An intervention method for insomnia based on artificial intelligence and cognitive behavior

Abstract

The invention relates to the technical field of insomnia analysis intervention, and discloses an insomnia disorder intervention method based on artificial intelligence and cognitive behavior, comprising: obtaining the respiratory frequency of a target person, forming sleep analysis data, forming insomnia analysis data, constructing an insomnia knowledge graph, forming comparison data, forming allocation data, and pushing a diagnosis and treatment doctor to the target person. The insomnia disorder intervention method based on artificial intelligence and cognitive behavior judges and determines whether the patient has an insomnia problem in his sleep according to the patient's respiratory frequency, and intervenes in time, judges and determines whether there are external factors in the patient's non-sleep period according to the patient's heart rate, causing the patient's sleep to be disturbed, and judges and determines the number of days that the external factors affect the patient's sleep according to the patient's historical heart rate data, so that the patient can receive corresponding intervention treatment in time at the early stage of insomnia to prevent more serious health problems.

Description

Technical Field

[0001] The present invention relates to the technical field of insomnia analysis and intervention, and in particular to an insomnia disorder intervention method based on artificial intelligence and cognitive behavior. Background Art

[0002] Insomnia is a common sleep disorder that often affects the daily quality of life and health of patients. Insomnia intervention methods based on artificial intelligence and cognitive behavior can help patients improve their sleep quality and reduce insomnia symptoms. Artificial intelligence technology can analyze patients' sleep patterns and sleep habits to help doctors understand patients' insomnia symptoms and possible causes. By monitoring patients' sleep data and behavioral patterns, artificial intelligence systems can provide personalized insomnia intervention programs. Cognitive behavioral therapy is a commonly used insomnia intervention method. By changing patients' sleep concepts and behavioral habits, they can help them establish healthy sleep habits. Artificial intelligence systems can interact with patients, provide guidance and support for cognitive behavioral therapy, and help patients improve their sleep quality. Artificial intelligence systems can also help patients relax their emotions, reduce anxiety and tension, and promote sleep through audio relaxation and psychological advice. Through continuous monitoring and intervention, artificial intelligence systems can help patients establish healthy sleep habits and reduce the occurrence of insomnia symptoms. In summary, insomnia intervention methods based on artificial intelligence and cognitive behavior combine the advantages of technology and psychotherapy, which can help patients effectively manage insomnia problems, improve sleep quality, and improve the quality of life. With the continuous development of artificial intelligence technology, this intervention method will be increasingly valued and applied.

[0003] Existing intervention methods for insomnia disorders based on artificial intelligence and cognitive behavior cannot judge and determine whether the patient has insomnia problems based on the patient's breathing frequency and intervene in time. They cannot judge and determine whether there are external factors in the patient's non-sleep period that disturb the patient's sleep based on the patient's heart rate. They cannot judge and determine the number of days that external factors affect the patient's sleep based on the patient's historical heart rate data, so that the patient cannot receive corresponding intervention treatment in time in the early stages of insomnia, which affects the patient's health. Its practicality has certain limitations. Summary of the invention

[0004] The present invention provides an intervention method for insomnia disorder based on artificial intelligence and cognitive behavior, which is used to promote the solution of the problems raised in the background technology.

[0005] The present invention provides the following technical solution: an intervention method for insomnia based on artificial intelligence and cognitive behavior, comprising:

[0006] Get the target person's breathing rate;

[0007] According to the breathing frequency of the target person, sleep analysis data is formed through a sleep analysis strategy to determine the target person's sleep state, which includes wakefulness, light sleep, and deep sleep, so as to determine whether the target person has insomnia, and perform insomnia disorder intervention on the target person who has insomnia;

[0008] Based on the sleep analysis data, insomnia analysis data is formed through insomnia analysis strategies to push questionnaires to target persons with insomnia and extract key information data from the answers of the target persons, so as to determine the insomnia type of the target persons;

[0009] Constructing an insomnia knowledge graph to match key information data in the target person's questionnaire answers with elements in the insomnia knowledge graph, thereby determining the target person's insomnia type;

[0010] Based on the insomnia analysis data and the insomnia knowledge graph, a data comparison strategy is used to form comparison data to determine the target person's insomnia type;

[0011] Based on the comparison data, the diagnosis and treatment allocation strategy is used to form allocation data to push the diagnosis and treatment doctors to the target person, thereby providing the target person with online intervention diagnosis and treatment;

[0012] Based on the allocated data, the treating doctors are pushed to the target person to provide online intervention treatment for the target person.

