CN-122006152-A - Depression patient emotion analysis platform based on transcranial focusing ultrasonic therapy

CN122006152ACN 122006152 ACN122006152 ACN 122006152ACN-122006152-A

Abstract

The invention discloses a depression patient emotion analysis platform based on transcranial focusing ultrasonic treatment, which relates to the technical field of medical intelligent signal processing and comprises a data acquisition module for synchronously acquiring physiological signals, behavior data and subjective emotion feedback of a patient, a treatment parameter synchronization module for acquiring ultrasonic treatment core parameters, an emotion feature extraction module for preprocessing the data and extracting high-dimensional emotion feature vectors, an intelligent emotion analysis module for evaluating depression degree and treatment response through a fusion algorithm model, a feedback regulation and control module for pushing parameter adjustment advice, intervention prompt and emotion regulation guidance to construct a closed-loop regulation and control system, a data storage and management module for storing data in a classified mode, and a man-machine interaction module for providing data display and operation functions and adapting requirements of a medical care end and a patient end. The invention integrates multi-modal data and intelligent analysis, accurately evaluates the emotional state and treatment response of a patient, and promotes standardized and accurate development of depression treatment through multidisciplinary collaboration and data security guarantee.

Inventors

YUAN BO
CAI XIN
LONG XIAOJING
SUN WEI
ZHENG XIAODAN

Assignees

深圳市第四人民医院(深圳市萨米医疗中心)

Dates

Publication Date: 20260512
Application Date: 20260206

Claims (10)

