data of the patient behavior based on the Internet of things according to the patient motion data, the patient voice data, the patient writing data and the patient face data.
Specifically, by collecting patient motion data, patient voice data, patient writing data and patient face data, real-time patient behavior data based on the Internet of things is determined, and data support can be provided for early diagnosis of parkinsonism quantification of a patient.
The quantitative diagnosis platform is used for preprocessing and analyzing real-time data of patient behaviors based on the Internet of things, performing parkinsonism quantitative early diagnosis on a patient, and performing intelligent management and control treatment on the patient according to a patient diagnosis result.
Specifically, write the collection unit, write the pen including intelligence and write and shoot the device, wherein, the intelligence write the pen with write and shoot and establish wireless communication connection between the device:
The intelligent writing pen collects the writing tremble times and the writing pressure of a patient in real time in a unit time;
comparing the writing tremble times and the writing pressure with a preset writing tremble times threshold and a preset writing pressure threshold respectively;
When the writing tremble times exceeds a preset writing tremble times threshold, but the writing pressure does not exceed the writing pressure threshold, triggering a writing photographing device to photograph through an intelligent writing pen, and obtaining a first writing image;
When the writing tremble times do not exceed a preset writing tremble times threshold, but the writing pressure exceeds a writing pressure threshold, comprehensively judging whether to trigger a writing photographing device to photograph or not according to the writing tremble times and corresponding writing pressure values;
when the writing tremble times and the writing pressure do not exceed the corresponding writing tremble times threshold and writing pressure threshold, controlling a writing photographing device to photograph according to a preset image acquisition frequency;
And setting the evaluation weight value of the first writing image to be higher than the corresponding evaluation weight value of the second writing image.
The technical effect of the technical scheme is that the intelligent writing pen can be used for collecting the writing tremble times and the writing pressure of a patient in real time, so that the collection rate of the writing state information of the patient and the timeliness of data acquisition can be effectively improved. Meanwhile, whether the writing photographing device is triggered to photograph or not is judged according to the preset writing tremble frequency threshold and the writing pressure threshold, so that the accuracy of judging abnormal conditions in the writing process of a captured patient can be effectively improved, the necessity of triggering is improved, and the problem that important and key writing images cannot be successfully captured when the image photographing is carried out according to the fixed photographing frequency is prevented.
When one of the number of writing tremors or writing pressure exceeds a threshold, the system may take a different photographing strategy. And when the number of the writing judder exceeds the threshold value, shooting is carried out according to the preset image acquisition frequency. The shooting triggering strategy can effectively improve shooting pertinence, and further improve accuracy of subsequent parkinsonism assessment. On the other hand, the system sets the evaluation weight value of the first written image (i.e., the photographing triggered under the specific condition) to be higher than the corresponding evaluation weight value of the second written image (i.e., the image photographed at the preset frequency). The weight assignment method can be used for focusing on key image data in the analysis process, so that the accuracy and the effectiveness of evaluation are improved.
Specifically, whether the writing photographing device is triggered to photograph or not is comprehensively judged through corresponding values of the writing tremble times and the writing pressure, and the method comprises the following steps:
When the writing pressure exceeds a writing pressure threshold, extracting the writing tremble times corresponding to unit time when the writing tremble times do not exceed a preset writing tremble times threshold, and taking the writing tremble times as a target tremble times data set;
Generating a comprehensive writing state parameter by using the target tremble frequency data set and the writing pressure, wherein the comprehensive writing state parameter is used for evaluating whether the writing state of the current patient has the image acquisition necessity or not;
Wherein, the comprehensive writing state parameter is obtained by the following formula:
The method comprises the steps of S representing comprehensive writing state parameters, P representing writing pressure corresponding to a writing pressure threshold, Py representing the writing pressure threshold, N representing the number of data contained in a target trembling frequency data set, wherein the number of data is consistent with the number of unit time when the number of the writing trembling frequency does not exceed a preset writing trembling frequency threshold, Ci representing the number of trembling frequencies corresponding to the ith trembling data in the target trembling frequency data set, N representing the number of unit time which a patient has written, N0 representing a preset unit time number reference value, and Cy representing a preset writing trembling frequency threshold;
Comparing the comprehensive writing state parameter with a preset parameter threshold;
when the comprehensive writing state parameter is lower than a preset parameter threshold, judging that the writing photographing device is not required to be triggered to photograph;
When the comprehensive writing state parameter is not lower than a preset parameter threshold, judging that the writing photographing device needs to be triggered to photograph;
Taking a writing image obtained when the writing tremble times do not exceed a preset writing tremble times threshold value but the writing pressure exceeds a writing pressure threshold value as a third writing image;
and setting the evaluation weight value of the third writing image to be higher than the evaluation weight value corresponding to the second writing image, but lower than the evaluation weight value corresponding to the first writing image.
The technical effect of the technical scheme is that the technical scheme can evaluate the writing state of the patient more finely by introducing the comprehensive writing state parameter S. The accuracy of the evaluation of the real writing situation of the patient can be effectively improved by comprehensively evaluating the writing judder times and the writing pressure. Meanwhile, the judder times and pressure changes in different unit time in the writing process of the patient are utilized in the technical scheme, the comprehensive writing state parameters are obtained by extracting the target judder times data set, the following performance between the comprehensive writing state parameters and the fine changes of the writing state of the patient can be effectively improved, and further the capturing sensitivity and the capturing accuracy of the fine changes of the writing state of the patient are effectively improved. Meanwhile, in the technical scheme, weight assignment is carried out on the writing images acquired under different conditions. The evaluation weight value of the third written image is higher than that of the second written image but lower than that of the first written image, the weight assignment mode not only considers the importance difference of the images under different conditions, but also keeps the rationality and continuity of weight assignment, and the importance distinction degree of the first written image, the second written image and the third written image can be effectively improved, so that the accuracy of early judgment of subsequent Parkinson is improved.
In the embodiment, the quantitative diagnosis platform comprises a behavior data preprocessing unit and an early diagnosis evaluation unit.
The behavior data preprocessing unit is used for cleaning, transforming, integrating and extracting characteristics of the patient behavior real-time data based on the Internet of things, and determining patient behavior characteristic data based on the Internet of things;
it should be noted that, the behavior data preprocessing unit includes:
The behavior data cleaning module is used for cleaning the real-time data of the patient behaviors based on the Internet of things;
the method comprises the steps of checking real-time data of patient behaviors based on the Internet of things based on a Python library, identifying repeated values, missing values and abnormal values in the real-time data of the patient behaviors based on the Internet of things, and deleting the repeated values, the missing values and the abnormal values in the real-time data of the patient behaviors based on the Internet of things;
the behavior data transformation module is used for transforming the real-time data of the patient behavior based on the Internet of things;
the method comprises the steps of normalizing real-time data of patient behaviors based on the Internet of things based on Z-Score standardization, converting the real-time data of the patient behaviors based on the Internet of things into standard normal distribution with a mean value of 0 and a standard deviation of 1, and removing dimension differences among the real-time data of the patient behaviors based on the Internet of things;
the behavior data integration module is used for integrating the real-time data of the patient behaviors based on the Internet of things;
Integrating the real-time data of the patient behaviors based on the Internet of things based on the Python library, integrating the real-time data of the patient behaviors based on the Internet of things into a unified data view, verifying the integrated real-time data of the patient behaviors based on the Internet of things, and judging whether the integrated real-time data of the patient behaviors based on the Internet of things is absent;