[0013] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the sleep analysis strategy is specifically:

[0014] Get the initial collection time point and collection interval duration;

[0015] The initial collection time point is used as the starting time point for collecting the respiratory frequency of the target person, and a collection time point is formed at each interval, and the respiratory frequency of the target person is recorded;

[0016] A frequency set is formed according to each acquisition time point and the breathing frequency of the target person recorded at each acquisition time point;

[0017] According to each acquisition time point in the frequency set and the breathing frequency of the target person recorded at each acquisition time point, a sleep statistics graph of the target person is formed;

[0018] Obtain a first frequency data point, a second frequency data point, and a third frequency data point in a target person's sleep statistics graph;

[0019] Extracting the starting acquisition time points corresponding to the first frequency data point, the second frequency data point and the third frequency data point respectively, and defining them as the first starting time point, the second starting time point and the third starting time point respectively;

[0020] Extracting the termination collection time points corresponding to the first frequency data point, the second frequency data point and the third frequency data point respectively, and defining them as the first termination time point, the second termination time point and the third termination time point respectively;

[0021] Calculate the first sleep duration, first sleep duration = (second starting time point - first starting time point) + (first ending time point - second ending time point);

[0022] Calculate the second sleep duration, the second sleep duration = (third starting time point - second starting time point) + (second ending time point - third ending time point);

[0023] Calculate the third sleep duration, where the third sleep duration = the third end time point - the third start time point;

[0024] Then the total sleep time = first sleep time + second sleep time + third sleep time;

[0025] Execute the sleep determination strategy.

[0026] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the sleep determination strategy is specifically:

[0027] Obtaining a light sleep ratio determination value and a deep sleep ratio determination value;

[0028] Calculate the target person's light sleep ratio, light sleep ratio = second sleep duration ÷ total sleep duration;

[0029] Calculate the target person's deep sleep ratio, deep sleep ratio = third sleep duration ÷ total sleep duration;

[0030] If the light sleep ratio ≥ the light sleep ratio judgment value and the deep sleep ratio ≥ the deep sleep ratio judgment value, it means that the target person has enough sleep;

[0031] If the light sleep ratio is less than the light sleep ratio judgment value or the deep sleep ratio is less than the deep sleep ratio judgment value, it means that the target person does not get enough sleep;

[0032] Obtain the determination period, and set the total number of days in the determination period as the determination days;

[0033] Obtain the number of days during the determination period when the target person does not get enough sleep, and define it as insufficient sleep days;

[0034] If the number of days with insufficient sleep ÷ the number of days determined ≥ 43%, the target person is determined to have insomnia, and the insomnia data correction strategy is executed.

[0035] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the insomnia data correction strategy is specifically as follows:

[0036] The period when the target person is not in a sleeping state is defined as the correction period;

[0037] The day on which the target person does not get enough sleep is set as the correction collection day;

[0038] Obtain the target person's heart rate during the calibration period of each calibration collection day within the determination period, and set it as the target heart rate;

[0039] If the collected target person's heart rate is ≥ target heart rate × (1 + 45%), it is considered as an abnormal heart rate;

[0040] Calculate the number of valid days in the determination period;

[0041] If an abnormal heart rate exists within the calibration period of the calibration collection day, it means that there are external factors that affect the target person's emotions on the calibration collection day, and the target determination result determined on the calibration collection day is an invalid result, and the result verification strategy is executed on the invalid result;

[0042] Get the number of days corresponding to the invalid results, then the number of valid days = the number of days with insufficient sleep - the number of days corresponding to the invalid results;

[0043] If there is no abnormal heart rate during the calibration period of the calibration collection day, it means that there are no external factors that affect the target person's emotions on the calibration collection day, and the determination result that the target person does not get enough sleep on the calibration collection day is a valid result, then the number of valid days = the number of days with insufficient sleep;

[0044] If the number of effective days ÷ the number of determined days ≥ 43%, it is determined that the target person suffers from insomnia without the influence of external factors, and the insomnia analysis strategy is executed.

[0045] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the result verification strategy is specifically as follows:

[0046] Get all historical heart rate data;

[0047] Extract the date corresponding to each abnormal heart rate and set it as the target date;

[0048] Get the light sleep ratio and deep sleep ratio corresponding to each target date;

[0049] If the target person is judged to have enough sleep on each target date, it means that external factors will not interfere with the target person's sleep state, and the invalid result is regarded as a valid result;

[0050] If the target person is judged to have insufficient sleep on the target date, it means that external factors will interfere with the target person's sleep state;

[0051] Obtain the end date of the determination period corresponding to each target date respectively, and set it as the target end date;

[0052] The target date is used as the start date and the target end date is used as the end date to form a verification time period;

[0053] Obtain the date during the verification period corresponding to each target date, on which the target person is judged to have insufficient sleep, and set it as the verification date;

[0054] Get the number of verification dates and set it as the verification number;

[0055] If the number of verifications is greater than 1 and the interval between the verification date and the target date is greater than 1 or the number of verifications is equal to 1, it means that the number of days that the target person's sleep is affected by external factors is 1 day, and the number of days corresponding to the invalid result is 1;

[0056] If the verification number is greater than 1 and the interval between the verification date and the target date is 1, the number of consecutive verification dates is obtained and defined as the consecutive number. This means that the number of days the target person's sleep is affected by external factors is the consecutive number, and the number of days corresponding to invalid results is the consecutive number.

[0057] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the insomnia analysis strategy is specifically:

[0058] Push a number of target questions to the target person;

[0059] Get the target person's answer to each target question;

[0060] The keywords in each answer of the target person are extracted respectively, defined as target keywords, and a keyword set is formed.

[0061] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the construction of the insomnia knowledge graph is specifically as follows:

[0062] Get all insomnia types;

[0063] Obtain all symptoms corresponding to each insomnia type and define them as insomnia symptoms;

[0064] Obtain all the causes corresponding to each insomnia symptom and determine them as the causes of insomnia;

[0065] According to the insomnia types and the insomnia symptoms corresponding to each insomnia type, a first-level map is formed;

[0066] A secondary map is formed based on the insomnia symptoms and the insomnia causes corresponding to each insomnia symptom;

[0067] Integrate the primary graph and the secondary graph to form an insomnia knowledge graph.

[0068] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the data comparison strategy is specifically as follows:

[0069] Obtain keyword set and insomnia knowledge graph;

[0070] Map each keyword in the keyword set to the insomnia knowledge graph and mark it;

[0071] Extract the labeled insomnia knowledge graph and define it as the labeled graph;

[0072] The data chain in which both the insomnia symptom and the insomnia cause corresponding to the insomnia symptom are marked in the marked atlas is defined as the marked secondary data chain;

[0073] If there is a marked secondary data chain in the marked atlas, the insomnia symptom corresponding to the marked secondary data chain is extracted and determined as the target symptom;

[0074] Extract the insomnia type corresponding to the target symptom in the primary spectrum and identify it as the target type;

[0075] If there is no marked secondary data chain in the marked map, the number of marked elements under each insomnia type is defined as the number of marks;

[0076] The insomnia types are sorted from large to small according to the numerical value of the number of marks, and the insomnia type corresponding to the largest number of marks is identified as the target type.

[0077] As an optional solution of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention, the diagnosis and treatment allocation strategy is specifically:

[0078] Obtain the insomnia type of the target person and set it as the target type;

[0079] Acquire a diagnosis and treatment database, wherein the diagnosis and treatment database includes diagnosis and treatment doctors and the types of insomnia that each diagnosis and treatment doctor is good at treating;

[0080] Extract the treating doctors corresponding to the target type and set them as target doctors;

[0081] The target doctor's diagnosis and treatment end is connected with the target person's diagnosis and treatment end. The target doctor determines the target person's cognition of insomnia based on the target person's insomnia type and the target person's answer to each target question, and determines it as the target cognition. The target person's sleep action before sleep is extracted and determined as the target action.

[0082] The target doctor implements online intervention diagnosis and treatment by conducting cognitive adjustment and relaxation training on the target person based on the target cognition and target action.

[0083] The present invention has the following beneficial effects:

[0084] 1. The insomnia intervention method based on artificial intelligence and cognitive behavior obtains the target person's breathing frequency during sleep, analyzes the target person's light sleep ratio and deep sleep ratio, and determines whether the target person has insomnia problems. If the light sleep ratio and deep sleep ratio are both greater than or equal to the judgment value, it means that the target person has enough sleep. If the light sleep ratio is less than the judgment value or the deep sleep ratio is less than the judgment value, it means that the target person does not have enough sleep. If the target person has too many days of insufficient sleep in a week, it is determined that the target person has insomnia problems. Questionnaires are pushed to patients with insomnia problems in a timely manner, keywords are extracted based on the patient's answers, and the keywords are matched with the insomnia knowledge graph and marked to determine the patient's insomnia type. According to the patient's insomnia type, the patient is pushed to a suitable doctor, and online intervention diagnosis and treatment is provided to the patient, so that the patient can receive corresponding intervention treatment in time at the early stage of insomnia to prevent more serious health problems.

[0085] 2. This insomnia intervention method based on artificial intelligence and cognitive behavior obtains the heart rate of each non-sleep period of the day when the patient with insomnia problems is judged to have insufficient sleep. If there is an abnormal heart rate that is higher than the patient's normal heart rate, it means that there may be external factors that affect the target person's emotions during the day, which makes the target person's sleep disturbed by emotions, resulting in insufficient sleep on that day. In this case, the judgment result of insufficient sleep is invalid, and the number of days of insufficient sleep of the target person needs to be corrected so that the number of days of insufficient sleep of the target person does not have invalid results caused by external factors, so that the data is more accurate and excessive treatment due to misjudgment is prevented.

[0086] 3. The insomnia intervention method based on artificial intelligence and cognitive behavior obtains the historical heart rate data of the target person, determines the date corresponding to each abnormal heart rate, and brings up the sleep data of the target person for one week corresponding to the date. If the sleep results on the day corresponding to each abnormal heart rate are all sufficient sleep, it means that the target person's sleep will not be affected by external factors and emotions. At this time, there is no invalid result, and the number of days of insufficient sleep of the target person remains unchanged. If the sleep result on the day corresponding to the abnormal heart rate is insufficient sleep, the date corresponding to the abnormal heart rate and the number of consecutive days of insufficient sleep are obtained. The number of days is the number of days when the target person's sleep is affected by external factors and emotions. At this time, the number of days with invalid results is the number of days with affected sleep. The corrected number of days of insufficient sleep of the target person is the original number of days of insufficient sleep minus the number of days with invalid results, so that the data is more accurate and excessive treatment due to misjudgment is prevented. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 This is a block diagram of the insomnia intervention method based on artificial intelligence and cognitive behavior of the present invention;

[0088] Figure 2 This is a schematic diagram of the sleep statistics of the target person of the present invention;

[0089] Figure 3 It is a schematic diagram of the insomnia knowledge map of the present invention. DETAILED DESCRIPTION

[0090] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] Embodiment 1, a method for intervening insomnia based on artificial intelligence and cognitive behavior, see Figure 1 ,include:

[0092] Get the target person's breathing rate;

[0093] According to the breathing frequency of the target person, sleep analysis data is formed through a sleep analysis strategy to determine the target person's sleep state, which includes wakefulness, light sleep, and deep sleep, so as to determine whether the target person has insomnia, and perform insomnia disorder intervention on the target person who has insomnia;

[0094] Based on the sleep analysis data, insomnia analysis data is formed through insomnia analysis strategies to push questionnaires to target persons with insomnia and extract key information data from the answers of the target persons, so as to determine the insomnia type of the target persons;

[0095] Constructing an insomnia knowledge graph to match key information data in the target person's questionnaire answers with elements in the insomnia knowledge graph, thereby determining the target person's insomnia type;

[0096] Based on the insomnia analysis data and the insomnia knowledge graph, a data comparison strategy is used to form comparison data to determine the target person's insomnia type;

[0097] Based on the comparison data, the diagnosis and treatment allocation strategy is used to form allocation data to push the diagnosis and treatment doctors to the target person, thereby providing the target person with online intervention diagnosis and treatment;

[0098] Based on the allocated data, the treating doctors are pushed to the target person to provide online intervention treatment for the target person.

[0099] Through the above method, based on the patient's breathing frequency, it is judged and determined whether the patient has insomnia problems and timely intervention is carried out. Based on the patient's heart rate, it is judged and determined whether there are external factors in the non-sleep period that disturb the patient's sleep. Based on the patient's historical heart rate data, it is judged and determined how many days the external factors affect the patient's sleep. The questionnaire is pushed to the patients with insomnia problems in a timely manner, keywords are extracted according to the patients' answers, the keywords are matched with the insomnia knowledge graph and marked, the patient's insomnia type is determined, and according to the patient's insomnia type, the appropriate diagnosis and treatment doctor is pushed to the patient, and the patient is provided with online intervention diagnosis and treatment, so that the patient can get the corresponding intervention treatment in time at the early stage of insomnia to prevent more serious health problems.

[0100] Embodiment 2: This embodiment is an improvement made on the basis of embodiment 1. Figure 2 The insomnia intervention method based on artificial intelligence and cognitive behavior, the sleep analysis strategy, specifically:

[0101] Acquire an initial collection time point and a collection interval duration, wherein the initial collection time point is the time point at which the respiratory frequency of the target person is first collected, and the collection interval duration is the interval duration between the previous collection of the respiratory frequency of the target person and the next collection of the respiratory frequency of the target person;

[0102] The initial collection time point is used as the starting time point for collecting the respiratory frequency of the target person, and a collection time point is formed at each interval, and the respiratory frequency of the target person is recorded;

[0103] According to each acquisition time point and the breathing frequency of the target person recorded at each acquisition time point, a frequency set is formed, wherein the frequency set includes the initial acquisition time point and the breathing frequency of the target person acquired at the initial acquisition time point;

[0104] According to each acquisition time point in the frequency set and the target person's breathing frequency recorded at each acquisition time point, a target person's sleep statistics graph is formed, such as Figure 2 As shown;

[0105] The first frequency data point, the second frequency data point and the third frequency data point in the target person's sleep statistics graph are obtained, wherein the first frequency data point is the target person's breathing frequency when he or she just falls asleep, the second frequency data point is the target person's breathing frequency when he or she just enters light sleep, and the third frequency data point is the target person's breathing frequency when he or she just enters deep sleep. Figure 2 As shown;

[0106] The starting acquisition time points corresponding to the first frequency data point, the second frequency data point and the third frequency data point are respectively extracted and defined as the first starting time point, the second starting time point and the third starting time point, such as Figure 2 As shown;

[0107] The termination acquisition time points corresponding to the first frequency data point, the second frequency data point, and the third frequency data point are respectively extracted and defined as the first termination time point, the second termination time point, and the third termination time point, such as Figure 2 As shown;

[0108] Calculate the first sleep duration, where the first sleep duration = (the second starting time point - the first starting time point) + (the first ending time point - the second ending time point), and the first sleep duration is the duration from falling asleep to light sleep;

[0109] Calculate the second sleep duration, the second sleep duration = (third starting time point - second starting time point) + (second ending time point - third ending time point), the second sleep duration is the light sleep duration;

[0110] Calculate the third sleep duration, where the third sleep duration=the third end time point-the third start time point, and the third sleep duration is the deep sleep duration;

[0111] Then the total sleep time = first sleep time + second sleep time + third sleep time;

[0112] Execute the sleep determination strategy.

[0113] The sleep determination strategy is specifically as follows:

[0114] Obtaining a light sleep ratio determination value and a deep sleep ratio determination value, wherein the light sleep ratio determination value is a determination value for determining whether the target person has sufficient light sleep, and the deep sleep ratio determination value is a determination value for determining whether the target person has sufficient deep sleep;

[0115] Calculate the target person's light sleep ratio, light sleep ratio = second sleep duration ÷ total sleep duration;

[0116] Calculate the target person's deep sleep ratio, deep sleep ratio = third sleep duration ÷ total sleep duration;

[0117] If the light sleep ratio ≥ the light sleep ratio judgment value and the deep sleep ratio ≥ the deep sleep ratio judgment value, it means that the target person has enough sleep;

[0118] If the light sleep ratio is less than the light sleep ratio judgment value or the deep sleep ratio is less than the deep sleep ratio judgment value, it means that the target person does not get enough sleep;

[0119] Obtain a determination period, and define the total number of days in the determination period as the determination days. If the determination period is one week, the determination days are seven days.

[0120] Obtain the number of days during the determination period when the target person does not get enough sleep, and define it as insufficient sleep days;

[0121] If the number of days with insufficient sleep ÷ the number of days determined ≥ 43%, the target person is determined to have insomnia, and the insomnia data correction strategy is executed.

[0122] The insomnia data correction strategy is specifically as follows:

[0123] The period when the target person is not in a sleeping state is defined as the correction period;

[0124] The day on which the target person does not get enough sleep is set as the correction collection day;

[0125] Obtaining the heart rate of the target person during the calibration period of each calibration collection day within the determination period, and setting it as the target heart rate, wherein the target heart rate is the heart rate of the target person in a stable mood;

[0126] If the collected target person's heart rate is ≥ target heart rate × (1 + 45%), it is considered as an abnormal heart rate;

[0127] Calculate the number of valid days in the determination period;

[0128] If an abnormal heart rate exists within the calibration period of the calibration collection day, it means that there are external factors that affect the target person's emotions on the calibration collection day, and the target determination result determined on the calibration collection day is an invalid result, and the result verification strategy is executed on the invalid result;

[0129] Get the number of days corresponding to the invalid results, then the number of valid days = the number of days with insufficient sleep - the number of days corresponding to the invalid results;

[0130] If there is no abnormal heart rate during the calibration period of the calibration collection day, it means that there are no external factors that affect the target person's emotions on the calibration collection day, and the determination result that the target person does not get enough sleep on the calibration collection day is a valid result, then the number of valid days = the number of days with insufficient sleep;

[0131] If the number of effective days ÷ the number of determined days ≥ 43%, it is determined that the target person suffers from insomnia without the influence of external factors, and the insomnia analysis strategy is executed.

[0132] The result verification strategy is specifically as follows:

[0133] Acquire all historical heart rate data, where the historical heart rate data is the heart rate collected during the calibration period of each day;

[0134] Extract the date corresponding to each abnormal heart rate and set it as the target date;

[0135] Get the light sleep ratio and deep sleep ratio corresponding to each target date;

[0136] If the target person is judged to have enough sleep on each target date, it means that external factors will not interfere with the target person's sleep state, and the invalid result is regarded as a valid result;

[0137] If the target person is judged to have insufficient sleep on the target date, it means that external factors will interfere with the target person's sleep state;

[0138] Obtain the end date of the determination period corresponding to each target date respectively, and set it as the target end date. If the determination period is seven days, such as Monday to Sunday, the end date is Sunday;

[0139] The target date is used as the start date and the target end date is used as the end date to form a verification time period. For example, if the target date is Wednesday and the target end date is Sunday, the verification time period is from Wednesday to Sunday.

[0140] Obtaining the date on which the target person is determined to have insufficient sleep within the verification time period corresponding to each target date, and setting it as the verification date, wherein the verification date includes the target date;

[0141] Get the number of verification dates and set it as the verification number;

[0142] If the verification number is greater than 1 and the interval between the verification date and the target date is greater than 1 or the verification number is equal to 1, it means that the number of days that the target person's sleep is affected by external factors is 1 day, and the number of days corresponding to the invalid result is 1. For example, if only the target date Wednesday is used to determine that the target person does not get enough sleep, the verification date is Wednesday, and the number of days corresponding to the invalid result is 1. If the verification dates for determining that the target person does not get enough sleep are Wednesday, Friday, and Saturday, then the interval between the verification date and the target date = Friday - Wednesday = 2, indicating that the number of days that the target person's sleep is affected by external factors is 1 day, and the number of days corresponding to the invalid result is 1.

[0143] If the verification quantity is greater than 1 and the number of days between the verification date and the target date is 1, then the number of consecutive verification dates is obtained and defined as the consecutive number. This means that the number of days the target person's sleep is affected by external factors is a consecutive number, and the number of days corresponding to invalid results is a consecutive number. For example, if the verification dates for determining that the target person does not get enough sleep are Wednesday, Thursday and Saturday, then the number of days between the verification date and the target date = Thursday - Wednesday = 1, indicating that Wednesday and Thursday are consecutive verification dates. The consecutive number is 2, indicating that the number of days the target person's sleep is affected by external factors is 2 days, and the number of days corresponding to invalid results is 2.

[0144] Embodiment 3: This embodiment is an improvement made on the basis of embodiment 2. Figure 3 In this embodiment, the insomnia analysis strategy is specifically:

[0145] Pushing a number of target questions to the target person, wherein the target questions are used to determine whether the target person suffers from insomnia, whether the insomnia is serious, and the type of insomnia;

[0146] Get the target person's answer to each target question;

[0147] The keywords in each answer of the target person are extracted respectively, determined as target keywords, and a keyword set is formed. For example, the target keywords are anxiety, stress, difficulty falling asleep, and poor sleep quality.

[0148] This embodiment also provides that the construction of the insomnia knowledge graph is specifically:

[0149] Get all insomnia types, such as primary insomnia, secondary insomnia and sleep-wake rhythm disorder, corresponding to Figure 3 Insomnia type A, Insomnia type B and Insomnia type C;

[0150] Get all the symptoms corresponding to each insomnia type and define them as insomnia symptoms. For example, the insomnia symptoms of insomnia type A are difficulty falling asleep and poor sleep quality, which correspond to Figure 3Insomnia symptoms A1-Insomnia symptoms An, Insomnia symptoms of Insomnia type 2 are difficulty falling asleep, sleep maintenance disorder, early awakening, daytime fatigue and irritability, etc., corresponding to Figure 3 Insomnia symptoms B1-Bn, insomnia symptoms of insomnia type 3 include difficulty falling asleep, early awakening and daytime sleepiness, etc., corresponding to Figure 3 Insomnia symptoms C1-Insomnia symptoms Cn;

[0151] Obtain all the causes corresponding to each insomnia symptom and define them as the causes of insomnia. For example, the causes of insomnia corresponding to insomnia symptoms A1-An include anxiety, stress, and chronic pain, etc., the causes of insomnia corresponding to insomnia symptoms B1-Bn include depression, anxiety, chronic pain, and drug abuse, etc., and the causes of insomnia corresponding to insomnia symptoms C1-Cn include traveling across time zones, working the night shift, or irregular sleep-wake schedules, etc.;

[0152] According to the insomnia types and the insomnia symptoms corresponding to each insomnia type, a first-level map is formed;

[0153] A secondary map is formed based on the insomnia symptoms and the insomnia causes corresponding to each insomnia symptom;

[0154] Integrate the primary graph and the secondary graph to form an insomnia knowledge graph, such as Figure 3 shown.

[0155] This embodiment also provides that the data comparison strategy is specifically:

[0156] Obtain keyword set and insomnia knowledge graph;

[0157] Map each keyword in the keyword set to the insomnia knowledge graph and mark it;

[0158] Extract the labeled insomnia knowledge graph and define it as the labeled graph;

[0159] The data chain in which both the insomnia symptom and the insomnia cause corresponding to the insomnia symptom are marked in the marked atlas is defined as the marked secondary data chain. Figure 3 For example, if Figure 3 Insomnia symptom A1 and insomnia cause A1C1 are marked, then the data chain formed by insomnia symptom A1 and insomnia cause A1C1 is identified as a marked secondary data chain;

[0160] If there is a marked secondary data chain in the marked spectrum, the insomnia symptom corresponding to the marked secondary data chain is extracted and determined as the target symptom. For example, if the marked secondary data chain is a data chain formed by insomnia symptom A1 and insomnia cause A1C1, the target symptom is insomnia symptom A1;

[0161] Extract the insomnia type corresponding to the target symptom in the primary spectrum and identify it as the target type. For example, if the target symptom is insomnia symptom A1, and the insomnia type corresponding to insomnia symptom A1 is insomnia type A, then the target type is insomnia type A.

[0162] If there is no secondary data chain in the marked map, the number of marked elements under each insomnia type is defined as the number of marks. Figure 3 For example, if Figure 3 The marked elements are insomnia symptom A4, insomnia symptom B8, insomnia symptom C7, insomnia cause C1C2, insomnia cause C2C1, insomnia cause C4C7. Then the number of marks for insomnia type A is 1, the number of marks for insomnia type B is 1, and the number of marks for insomnia type C is 4.

[0163] Sort the insomnia types according to the numerical value of the number of marks from large to small, and identify the insomnia type corresponding to the number of marks with the largest numerical value as the target type. For example, the number of marks for insomnia type A is 1, the number of marks for insomnia type B is 1, and the number of marks for insomnia type C is 4, then the target type is insomnia type C.

[0164] This embodiment also provides that the diagnosis and treatment allocation strategy is specifically:

[0165] Obtain the insomnia type of the target person and set it as the target type;

[0166] Acquire a diagnosis and treatment database, wherein the diagnosis and treatment database includes diagnosis and treatment doctors and the types of insomnia that each diagnosis and treatment doctor is good at treating;

[0167] Extract the treating doctors corresponding to the target type and set them as target doctors;

[0168] The target doctor's diagnosis and treatment end is connected with the target person's diagnosis and treatment end. The target doctor determines the target person's cognition of insomnia based on the target person's insomnia type and the target person's answer to each target question, and determines it as the target cognition. The target person's sleep action before sleep is extracted and determined as the target action. The target person's cognition of insomnia is the target person's view on insomnia, such as anxiety, etc. The sleep action includes playing games, watching TV, chatting, etc.;

[0169] The target doctor implements online intervention diagnosis and treatment by conducting cognitive adjustment and relaxation training on the target person based on the target cognition and target action.

[0170] In this embodiment, based on the patient's breathing frequency, it is judged and determined whether the patient has insomnia problems and timely intervention is carried out. Based on the patient's heart rate, it is judged and determined whether there are external factors in the patient's non-sleep period that disturb the patient's sleep. Based on the patient's historical heart rate data, it is judged and determined the number of days when external factors affect the patient's sleep. A questionnaire is pushed to patients with insomnia problems in a timely manner, keywords are extracted based on the patient's answers, the keywords are matched with the insomnia knowledge graph and marked, the patient's insomnia type is determined, and based on the patient's insomnia type, a suitable doctor is pushed to the patient, and online intervention diagnosis and treatment is provided to the patient, so that the patient can receive corresponding intervention treatment in a timely manner in the early stages of insomnia to prevent more serious health problems.