1. A depression patient emotion analysis platform based on transcranial focused ultrasound therapy, comprising the following modules: The data acquisition module is provided with a high-precision electrocardiosignal, an electroencephalogram acquisition electrode, a piezoelectric sensor, a high-definition facial image camera, a high-fidelity voice microphone and a digital self-evaluation scale interaction unit, synchronously acquires physiological signals, behavior data and subjective emotion feedback during treatment of a patient, and transmits the physiological signals, the behavior data and subjective emotion feedback to the core processing module after acquisition, signal amplification, filtering and noise reduction treatment; The treatment parameter synchronization module adopts an industrial Ethernet and 5G double-link communication protocol to establish stable connection with transcranial focusing ultrasonic treatment equipment and acquire core parameters in real time; The emotion feature extraction module is used for carrying out systematic preprocessing on the original data, and extracting high-dimensional structured emotion feature vectors by fusing multi-source data; The intelligent emotion analysis module is used for carrying a fusion algorithm model of a random forest, a convolutional neural network and a long-term and short-term memory network, inputting emotion feature vectors and synchronous treatment parameters, dynamically evaluating the condition of a patient and generating a quantized emotion score and a qualitative analysis report; The feedback regulation and control module receives the intelligent analysis result, sends parameter adjustment advice to the treatment equipment, pushes an evaluation report and an intervention prompt to the medical care end, and sends emotion regulation guidance to the patient end; The data storage and management module adopts a local encryption storage and cloud distributed server dual architecture, stores various data in a classified manner, and supports multidimensional inquiry, export and traceability; And the man-machine interaction module is used for configuring an interaction interface between the medical care end professional workstation and the patient end mobile terminal, displaying related data and information in real time, supporting the operation adjustment of medical staff, submitting data by a patient and viewing guidance content.
2. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising an emotion state evaluation module integrating the multidimensional features output by the emotion feature extraction module, constructing a comprehensive evaluation model by a multi-factor weighted fusion algorithm, and formulating the comprehensive evaluation model by a formula Calculating a patient's real-time mood score, wherein For the comprehensive score of the emotion, As a result of the physiological signal feature weight coefficients, For the normalization of the score to the physiological signal, For the characteristic weighting coefficients of the behavioural data, For the performance data to be normalized for score, For the characteristic weight coefficient of the main scale, The score is normalized for the subjective scale, In order to treat the response characteristic weighting coefficients, The scores were normalized for treatment response.
3. The system of claim 1, further comprising a therapy response prediction and parameter optimization module for predicting the effect of a patient's subsequent therapy response by a time series trend analysis algorithm based on historical evaluation data and therapy parameter records of the intelligent emotion analysis module, and a formula Calculating an optimal adjustment of intensity of ultrasound therapy, wherein For the intensity adjustment of the ultrasonic treatment, In order to respond to the sensitivity coefficient of the therapy, The difference of emotion comprehensive scores for two adjacent treatment periods is , For the duration of the current treatment cycle, And generating personalized treatment parameter optimization suggestions for the current treatment pulse repetition frequency according to the adjustment quantity.
4. The depression patient emotion analysis platform based on transcranial focusing ultrasonic treatment according to claim 1, further comprising a remote monitoring and abnormality early warning module, wherein the module completes real-time data transmission between a medical care end and a patient treatment scene through a 5G ultra-wideband communication technology, sets three classification early warning thresholds of emotion fluctuation, physiological abnormality and treatment response, immediately triggers a classification early warning mechanism when detecting that the emotion score fluctuation of a patient exceeds a set range, abnormal peaks occur in physiological signals and emotion score rising of two continuous treatment periods is insufficient to set values, synchronously records abnormality occurrence time, related data and early warning grades, and supports medical staff to remotely check real-time data, give emergency intervention instructions and adjust treatment schemes.
5. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising a model self-learning optimization module, wherein the model self-learning optimization module continuously collects multi-center clinical treatment data, emotion analysis results and treatment effect feedback, a large-scale labeling data set is constructed, a fusion algorithm model of the intelligent emotion analysis module is subjected to iterative training through an incremental learning algorithm, and features of the dynamic optimization model extract weights, classification decision boundaries and evaluation parameters, and meanwhile personalized analysis submodels are generated according to patient ages, disease courses, disease severity, treatment targets and basic health condition differences.
6. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising a multi-mode data fusion enhancement module, wherein the multi-mode data fusion enhancement module adopts a concentration mechanism and feature level fusion algorithm to deeply fuse physiological signals, behavior data, subjective scale data and treatment parameter data, dynamically distributes information with characteristic weights which are highly relevant to emotion states, extracts specific electroencephalogram characteristics aiming at an electroencephalogram signal and ultrasonic treatment target point relevant area, and combines emotion consistency verification of facial expression and voice intonation to generate a high-dimensional emotion characteristic vector with more identification degree.
7. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising a medical care cooperative interaction module supporting authority hierarchical management of multi-disciplinary medical care teams, dividing operation authorities and data access ranges of different roles, enabling an attending doctor to view complete emotion analysis data, adjusting treatment parameter suggestions, signing intervention schemes, enabling a nurse to receive early warning information, recording clinical observations, enabling a rehabilitee to obtain emotion improvement trends, formulating personalized emotion adjustment training schemes, enabling a psychotherapist to view emotion fluctuation details, enabling operation logs of all roles to be recorded in real time and not tampered, supporting intra-team online communication, opinion cooperation, and generating a unified patient treatment and emotion management scheme.
8. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising a patient emotion adjustment guidance module that automatically matches personalized emotion adjustment schemes based on the assessment results and treatment phases of the intelligent emotion analysis module, pushes respiratory relaxation training, positive meditation guiding audio, positive psychological implication graphics-text courses for low emotion states, provides cognitive behavior adjustment skills, stress management advice, emotion diary templates for emotion fluctuation states, pushes rehabilitation progress motivation content, interest activity advice, social interaction guidance for treatment response well-being states, scheme pushing combines patient daily work and rest time reasonable arrangement, and records patient adjustment schemes.
9. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising a data security and privacy protection module, wherein the data security and privacy protection module adopts an end-to-end encryption algorithm to carry out whole-course encryption transmission and storage on sensitive information, the tracing and tamper resistance of key medical data are completed through a blockchain technology, a biological recognition and dynamic password dual verification mechanism is adopted, the related standard of medical data privacy protection is strictly followed, the desensitization treatment is carried out on patient identity information, a data access audit log is set, and the data operation is comprehensively recorded.
10. The depression patient emotion analysis platform based on transcranial focused ultrasound therapy according to claim 1, further comprising a multi-center data sharing and model iteration module, wherein the module adopts standardized data formats and interface protocols, supports platform data interconnection and secure sharing of different medical institutions, establishes a multi-center combined scientific research database, performs data summarization analysis after removing patient privacy information, performs cross-institution training on an intelligent emotion analysis model by combining multi-center data on the premise of not sharing original sensitive data through a federal learning algorithm, and integrates clinical experience and data resources of different regions and different diagnosis and treatment levels.

