Psychiatry anti-suicide risk early warning method and system

Abstract

The invention discloses a psychiatry anti-suicide risk early warning method and system, and relates to the technical field of data encryption, and the method comprises the steps: deploying a sensor at a key part of the head of a patient, collecting the concentration data of oxyhemoglobin of the patient, collecting the medication data of the patient, and carrying out the preprocessing; extracting psychological state characteristics from the preprocessed oxyhemoglobin concentration data, and extracting medicine characteristics from the patient medication data; establishing a psychological and physiological prediction model through the psychological state features and the drug features to obtain a psychological and physiological prediction value of the patient; according to the psychological and physiological predicted value of the patient, suicide risk assessment is carried out on the patient through the experience of psychiatric experts, and risk early warning is carried out according to the risk assessment result; according to the risk early warning result, personalized intervention suggestions for the patient are given. According to the method, oxyhemoglobin concentration data and medication data of a patient are comprehensively analyzed, a psychological and physiological prediction model is constructed to evaluate the suicide risk, and personalized intervention suggestions are given accordingly.

Description

Technical Field

[0001] The present invention relates to the field of medical health, and in particular to a psychiatric suicide risk warning method and system. Background Art

[0002] Suicide is a serious public health problem, and its early identification and intervention are crucial. With the development of neuroscience technology, it has become possible to assess an individual's mental state by monitoring brain activity. Near-infrared spectroscopy (NIRS) is increasingly used in the field of mental health due to its non-invasiveness and portability. NIRS can measure changes in blood flow in the cerebral cortex area, thereby reflecting the functional state of the brain, which is of great significance for understanding emotional states and assessing suicide risks.

[0003] In addition, the patient's medication record is also an important factor in assessing suicide risk. Traditional suicide risk assessment mainly relies on the experience and subjective judgment of clinicians, which may lead to inconsistent assessment results with reality. Therefore, it is particularly important to develop a suicide risk warning system based on objective physiological data and technical means. Summary of the invention

[0004] In view of the above existing problems, the present invention provides a psychiatric suicide risk warning method and system, which solves the problem that the suicide risk assessment results of patients by psychiatric clinicians are inconsistent with the reality.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides a psychiatric suicide risk warning method, which includes deploying sensors at key parts of a patient's head to collect the patient's oxygenated hemoglobin concentration data, and at the same time collect the patient's medication data, and perform preprocessing; Extract psychological state features from preprocessed oxyhemoglobin concentration data and extract drug features from patient medication data; Through the psychological state characteristics and drug characteristics, a psychological and physiological prediction model is established to obtain the patient's psychological and physiological prediction value; Based on the patient's psychological and physiological prediction values ​​and the experience of psychiatric experts, the patient's suicide risk is assessed and risk warnings are issued based on the risk assessment results; Based on the risk warning results, personalized intervention recommendations are given to patients.

[0006] As a preferred embodiment of the psychiatric suicide risk warning method of the present invention, the key parts of the patient's head are the frontopolar region, frontal lobe and temporal lobe; the sensor is a near-infrared spectral sensor; the patient's medication data includes medication dosage, medication frequency and medication time; and the preprocessing includes time synchronization, medication dosage standardization, data interpolation and data smoothing.

[0007] As a preferred solution of the psychiatric suicide risk warning method of the present invention, wherein: the psychological state characteristics are extracted from the pre-processed oxyhemoglobin concentration data, and the psychological state characteristics are composed of peak characteristics and skewness characteristics. The specific steps are as follows: Extracting spike features from oxygenated hemoglobin concentration data; Extract skewness features from oxyhemoglobin concentration data.

[0008] As a preferred solution of the psychiatric suicide risk warning method of the present invention, drug features are extracted from medication data.

[0009] As a preferred embodiment of the psychiatric suicide risk warning method of the present invention, a psychological and physiological prediction model is established through psychological state characteristics and drug characteristics to obtain the patient's psychological and physiological prediction results.