the behavior data extraction module is used for extracting real-time data of patient behaviors based on the Internet of things;
The method comprises the steps of extracting characteristics related to Parkinson's disease from real-time data of patient behaviors based on the Internet of things, carrying out weighted fusion on the extracted characteristic vectors, and determining the patient behavior characteristic data based on the Internet of things.
Specifically, patient behavior characteristic data based on the Internet of things is determined by cleaning, transforming, integrating and extracting characteristics of the patient behavior real-time data based on the Internet of things, wherein the patient behavior characteristic data comprises gait parameters, voice characteristics, writing dynamics characteristics, facial expression characteristics and the like.
The early diagnosis evaluation unit is used for analyzing the patient behavior characteristic data based on the Internet of things, carrying out parkinsonism quantitative early diagnosis on the patient, and carrying out intelligent management and control treatment on the patient according to the diagnosis result of the patient.
The early diagnosis evaluation unit includes:
The diagnosis model training module is used for training an early diagnosis model of parkinsonism quantification;
Collecting patient behavior historical data based on the Internet of things according to parkinsonism quantification early diagnosis requirements based on behavior feature recognition, dividing the collected patient behavior historical data based on the Internet of things, and dividing the patient behavior historical data based on the Internet of things into a training set and a testing set;
based on a deep learning technology, training a deep learning model by adopting a training set, enabling the deep learning model to autonomously learn a parkinsonism quantitative early diagnosis process based on behavior feature recognition, and determining a parkinsonism quantitative early diagnosis model based on the behavior feature recognition;
Based on a test set, testing the parkinsonism quantitative early diagnosis model based on behavior feature recognition, evaluating the performance of the parkinsonism quantitative early diagnosis model based on the behavior feature recognition based on the accuracy, recall rate and F1 value index, and judging whether the parkinsonism quantitative early diagnosis model based on the behavior feature recognition can achieve the expected effect;
Parameters and structures of the parkinsonism quantitative early diagnosis model based on behavior feature recognition are adjusted according to the test evaluation result, and the optimal parkinsonism quantitative early diagnosis model is determined through continuous iterative optimization;
the diagnosis and evaluation management and control module is used for carrying out parkinsonism quantitative early diagnosis and evaluation management and control on patients;
acquiring an optimal parkinsonism quantitative early diagnosis model, and disposing the optimal parkinsonism quantitative early diagnosis model in an actual parkinsonism quantitative early diagnosis environment based on behavior feature recognition;
Inputting the patient behavior characteristic data based on the Internet of things into an optimal parkinsonism quantitative early diagnosis model, analyzing the patient behavior characteristic data based on the Internet of things based on the optimal parkinsonism quantitative early diagnosis model, performing parkinsonism quantitative early diagnosis on a patient, and determining a patient diagnosis result;
the system comprises a patient diagnosis module, a patient monitoring module and a patient monitoring module, wherein the patient behavior characteristic data are analyzed based on the patient diagnosis result, a patient condition evaluation report is determined, intelligent management and control treatment is carried out on the patient based on the patient condition evaluation report, the treatment effect is monitored in real time, and the management and control treatment scheme of the patient is dynamically adjusted according to the monitoring condition.
Therefore, based on the parkinsonism quantitative early diagnosis model, the patient behavior feature data based on the Internet of things are analyzed, parkinsonism quantitative early diagnosis is carried out on the patient, the patient diagnosis result is determined, based on the patient diagnosis result, the patient behavior feature data is analyzed, the patient condition evaluation report is determined, intelligent management and control treatment is carried out on the patient based on the patient condition evaluation report, the treatment effect is monitored in real time, the management and control treatment scheme of the patient is dynamically adjusted according to the monitoring condition, parkinsonism quantitative early diagnosis can be carried out on the patient based on the behavior feature identification, and the parkinsonism patient diagnosis effect can be improved.
It is noted that relational terms such as first and second, and the like are 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. Moreover, 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.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims (10)