Description

Depression patient emotion analysis platform based on transcranial focusing ultrasonic therapy Technical Field The invention relates to the technical field of medical intelligent signal processing, in particular to a depression patient emotion analysis platform based on transcranial focusing ultrasonic treatment. Background Depression is regarded as a global high-rise mental and psychological disease, seriously affects physical and psychological health and social functions of patients, and transcranial focused ultrasound treatment has become one of clinically important treatment means due to the advantages of noninvasive property, strong targeting property and the like. The emotional state is used as a core index for evaluating the treatment effect and adjusting the treatment scheme, and the accurate monitoring and dynamic analysis of the emotional state are important to improving the treatment effectiveness. However, in the current depression treatment, emotion assessment depends on subjective observation of medical staff and periodic self-assessment feedback of patients, so that problems of assessment lag, strong subjectivity, data fragmentation and the like exist, fine fluctuation of emotion in the treatment process can not be captured in real time, and correlation between emotion change and treatment parameters is difficult to establish, so that the treatment scheme adjustment lacks timely and objective data support. The prior emotion analysis related technology has obvious limitation and is difficult to adapt to the clinical requirement of transcranial focusing ultrasonic treatment. In the aspect of data acquisition, the method is limited to single-dimension data, or only physiological signals are acquired, or only subjective scales are relied on, and the synchronous acquisition and integration of behavior data and treatment parameters are lacked, so that the emotion assessment dimension is on one side, and the true emotion state of a patient cannot be comprehensively reflected. In the aspect of analysis model, a single algorithm is mostly adopted, complementarity of multi-mode data is not fully fused, targeted optimization of a treatment scene is lacking, emotion assessment accuracy is insufficient, and treatment response effect and emotion improvement trend are difficult to predict. In the aspect of a feedback mechanism, a complete closed loop of acquisition-analysis-feedback-intervention is not constructed, an analysis result is only used as a reference, and treatment parameter adjustment and personalized intervention cannot be directly guided, so that treatment and emotion management are disjointed. Insufficient intelligence, cooperativity and data security further restrict clinical application. The existing system lacks self-learning capability, and can not adapt to the personalized demand capability of patients with different ages, disease course and disease severity according to the multi-center clinical data iterative optimization model. The multidisciplinary medical team cooperation mechanism is missing, and different roles are difficult to efficiently share data and cooperatively formulate a scheme, so that the timeliness and the integrity of intervention are affected. The medical data privacy protection measures are imperfect, the sensitive information of the patient is exposed and abused, and a trans-institution data sharing mechanism is lacked, so that the high-quality clinical resource pushing technology iteration is difficult to integrate. These problems lead to the difficulty in improving the accuracy and individuation level of transcranial focused ultrasound treatment, and restrict the further optimization of the treatment effect of depression. Disclosure of Invention The invention provides a depression patient emotion analysis platform based on transcranial focused ultrasound therapy, which aims to solve the problems in the prior art. In order to achieve the aim, the invention adopts the following technical scheme that the emotion analysis platform for the depression patient based on transcranial focusing ultrasonic therapy comprises the following modules: The data acquisition module is provided with a high-precision electrocardiosignal, an electroencephalogram acquisition electrode, a piezoelectric sensor, a high-definition facial image camera, a high-fidelity voice microphone and a digital self-evaluation scale interaction unit, synchronously acquires physiological signals, behavior data and subjective emotion feedback during treatment of a patient, and transmits the physiological signals, the behavior data and subjective emotion feedback to the core processing module after acquisition, signal amplification, filtering and noise reduction treatment; The treatment parameter synchronization module adopts an industrial Ethernet and 5G double-link communication protocol to establish stable connection with transcranial focusing ultrasonic treatment equipment and acquire core parameters in