[0010] As a preferred embodiment of the psychiatric suicide risk warning method of the present invention, the suicide risk of the patient is assessed based on the patient's psychological and physiological prediction values ​​and the experience of psychiatric experts, and a risk warning is performed based on the risk assessment results. The specific steps are as follows: Based on the experience of psychiatric experts, two suicide risk thresholds were determined, namely low suicide risk threshold and medium suicide risk threshold; When the patient's psychological and physiological prediction value is less than or equal to the low suicide risk threshold, it indicates that the patient is at a low suicide risk state, and no warning is issued at this time; When the patient's psychological and physiological prediction value is greater than the low suicide risk threshold and less than or equal to the medium suicide risk threshold, it indicates that the patient is at medium suicide risk, and a first-level warning is issued at this time; When the patient's psychological and physiological prediction value is greater than the medium suicide risk threshold, it indicates that the patient is at high risk of suicide, and a second-level warning is issued.

[0011] As a preferred solution of the psychiatric suicide risk warning method of the present invention, wherein: according to the risk warning result, personalized intervention suggestions are given to the patient, and the specific steps are as follows: For patients with a low risk of suicide, only regular follow-up and routine treatment are required; Patients at risk of suicide at level 1 need increased psychological support and monitoring frequency; Patients at level 2 high risk of suicide require urgent hospitalization and intensive monitoring; Experts adjust assessment criteria and intervention plans based on the patient's actual situation.

[0012] In a second aspect, the present invention provides an automatic vending system based on a pressure sensor, comprising a data collection and preprocessing module, a psychophysiological feature extraction module, a psychophysiological prediction model construction module, a suicide risk assessment module, and a personalized intervention suggestion module; The data collection and preprocessing module is used to deploy sensors at key parts of the patient's head to collect the patient's oxygenated hemoglobin concentration data, and at the same time collect the patient's medication data and perform preprocessing; The psychological and physiological feature extraction module is used to extract psychological state features and drug features from the pre-processed oxyhemoglobin concentration data and the patient medication data, and fuse them to form a psychological and physiological feature set; The psychophysiological prediction model building module is used to establish a psychophysiological prediction model through a psychophysiological feature set, using a convolutional neural network and a recurrent neural network, to obtain a patient's psychophysiological prediction value; The suicide risk assessment module is used to assess the patient's suicide risk based on the patient's psychological and physiological prediction values ​​and through the experience of psychiatric experts, and to issue risk warnings based on the risk assessment results; The personalized intervention suggestion module is used to give personalized intervention suggestions to patients according to the risk warning results.

[0013] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the psychiatric suicide risk warning method as described in the first aspect of the present invention is implemented.

[0014] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the psychiatric suicide risk warning method as described in the first aspect of the present invention is implemented.

[0015] The beneficial effects of the present invention are as follows: by using near-infrared spectroscopy technology to monitor brain activity, extract psychological state characteristics, and combine with drug use pattern analysis, the patient's suicide risk can be more comprehensively assessed, and the objectivity and accuracy of the assessment can be improved; based on the psychophysiological prediction model, personalized intervention suggestions can be provided for each patient, and different intervention measures can be formulated according to the risk level, such as regular follow-up, increased psychological support, emergency hospitalization, etc., which helps to improve the treatment effect; through continuous monitoring and evaluation, early warnings can be issued in time when the risk of suicide increases, providing patients and clinicians with timely intervention opportunities and reducing the occurrence of tragedies; the automated data collection and analysis process reduces the workload of doctors, allowing them to focus more on the treatment and care of patients, thereby improving the efficiency of medical services. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 This is a flow chart of the psychiatric suicide risk warning method in Example 1.

[0018] Figure 2 This is a diagram for determining the suicide risk level of patients in Example 1. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0020] Example 1 Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a psychiatric suicide risk warning method, comprising the following steps: S1. Deploy sensors at key parts of the patient's head to collect the patient's oxygenated hemoglobin concentration data, collect the patient's medication data at the same time, and perform preprocessing, including the following steps: the key parts of the patient's head are the frontal pole area, frontal lobe and temporal lobe; the sensor is a near-infrared spectral sensor; the patient's medication data includes medication dosage, medication frequency and medication time; preprocessing includes time synchronization, medication dosage standardization, data interpolation and data smoothing; It should be noted that because these areas are closely related to emotion regulation and cognitive function, monitoring blood flow changes in these specific brain regions provides objective physiological data, which helps to more accurately assess the patient's psychological state.

[0021] Analyzing patients' medication patterns to assess their emotional state and treatment compliance can assist in suicide risk assessment, revealing patients' response to treatment and mood swings, providing an additional source of information for risk assessment.

[0022] Time synchronization ensures that different data sources are aligned in time for joint analysis; drug dosage is standardized to eliminate differences in different drug dosage units, making the data comparable; data interpolation fills in missing data points to ensure data integrity; data smoothing is used to reduce noise and improve data quality.

[0023] S2. Extracting psychological state features from the preprocessed oxygenated hemoglobin concentration data and extracting drug features from the patient medication data, including the following steps. The psychological state features are composed of peak features and skewness features. The peak features are extracted from the oxygenated hemoglobin concentration data and are expressed as: ; In the formula, is the peak eigenvalue, is the number of oxygenated hemoglobin concentration data points, is the index of the time point, For at time point The oxygenated hemoglobin concentration data value, is the mean of the oxygenated hemoglobin concentration data, is the standard deviation of the oxygenated hemoglobin concentration data; Extract the skewness feature from the oxyhemoglobin concentration data, expressed as: ; In the formula, is the skewness eigenvalue; Extract drug features from medication data, expressed as: DF =[ Do t - Do t - 1 Δ t , σ D μ D , Fr t - Fr t - 1 Δ t , ∑ i=1 n D - 1 ( Tm i+1 - Tm i ) n D - 1 ] ; In the formula, is the medication feature vector, is the index of the time point, For the current time point The dosage of For time point The dosage of the drug, is the time interval, is the standard deviation of the dosage, is the mean dosage of the drug, For the current time point The frequency of medication, For time point The frequency of medication, For the The time of medication, For the The time of medication, The number of times the medication is used.

[0024] It should be noted that peak features are extracted from the oxygenated hemoglobin concentration data to evaluate the peak degree of the oxygenated hemoglobin concentration data distribution, thereby reflecting the stability of the patient's psychological state; skewness features are extracted from the oxygenated hemoglobin concentration data to evaluate the skewness of the oxygenated hemoglobin concentration data distribution, thereby reflecting the asymmetry of the patient's psychological state. The skewness feature can capture the asymmetry of the data distribution and help identify potential psychological pressure or mood swings; medication features are extracted from the medication data to evaluate the patient's emotional state and treatment compliance by analyzing the patient's medication pattern, thereby assisting in suicide risk assessment. Medication features can reveal the patient's response to treatment and mood swings, providing an additional source of information for risk assessment.

[0025] S3. Establishing a psychological and physiological prediction model through psychological and psychological state characteristics and drug characteristics to obtain the patient's psychological and physiological prediction results includes the following steps. The psychological and physiological prediction model is expressed as: ; In the formula, is the patient's psychological and physiological prediction value, is the weight coefficient of the skewness feature relative to the peak feature, is the weight coefficient of drug characteristics relative to psychological state characteristics, is the length of the medication feature vector, Represented as an element of the medication feature vector, For the The normalization factor for each drug feature, is the integral index, is the power of the drug feature vector, which is used to adjust the influence of drug features. is the index of the discrete time point.

[0026] It should be noted that the model can take into account multiple factors, including psychological state and drug use, to provide a comprehensive risk assessment result. Quantifying the patient's psychophysiological state through mathematical models can help to more accurately identify individuals at higher risk of suicide. The model can be updated over time and as the patient's condition changes, providing a dynamic risk assessment.