CN202510267356.XA2025-03-07 Parkinson's disease quantitative early diagnosis system based on behavioral feature recognitionActiveCN120167898B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202510267356.XACN120167898B (en)2025-03-07 Parkinson's disease quantitative early diagnosis system based on behavioral feature recognition

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202510267356.XACN120167898B (en)2025-03-07 Parkinson's disease quantitative early diagnosis system based on behavioral feature recognition

Publications (2)

Publication NumberPublication Date
CN120167898Atrue CN120167898A (en)2025-06-20
CN120167898B CN120167898B (en)2025-10-14

Family

ID=

Citations (6)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN111110244A (en)*2019-12-282020-05-08苏州同启苏沐软件有限公司Parkinson's syndrome diagnostic device
CN115101191A (en)*2022-08-262022-09-23大连理工大学Parkinson disease diagnosis system
CN115346661A (en)*2021-09-032022-11-15中国人民解放军总医院Tremor symptom detection method and equipment based on electronic handwriting
CN115985490A (en)*2023-03-172023-04-18四川大学华西医院 An objective and quantitative early diagnosis system and storage medium for Parkinson's disease
CN118490225A (en)*2024-05-082024-08-16北京健康有益科技有限公司 A diversified physiological and psychological signal index collection method and system
CN119296765A (en)*2024-09-232025-01-10中国人民解放军总医院第二医学中心 Early diagnosis system of Parkinson's disease based on handwriting behavior feature recognition

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN111110244A (en)*2019-12-282020-05-08苏州同启苏沐软件有限公司Parkinson's syndrome diagnostic device
CN115346661A (en)*2021-09-032022-11-15中国人民解放军总医院Tremor symptom detection method and equipment based on electronic handwriting
CN115101191A (en)*2022-08-262022-09-23大连理工大学Parkinson disease diagnosis system
CN115985490A (en)*2023-03-172023-04-18四川大学华西医院 An objective and quantitative early diagnosis system and storage medium for Parkinson's disease
CN118490225A (en)*2024-05-082024-08-16北京健康有益科技有限公司 A diversified physiological and psychological signal index collection method and system
CN119296765A (en)*2024-09-232025-01-10中国人民解放军总医院第二医学中心 Early diagnosis system of Parkinson's disease based on handwriting behavior feature recognition

Similar Documents

PublicationPublication DateTitle
CN110507335B (en)Multi-mode information based criminal psychological health state assessment method and system
CN111728609B (en)Electroencephalogram signal classification method, classification model training method, device and medium
CN111046823A (en)Student classroom participation degree analysis system based on classroom video
CN113537005B (en)Online examination student behavior analysis method based on attitude estimation
CN110428908B (en)Eyelid motion function evaluation system based on artificial intelligence
CN115299947B (en) Confidence evaluation method and system for psychological scales based on multimodal physiological data
CN117438048A (en)Method and system for assessing psychological disorder of psychiatric patient
CN111222464B (en)Emotion analysis method and system
CN118692662A (en) An audio and video recognition system based on multimodal analysis
CN111523445A (en) An Examination Behavior Detection Method Based on Improved Openpose Model and Facial Micro-expressions
CN115274104A (en)Digital biomarker
CN111145851A (en)Mental state monitoring and evaluating system based on intelligent bracelet
CN117496575A (en) A method and system for classroom student status analysis based on face monitoring
CN120015351A (en) Multimodal large model recognition and intelligent intervention methods and systems for mental disorders
CN113974627B (en)Emotion recognition method based on brain-computer generated confrontation
CN118512173B (en)Deep learning-based children hearing detection method and system
CN120167898B (en) Parkinson's disease quantitative early diagnosis system based on behavioral feature recognition
CN120167898A (en) Parkinson's disease quantitative early diagnosis system based on behavioral feature recognition
Gaurav et al.Characterizing neural activity from EEG signals during game interaction using persistent homology
CN115251901A (en)Parkinson disease early diagnosis method based on eye-surface motion capture technology
CN117158965A (en)Autism auxiliary diagnosis system
CN114387678A (en)Method and apparatus for evaluating language readability using non-verbal body symbols
CN119600865B (en) An artificial intelligence-based pediatric massage teaching auxiliary system
WangRapid recognition of athlete's anxiety emotion based on multimodal fusion
Guo et al.Upper limb motion recognition for unsupervised stroke rehabilitation based on Support Vector Machine

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp