AI-based early Alzheimer's disease intelligent auxiliary diagnosis method and system

Abstract

The invention discloses an AI-based intelligent auxiliary diagnosis method and system for early Alzheimer's disease. The method comprises the following steps that a patient fills in necessary personal information in the AI diagnosis system to complete registration to form an ID account number corresponding to the patient and automatically log in, and the AI diagnosis system obtains facial expression changes of the patient, obtains voices and intonations of the patient and monitors gaits and reaction time of the patient; the AI diagnosis system sends a voice prompt to guide the patient to fill in a neuropsychological scale on an interactive interface, guides the patient to complete self-evaluation, and inputs the facial expression change of the patient, the voice and intonation of the patient, the monitoring of the gait and reaction time of the patient and the filling in of the neuropsychological scale into a trained machine learning model; according to the Alzheimer's disease diagnosis system and method, the Alzheimer's disease can be rapidly screened and preliminarily diagnosed, and the workload of medical staff is relieved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence medical technology, and specifically to an AI-based intelligent auxiliary diagnosis method and system for early Alzheimer's disease. Background Art

[0002] Alzheimer's Disease (AD) is one of the most difficult diseases to cure in the elderly. AD is a degenerative, irreversible brain disease. As the disease progresses, more and more brain neurons lose function, neurons lose connection, and even die. The cortex and hippocampus of the brain shrink, and the ventricles enlarge. The external manifestation is the weakening or complete loss of memory and other cognitive abilities. Patients are divided into different stages of cognition: 1. Cognitively Normal (CN), that is, patients have no symptoms of cognitive impairment; 2. Mild cognitive impairment (MCI) 3. AD.

[0003] According to statistics, 44% of MCI patients will transform into AD within 3 years; MCI patients who transform into AD within the next 3 years are called progressive MCI (pMCI), and MCI patients who do not transform into AD within the next 3 years are called stable MCI (sMCI). Although MCI patients have some problems in language, memory, thinking ability, etc., the symptoms are not as severe as those of AD patients, so MCI patients are often mistaken for manifestations of aging and are not taken seriously. However, if MCI is completely transformed into AD, there is no effective drug or treatment to cure AD so far.

[0004] Given the irreversibility of AD, the prediction and prevention of patients during the MCI period are critical. Accurately predicting the development trend of MCI during the MCI stage of patients and timely drug treatment and psychiatric treatment can effectively slow down the deterioration of the disease and improve the quality of life of patients. Neuroimaging can intuitively reflect the pathological changes of patients, and due to its non-invasive advantages, it is more easily accepted by patients and their families. Among the methods of neuroimaging, magnetic resonance imaging (MRI) is a common method for diagnosing AD. It has a high resolution for brain soft tissue, can display brain tissue in three dimensions, and provide useful human brain anatomical information. Therefore, many algorithms based on MRI images to predict AD have been generated.

[0005] However, traditional neuroimaging examinations require patients to register at the hospital, see a doctor in person, and then get a prescription for the examination. The steps are rather cumbersome, and early Alzheimer's patients generally believe that they do not have the disease and are relatively resistant to going to the hospital for examination. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] Therefore, the purpose of the present invention is to provide an AI-based intelligent auxiliary diagnosis method and system for early Alzheimer's disease, so as to realize rapid screening and preliminary diagnosis of Alzheimer's disease, reduce the workload of medical staff, help intervene in the early stages of the disease, delay the progression of the disease, and improve the quality of life of patients.

[0008] To solve the above technical problems, according to one aspect of the present invention, the present invention provides the following technical solutions:

[0009] An AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease, the steps are as follows:

[0010] S1. The patient fills in the necessary personal information in the AI ​​diagnostic system to complete the registration and form an ID account corresponding to the patient and log in automatically;

[0011] S2, the AI ​​diagnostic system acquires changes in the patient's facial expressions, acquires the patient's voice tone, and monitors the patient's gait and reaction time;

[0012] S3, the AI ​​diagnostic system issues voice prompts to guide patients to fill in the neuropsychological scale on the interactive interface and guide patients to complete the self-assessment;

[0013] S4. The patient's facial expression changes, voice tone, gait and reaction time are monitored, and the neuropsychological scale is input into the trained machine learning model. The machine learning model conducts a comprehensive analysis of the patient, provides an Alzheimer's disease risk assessment, and automatically generates a diagnostic report, which is displayed to the patient and / or medical staff on an electronic screen.

[0014] As a preferred solution of the AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease described in the present invention, in step S2, the specific steps of the AI ​​diagnosis system obtaining the patient's facial expression changes, obtaining the patient's voice intonation and monitoring the patient's gait and reaction time are as follows:

[0015] The AI ​​diagnostic system monitors the patient’s facial expression changes, gait, and reaction time through video capture;

[0016] The AI ​​diagnostic system collects the patient's voice and intonation through voice sensors, and then cleans, denoises and standardizes the collected video and audio.

[0017] As a preferred embodiment of the AI-based intelligent assisted diagnosis method for early Alzheimer's disease described in the present invention, in step S3, the neuropsychological scale includes the MoCA scale, the depression rating scale, the sleep disorder rating scale and the ADAS-Cog scale.

[0018] As a preferred solution of the AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease described in the present invention, in step S3, the machine learning model is specifically trained as follows:

[0019] First, we used crawlers to obtain a large number of datasets of patients’ facial expressions, gait, reaction time, voice intonation, and neuropsychological scales;

[0020] Divide the dataset into training set, validation set and test set, with the proportions of 70%, 15% and 15% respectively;

[0021] Then, the model is trained, and the Adam optimizer with a learning rate of 0.0001 is used during the training process to minimize the Huber loss function, which is defined as:

[0022]

[0023] where a ij is the true value, is the predicted value, δ is the threshold to switch from quadratic loss to linear loss,

[0024] In each training round, training is performed first, then validation is performed, and the loss value is calculated.

[0025] A system for intelligent auxiliary diagnosis of early Alzheimer's disease based on AI, comprising:

[0026] The login unit is configured to: the patient fills in necessary personal information in the AI ​​diagnosis system to complete registration to form an ID account corresponding to the patient and automatically log in;

[0027] The symptom and sign detection unit is configured as follows: the AI ​​diagnostic system acquires changes in the patient's facial expression, acquires the patient's voice tone, and monitors the patient's gait and reaction time;

[0028] The self-assessment unit is configured as follows: the AI ​​diagnostic system issues voice prompts to guide the patient to fill in the neuropsychological scale on the interactive interface, guiding the patient to complete the self-assessment;

[0029] The data analysis unit is configured to: input the patient's facial expression changes, obtain the patient's voice intonation, monitor the patient's gait, reaction time, and fill out a neuropsychological scale into a trained machine learning model, and the machine learning model performs a comprehensive analysis of the patient;

[0030] The report feedback unit is configured as follows: the AI ​​diagnostic system provides an Alzheimer's disease risk assessment and automatically generates a diagnostic report, which is displayed to the patient and / or medical staff through an electronic screen.

[0031] As a preferred solution of the AI-based early Alzheimer's disease intelligent auxiliary diagnosis system described in the present invention, the symptom and sign detection unit includes a high-definition camera, a microphone and a data preprocessing module. The high-definition camera is used to capture and monitor the patient's facial expression changes, gait and reaction time. The microphone is used to collect the patient's voice and intonation. The data preprocessing module is used to clean, denoise and standardize the collected video and audio.

[0032] Compared with the prior art, the present invention has the beneficial effect that, by combining the patient's self-assessment of symptoms, physical sign detection, Montreal Cognitive Assessment Scale (MoCA), depression score, sleep disorder score, cognitive part of Alzheimer's Disease Assessment Scale and Clinical Dementia Rating Scale (CDR) and imaging, biomarkers, risk factors and other multi-dimensional information, the present invention combines electronic system automated analysis to achieve rapid diagnosis of Alzheimer's disease prevalence, reduce the workload of medical staff, help intervene in the early stages of the disease, delay the progression of the disease, improve the quality of life of patients, and provide rich data support for the study of Alzheimer's disease, and promote the exploration of disease pathogenesis and treatment methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the present invention will be described in detail below in combination with the accompanying drawings and detailed embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0034] Figure 1 This is a flow chart of an AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease of the present invention;

[0035] Figure 2 This is a principle block diagram of an AI-based early Alzheimer's disease intelligent auxiliary diagnosis system of the present invention. DETAILED DESCRIPTION

[0036] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0037] Secondly, the present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention in detail, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0038] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0039] The present invention provides an AI-based intelligent auxiliary diagnosis method and system for early Alzheimer's disease, which can realize rapid screening and preliminary diagnosis of Alzheimer's disease, reduce the workload of medical staff, help intervene in the early stages of the disease, delay the progression of the disease, and improve the quality of life of patients.

[0040] Example 1

[0041] like Figure 1 As shown, the present invention provides an AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease, and the specific steps are as follows:

[0042] S1. The patient fills in the necessary personal information in the AI ​​diagnostic system to complete the registration and forms an ID account corresponding to the patient and logs in automatically.

[0043] S2. The AI ​​diagnostic system acquires changes in the patient's facial expression, acquires the patient's voice and intonation, and monitors the patient's gait and reaction time. Specifically, in this embodiment, the AI ​​diagnostic system monitors changes in the patient's facial expression, as well as the patient's gait and reaction time through video capture. The AI ​​diagnostic system acquires the patient's voice and intonation through a voice sensor, and then cleans, denoises, and standardizes the acquired video and audio.

[0044] S3. The AI ​​diagnostic system issues voice prompts to guide the patient to fill in the neuropsychological scale on the interactive interface and guide the patient to complete the self-assessment. Specifically, in this embodiment, the neuropsychological scale includes the MoCA scale, the depression rating scale, the sleep disorder rating scale and the ADAS-Cog scale.

[0045] S4, inputting the patient's facial expression changes, obtaining the patient's voice tone, monitoring the patient's gait, reaction time, and filling out the neuropsychological scale into the trained machine learning model, the machine learning model conducts a comprehensive analysis of the patient, gives an Alzheimer's disease risk assessment, and automatically generates a diagnosis report, which is displayed to the patient and / or medical staff through an electronic screen. Specifically, in this embodiment, the machine learning model is specifically trained as follows:

[0046] First, we used crawlers to obtain a large number of datasets of patients’ facial expressions, gait, reaction time, voice intonation, and neuropsychological scales;

[0047] Divide the dataset into training set, validation set and test set, with the proportions of 70%, 15% and 15% respectively;

[0048] Then, the model is trained, and the Adam optimizer with a learning rate of 0.0001 is used during the training process to minimize the Huber loss function, which is defined as:

[0049]

[0050] where a ij is the true value, is the predicted value, δ is the threshold to switch from quadratic loss to linear loss,

[0051] In each training round, training is performed first, then validation is performed, and the loss value is calculated.

[0052] Example 2

[0053] To implement the AI-based early Alzheimer's disease intelligent auxiliary diagnosis method steps in the above embodiment 1, please refer to Figure 2 The present invention also provides a system for an AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease, which includes a login unit, a symptom and sign detection unit, a self-assessment unit, a data analysis unit, and a report feedback unit.

[0054] Specifically, the login unit is configured as follows: the patient fills in necessary personal information in the AI ​​diagnosis system to complete registration to form an ID account corresponding to the patient and automatically logs in;

[0055] The symptom and sign detection unit is configured as follows: the AI ​​diagnostic system obtains changes in the patient's facial expressions, obtains the patient's voice and tone, and monitors the patient's gait and reaction time. The symptom and sign detection unit includes a high-definition camera, a microphone, and a data preprocessing module. The high-definition camera is used to capture and monitor changes in the patient's facial expressions, as well as the patient's gait and reaction time. The microphone is used to collect the patient's voice and tone. The data preprocessing module is used to clean, denoise, and standardize the collected video and audio.

[0056] The self-assessment unit is configured as follows: the AI ​​diagnostic system issues voice prompts to guide the patient to fill in the neuropsychological scale on the interactive interface, guiding the patient to complete the self-assessment;

[0057] The data analysis unit is configured to: input the patient's facial expression changes, obtain the patient's voice intonation, monitor the patient's gait, reaction time, and fill out a neuropsychological scale into a trained machine learning model, and the machine learning model performs a comprehensive analysis of the patient;

[0058] The report feedback unit is configured as follows: the AI ​​diagnostic system provides an Alzheimer's disease risk assessment and automatically generates a diagnostic report, which is displayed to the patient and / or medical staff through an electronic screen.

[0059] In summary, the present invention combines the patient's self-assessment of symptoms, physical sign detection, Montreal Cognitive Assessment (MoCA), depression score, sleep disorder score, cognitive part of Alzheimer's Disease Assessment Scale and Clinical Dementia Rating Scale (CDR) and imaging, biomarkers, risk factors and other multi-dimensional information, combined with electronic system automated analysis, to achieve rapid diagnosis of Alzheimer's disease prevalence, reduce the workload of medical staff, help intervene in the early stages of the disease, delay the progression of the disease, improve the quality of life of patients, and provide rich data support for the study of Alzheimer's disease, and promote the exploration of disease pathogenesis and treatment methods.

[0060] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced by equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed in the present invention may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. An AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease, characterized in that: Here are the steps: S1. The patient fills in the necessary personal information in the AI ​​diagnostic system to complete the registration and form an ID account corresponding to the patient and log in automatically; S2, the AI ​​diagnostic system acquires changes in the patient's facial expressions, acquires the patient's voice tone, and monitors the patient's gait and reaction time; S3, the AI ​​diagnostic system issues voice prompts to guide patients to fill in the neuropsychological scale on the interactive interface and guide patients to complete the self-assessment; S4. The patient's facial expression changes, voice tone, gait, reaction time, and completed neuropsychological scale are input into the trained machine learning model. The machine learning model conducts a comprehensive analysis of the patient, provides an Alzheimer's disease risk assessment, and automatically generates a diagnostic report, which is displayed to the patient and / or medical staff on an electronic screen.

2. According to claim 1, an AI-based intelligent auxiliary diagnosis method for early Alzheimer's disease is characterized in that: In step S2, the specific steps of the AI ​​diagnosis system obtaining changes in the patient's facial expression, obtaining the patient's voice tone, and monitoring the patient's gait and reaction time are as follows: The AI ​​diagnostic system monitors the patient’s facial expression changes, gait, and reaction time through video capture; The AI ​​diagnostic system collects the patient's voice and intonation through voice sensors, and then cleans, denoises and standardizes the collected video and audio.

3. The AI-based intelligent assisted diagnosis method for early Alzheimer's disease according to claim 1, characterized in that: In step S3, the neuropsychological scales include the MoCA scale, the depression rating scale, the sleep disorder rating scale, and the ADAS-Cog scale.

4. The AI-based intelligent assisted diagnosis method for early Alzheimer's disease according to claim 1, characterized in that: In step S3, the machine learning model is specifically trained as follows: First, we used crawlers to obtain a large number of datasets of patients’ facial expressions, gait, reaction time, voice intonation, and neuropsychological scales; Divide the dataset into training set, validation set and test set, with the proportions of 70%, 15% and 15% respectively; Then, the model is trained, and the Adam optimizer with a learning rate of 0.0001 is used during the training process to minimize the Huber loss function, which is defined as: where a ij is the true value, is the predicted value, δ is the threshold for switching from quadratic loss to linear loss. In each training round, training is performed first, then validation is performed, and the loss value is calculated.

5. A system for implementing the AI-based intelligent assisted diagnosis method for early Alzheimer's disease as described in any one of claims 1 to 4, characterized in that: include: The login unit is configured to: the patient fills in necessary personal information in the AI ​​diagnosis system to complete registration to form an ID account corresponding to the patient and automatically log in; The symptom and sign detection unit is configured as follows: the AI ​​diagnostic system acquires changes in the patient's facial expression, acquires the patient's voice tone, and monitors the patient's gait and reaction time; The self-assessment unit is configured as follows: the AI ​​diagnostic system issues voice prompts to guide the patient to fill in the neuropsychological scale on the interactive interface, guiding the patient to complete the self-assessment; The data analysis unit is configured to: input the changes in the patient's facial expression, obtain the patient's voice tone, monitor the patient's gait, reaction time, and the completed neuropsychological scale into the trained machine learning model, and the machine learning model performs a comprehensive analysis of the patient; The report feedback unit is configured as follows: the AI ​​diagnostic system provides an Alzheimer's disease risk assessment and automatically generates a diagnostic report, which is displayed to the patient and / or medical staff through an electronic screen.

6. The AI-based early Alzheimer's disease intelligent auxiliary diagnosis system according to claim 5 is characterized in that: The symptom and sign detection unit includes a high-definition camera, a microphone and a data preprocessing module. The high-definition camera is used to capture and monitor the patient's facial expression changes, gait and reaction time. The microphone is used to collect the patient's voice intonation. The data preprocessing module is used to clean, denoise and standardize the collected video and audio.

Images