[0027] S4. Based on the patient's psychological and physiological prediction values ​​and the experience of psychiatric experts, the patient's suicide risk is assessed, and risk warning is carried out according to the risk assessment results, including the following steps: based on the experience of psychiatric experts, two suicide risk thresholds are determined, namely, a low suicide risk threshold and a medium suicide risk threshold; When the patient's psychological and physiological prediction value is less than or equal to the low suicide risk threshold, it indicates that the patient is at a low suicide risk state, and no warning is issued at this time; When the patient's psychological and physiological prediction value is greater than the low suicide risk threshold and less than or equal to the medium suicide risk threshold, it indicates that the patient is at medium suicide risk, and a first-level warning is issued at this time; When the patient's psychological and physiological prediction value is greater than the medium suicide risk threshold, it indicates that the patient is at high risk of suicide, and a second-level warning is issued.

[0028] It should be noted that the psychophysiological prediction score (PSRS) is a quantitative indicator of the patient's psychophysiological state calculated by a mathematical model by integrating the characteristics of the psychological state and the characteristics of the drug. The suicide risk threshold is a numerical value set by psychiatric experts based on experience to distinguish different levels of suicide risk. By quantifying the psychophysiological state, the patient's suicide risk can be assessed more accurately, thereby improving the accuracy and reliability of the assessment. It can detect signs of increased suicide risk in a timely manner, provide early warnings for patients and clinicians, and facilitate early intervention. The psychophysiological prediction value is linked to the suicide risk level to facilitate risk warning. Through an automated risk assessment and early warning system, the workload of clinicians can be reduced and the efficiency of medical services can be improved. Combining the experience of psychiatric experts with objective data can help enhance the trust of patients and their families in the assessment results.

[0029] S5. Based on the risk warning results, give personalized intervention recommendations to patients, including the following steps: For patients with low suicide risk, only regular follow-up and routine treatment are required; for patients in the first-level warning suicide risk state, psychological support and monitoring frequency need to be increased; for patients in the second-level warning suicide risk state, emergency hospitalization and intensive monitoring are required; experts adjust the assessment criteria and intervention plan according to the actual situation of the patient.

[0030] It should be noted that for patients with a low risk of suicide, regular follow-up is to arrange outpatient follow-up regularly, such as once every two weeks or once a month. The specific frequency can be determined according to the specific situation of the patient and the advice of the psychiatrist. During the follow-up, the patient's symptom changes, quality of life, and social function should be evaluated. Conventional treatment is to continue the existing treatment plan, including drug therapy and psychotherapy. Adjust the treatment plan regularly according to the patient's needs and progress. Encourage patients to maintain good communication with their families and seek support from family members. Provide family education materials to help family members understand the patient's condition and how to provide appropriate support. Educate patients on how to identify and cope with emotional fluctuations. Provide self-management skills training, such as relaxation training, mindfulness meditation, etc.

[0031] For patients at risk of suicide at level 1, arrange more frequent counseling or psychotherapy, such as once a week or more frequently as needed. Provide a crisis hotline number and encourage patients to seek help when they encounter difficulties. Increase the frequency of face-to-face or remote follow-up, such as once a week. Use mobile applications or online platforms for daily self-monitoring to track mood changes. Recommend patients to join support groups or community organizations to communicate with others who have similar experiences. Encourage patients to participate in activities that help regulate emotions, such as art therapy, exercise, etc. Assess whether there are risk factors in the home that may trigger suicidal behavior and take appropriate measures to reduce the risk. Discuss with family members how to create an environment conducive to recovery.

[0032] For patients at a high risk of suicide at the second-level warning level, emergency hospitalization is when the assessment shows that the patient has obvious suicidal tendencies, and hospitalization should be arranged immediately. A comprehensive physical and mental health assessment should be conducted during hospitalization. Intensive monitoring is to closely monitor the patient's behavior and emotional changes in the hospital environment. Implement 24-hour uninterrupted monitoring to ensure the safety of the patient. Crisis intervention is emergency support provided by a professional crisis intervention team, including psychotherapy and drug therapy. Provide a crisis intervention manual or guide to guide patients on how to respond to emergencies. Family participation is to invite family members to participate in the treatment process to understand the patient's needs and progress. Provide family support services to help family members master the skills to support patients. Discharge planning is to develop a detailed discharge plan, including follow-up treatment arrangements, family support, and community resource links. Conduct a final comprehensive assessment before discharge to ensure that the patient has sufficient self-care ability.

[0033] Regarding the experts’ adjustment of assessment criteria and intervention plans, regular review is to regularly review the patient’s progress and the effectiveness of the treatment plan, and adjust the assessment criteria and intervention measures according to the patient’s specific situation; personalized intervention is to develop a personalized intervention plan based on the individual differences of the patient. Including treatment methods and preventive measures for specific symptoms; interdisciplinary collaboration is to cooperate with psychiatrists, psychological counselors, social workers and other medical professionals to ensure that patients receive comprehensive support, hold regular case meetings to discuss the patient’s progress and next action plan; patient participation is to encourage patients to actively participate in the treatment process, express their feelings and needs, support patients in setting personal goals, and assist in achieving these goals.

[0034] This embodiment also provides a psychiatric suicide risk warning system, including a data collection and preprocessing module, a psychological and physiological feature extraction module, a psychological and physiological prediction model construction module, a suicide risk assessment module and a personalized intervention suggestion module; The data collection and preprocessing module is used to deploy sensors at key parts of the patient's head to collect the patient's oxygenated hemoglobin concentration data, as well as the patient's medication data, and perform preprocessing; The psychological and physiological feature extraction module is used to extract psychological state features and drug features from the pre-processed oxyhemoglobin concentration data and patient medication data, and fuse them to form a psychological and physiological feature set; The psychophysiological prediction model building module is used to establish a psychophysiological prediction model through a psychophysiological feature set, using a convolutional neural network and a recurrent neural network, and obtain the patient's psychophysiological prediction value; The suicide risk assessment module is used to assess the patient's suicide risk based on the patient's psychological and physiological prediction values ​​and the experience of psychiatric experts, and to issue risk warnings based on the risk assessment results; The personalized intervention recommendation module is used to give personalized intervention recommendations to patients based on risk warning results.

[0035] This embodiment also provides a computer device, which is applicable to the case of a psychiatric suicide risk warning method, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the psychiatric suicide risk warning method proposed in the above embodiment.

[0036] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0037] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for realizing the psychiatric suicide risk warning method proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0038] In summary, the present invention deploys near-infrared spectroscopy sensors at key parts of the patient's head to collect oxygenated hemoglobin concentration data and medication data, and performs preprocessing to extract psychological state characteristics and drug characteristics from these data. Using these characteristics, the patient's psychological and physiological prediction value "PSRS" is calculated by constructing a psychological and physiological prediction model, and the suicide risk threshold is determined in combination with the experience of psychiatric experts to conduct suicide risk assessment and early warning. According to the risk warning results, personalized intervention suggestions are provided to patients, including regular follow-up, increased psychological support and monitoring frequency, emergency hospitalization and enhanced monitoring. The present invention improves the accuracy and timeliness of suicide risk assessment, helps early intervention, and reduces the occurrence of suicide incidents.

[0039] Example 2 Referring to Table 1, which is the second embodiment of the present invention, in order to further verify the advancement of the present invention, experimental simulation data of a psychiatric suicide risk warning method are provided.

[0040] In order to verify the effectiveness of the technical solution of the present invention, an experiment was conducted involving 10 subjects who were initially diagnosed as having suicide risk. Near-infrared spectroscopy sensors were deployed in key parts of each subject's head (frontopolar region, frontal lobe and temporal lobe) to collect oxygenated hemoglobin concentration data. In addition, the subjects' medication data, including medication dosage, medication frequency and medication time, were also collected. These data were preprocessed, including time synchronization, medication dosage standardization, data interpolation and data smoothing to ensure data quality.

[0041] Psychological state features, including peak features ("Kur") and skewness features ("Ske"), were extracted from the preprocessed oxyhemoglobin concentration data. Drug features ("DF") were extracted from the medication data, which included the rate of change of medication dosage, the ratio of standard deviation of medication dosage to mean, the rate of change of medication frequency, and the average medication interval.

[0042] Next, based on the extracted psychological state characteristics and drug characteristics, a psychophysiological prediction model was constructed, which calculated the subject's psychophysiological prediction value (PSRS). The model took into account the influence of peak characteristics, skewness characteristics, and drug characteristics, and obtained the psychophysiological prediction value of each subject through a series of mathematical operations.

[0043] Finally, based on the psychophysiological prediction values, two suicide risk thresholds were set through the experience of psychiatric experts: low suicide risk threshold (LSRT) and medium suicide risk threshold (MSRT). According to these thresholds, the subjects were divided into different risk levels and corresponding personalized intervention recommendations were given.

[0044] The details are shown in Table 1: Table 1 Suicide risk assessment experiment record Test No. Peak feature Kur Skewness feature Ske Medication characteristics "DF" Psychophysiological predictive value PSRS Risk Level Interventions 01 0.25 -0.03 [0.05, 0.02, 0.1,0.03] 0.25 Low Regular follow-up 02 0.32 -0.15 [0.1, 0.03, 0.15,0.05] 0.35 middle Increase psychological support 03 0.40 0.2 [0.2, 0.04, 0.25,0.08] 0.45 high Emergency hospitalization 04 0.20 -0.02 [0.03, 0.01, 0.12,0.02] 0.23 Low Regular follow-up 05 0.35 -0.1 [0.12, 0.04, 0.18,0.06] 0.38 middle Increase psychological support 06 0.42 0.15 [0.25, 0.05, 0.3,0.1] 0.48 high Emergency hospitalization 07 0.22 -0.04 [0.04, 0.02, 0.13,0.03] 0.24 Low Regular follow-up 08 0.33 -0.12 [0.11, 0.03, 0.16,0.05] 0.36 middle Increase psychological support 09 0.41 0.18 [0.22, 0.04, 0.28,0.07] 0.47 high Emergency hospitalization 10 0.28 -0.05 [0.06, 0.02, 0.14,0.04] 0.30 Low Regular follow-up As can be seen from the table above, the psychophysiological prediction values ​​of subjects 03, 06 and 09 are relatively high, 0.45, 0.48 and 0.47 respectively, which indicates that they face a higher risk of suicide. In contrast, the psychophysiological prediction values ​​of subjects 01, 04, 07 and 10 are relatively low, ranging from 0.23 to 0.25, which means that their suicide risk is relatively low.

[0045] By analyzing the medication characteristics "DF", we can see that the medication dosage change rates of subjects 03, 06, and 09 are high, indicating that their medication dosage changes in a short period of time are large, which may be caused by poor treatment compliance or mood swings. In addition, their average medication interval is short, reflecting a high frequency of recent medication use.

[0046] Through the method of the present invention, we can clearly see that the peak feature and skewness feature are directly related to the patient's suicide risk. A higher peak feature value indicates that the hemoglobin concentration distribution is more concentrated, which may reflect the poor stability of the patient's psychological state; a larger skewness feature value indicates the asymmetry of the data distribution, suggesting that there may be significant emotional fluctuations or psychological pressure.

[0047] In addition, medication characteristics also have an important impact on suicide risk assessment. From the data, it can be seen that those subjects who show a larger rate of change in medication dosage and a shorter average medication interval tend to have higher psychophysiological prediction values, which indicates that monitoring medication patterns can effectively assess patients' treatment compliance and emotional state.

[0048] The present invention combines psychological state characteristics and drug characteristics to construct a psychological and physiological prediction model to assess suicide risk. Compared with traditional methods that rely solely on subjective judgment or a single indicator, it can provide more objective and comprehensive risk assessment results. This method not only helps to promptly identify signs of increased suicide risk, but also provides patients with more targeted personalized intervention suggestions, thereby effectively preventing the occurrence of suicidal behavior.

[0049] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A psychiatric suicide risk warning method, characterized by: include, Sensors are deployed at key locations on the patient’s head to collect data on the patient’s oxygenated hemoglobin concentration, as well as medication data, and pre-process it; Extract psychological state features from preprocessed oxyhemoglobin concentration data and extract drug features from patient medication data; Through the psychological state characteristics and drug characteristics, a psychological and physiological prediction model is established to obtain the patient's psychological and physiological prediction value; Based on the patient's psychological and physiological prediction values ​​and the experience of psychiatric experts, the patient's suicide risk is assessed and risk warnings are issued based on the risk assessment results; Based on the risk warning results, personalized intervention recommendations are given to patients.

2. The psychiatric suicide risk warning method according to claim 1, characterized in that: The key parts of the patient's head are the frontopolar region, frontal lobe and temporal lobe; the sensor is a near-infrared spectral sensor; the patient's medication data includes medication dosage, medication frequency and medication time; the preprocessing includes time synchronization, medication dosage standardization, data interpolation and data smoothing.

3. The psychiatric suicide risk warning method according to claim 2, characterized in that: The psychological state features are extracted from the preprocessed oxygenated hemoglobin concentration data. The psychological state features are composed of peak features and skewness features. The specific steps are as follows: Extracting spike features from oxygenated hemoglobin concentration data; Extract skewness features from oxyhemoglobin concentration data.

4. The psychiatric suicide risk warning method according to claim 3, characterized in that: Extract drug features from medication data.

5. The psychiatric suicide risk warning method according to claim 4, characterized in that: Through the characteristics of psychological state and drug characteristics, a psychological and physiological prediction model is established to obtain the psychological and physiological prediction results of patients.

6. The psychiatric suicide risk warning method according to claim 5, characterized in that: Based on the patient's psychological and physiological prediction values ​​and the experience of psychiatric experts, the patient's suicide risk is assessed and risk warning is issued based on the risk assessment results. The specific steps are as follows: Based on the experience of psychiatric experts, two suicide risk thresholds were determined, namely low suicide risk threshold and medium suicide risk threshold; When the patient's psychological and physiological prediction value is less than or equal to the low suicide risk threshold, it indicates that the patient is at a low suicide risk state, and no warning is issued at this time; When the patient's psychological and physiological prediction value is greater than the low suicide risk threshold and less than or equal to the medium suicide risk threshold, it indicates that the patient is at medium suicide risk, and a first-level warning is issued at this time; When the patient's psychological and physiological prediction value is greater than the medium suicide risk threshold, it indicates that the patient is at high risk of suicide, and a second-level warning is issued.

7. The psychiatric suicide risk warning method according to claim 6, characterized in that: According to the risk warning results, personalized intervention suggestions are given to patients. The specific steps are as follows: For patients with a low risk of suicide, only regular follow-up and routine treatment are required; Patients at risk of suicide at level 1 need increased psychological support and monitoring frequency. Patients at level 2 high risk of suicide require urgent hospitalization and intensive monitoring; Experts adjust assessment criteria and intervention plans based on the patient's actual situation.

8. A suicide risk warning system, based on the psychiatric suicide risk warning system according to any one of claims 1 to 6, characterized in that: It includes data collection and preprocessing module, psychological and physiological feature extraction module, psychological and physiological prediction model construction module, suicide risk assessment module and personalized intervention suggestion module; The data collection and preprocessing module is used to deploy sensors at key parts of the patient's head to collect the patient's oxygenated hemoglobin concentration data, and at the same time collect the patient's medication data and perform preprocessing; The psychological and physiological feature extraction module is used to extract psychological state features and drug features from the pre-processed oxyhemoglobin concentration data and the patient medication data, and fuse them to form a psychological and physiological feature set; The psychophysiological prediction model building module is used to establish a psychophysiological prediction model through a psychophysiological feature set, using a convolutional neural network and a recurrent neural network, to obtain a patient's psychophysiological prediction value; The suicide risk assessment module is used to conduct suicide risk assessment on patients based on the patient's psychological and physiological prediction values ​​and through the experience of psychiatric experts, and to issue risk warnings based on the risk assessment results; The personalized intervention suggestion module is used to give personalized intervention suggestions to patients according to the risk warning results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the psychiatric suicide risk warning method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the psychiatric suicide risk warning method according to any one of claims 1 to 6 are implemented.

Images