[0171] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0172] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An intervention method for insomnia based on artificial intelligence and cognitive behavior, characterized by: include: Get the target person's breathing rate; According to the breathing frequency of the target person, sleep analysis data is formed through a sleep analysis strategy to determine the target person's sleep state, which includes wakefulness, light sleep, and deep sleep, so as to determine whether the target person has insomnia, and perform insomnia disorder intervention on the target person who has insomnia; Based on the sleep analysis data, insomnia analysis data is formed through insomnia analysis strategies, so as to push questionnaires to target persons with insomnia and extract key information data from the answers of the target persons, so as to determine the insomnia type of the target persons; Constructing an insomnia knowledge graph to match key information data in the target person's questionnaire answers with elements in the insomnia knowledge graph, thereby determining the target person's insomnia type; Based on the insomnia analysis data and the insomnia knowledge graph, a data comparison strategy is used to form comparison data to determine the target person's insomnia type; Based on the comparison data, the diagnosis and treatment allocation strategy is used to form allocation data to push the diagnosis and treatment doctors to the target person, thereby providing the target person with online intervention diagnosis and treatment; The sleep analysis strategy is specifically: Get the initial collection time point and collection interval duration; The initial collection time point is used as the starting time point for collecting the respiratory frequency of the target person, and a collection time point is formed at each interval, and the respiratory frequency of the target person is recorded; A frequency set is formed according to each acquisition time point and the breathing frequency of the target person recorded at each acquisition time point; According to each acquisition time point in the frequency set and the breathing frequency of the target person recorded at each acquisition time point, a sleep statistics graph of the target person is formed; Obtain a first frequency data point, a second frequency data point, and a third frequency data point in a target person's sleep statistics graph; Extracting the starting acquisition time points corresponding to the first frequency data point, the second frequency data point and the third frequency data point respectively, and defining them as the first starting time point, the second starting time point and the third starting time point respectively; Extracting the termination collection time points corresponding to the first frequency data point, the second frequency data point and the third frequency data point respectively, and defining them as the first termination time point, the second termination time point and the third termination time point respectively; Calculate the first sleep duration, first sleep duration = (second starting time point - first starting time point) + (first ending time point - second ending time point); Calculate the second sleep duration, the second sleep duration = (third starting time point - second starting time point) + (second ending time point - third ending time point); Calculate the third sleep duration, where the third sleep duration = the third end time point - the third start time point; Then the total sleep time = first sleep time + second sleep time + third sleep time; Implement sleep determination strategy; The sleep determination strategy is specifically: Obtaining a light sleep ratio determination value and a deep sleep ratio determination value; Calculate the target person's light sleep ratio, light sleep ratio = second sleep duration ÷ total sleep duration; Calculate the target person's deep sleep ratio, deep sleep ratio = third sleep duration ÷ total sleep duration; If the light sleep ratio ≥ the light sleep ratio judgment value and the deep sleep ratio ≥ the deep sleep ratio judgment value, it means that the target person has enough sleep; If the light sleep ratio is less than the light sleep ratio judgment value or the deep sleep ratio is less than the deep sleep ratio judgment value, it means that the target person does not get enough sleep; Obtain the determination period, and set the total number of days in the determination period as the determination days; Obtain the number of days during the determination period when the target person does not get enough sleep, and define it as insufficient sleep days; If the number of days with insufficient sleep ÷ the number of days determined ≥ 43%, the target person is determined to have insomnia, and the insomnia data correction strategy is implemented; The insomnia data correction strategy is specifically: The period when the target person is not in a sleeping state is defined as the correction period; The day on which the target person does not get enough sleep is set as the correction collection day; Obtain the target person's heart rate during the calibration period of each calibration collection day within the determination period, and set it as the target heart rate; If the collected target person's heart rate is ≥ target heart rate × (1 + 45%), it is considered as an abnormal heart rate; Calculate the number of valid days in the determination period; If an abnormal heart rate exists within the calibration period of the calibration collection day, it means that there are external factors that affect the target person's emotions on the calibration collection day, and the target determination result determined on the calibration collection day is an invalid result, and the result verification strategy is executed on the invalid result; Get the number of days corresponding to the invalid results, then the number of valid days = the number of days with insufficient sleep - the number of days corresponding to the invalid results; If there is no abnormal heart rate during the calibration period of the calibration collection day, it means that there are no external factors that affect the target person's emotions on the calibration collection day, and the determination result that the target person does not get enough sleep on the calibration collection day is a valid result, then the number of valid days = the number of days with insufficient sleep; If the number of effective days ÷ the number of judged days ≥ 43%, the target person is judged to have insomnia without external factors, and the insomnia analysis strategy is implemented; The result verification strategy is specifically as follows: Get all historical heart rate data; Extract the date corresponding to each abnormal heart rate and set it as the target date; Get the light sleep ratio and deep sleep ratio corresponding to each target date; If the target person is judged to have enough sleep on each target date, it means that external factors will not interfere with the target person's sleep state, and the invalid result is regarded as a valid result; If the target person is judged to have insufficient sleep on the target date, it means that external factors will interfere with the target person's sleep state; Obtain the end date of the determination period corresponding to each target date respectively, and set it as the target end date; The target date is used as the start date and the target end date is used as the end date to form a verification time period; Obtain the date during the verification period corresponding to each target date, on which the target person is judged to have insufficient sleep, and set it as the verification date; Get the number of verification dates and set it as the verification number; If the number of verifications is greater than 1 and the interval between the verification date and the target date is greater than 1 or the number of verifications is equal to 1, it means that the number of days that the target person's sleep is affected by external factors is 1 day, and the number of days corresponding to the invalid result is 1; If the verification number is greater than 1 and the interval between the verification date and the target date is 1, the number of consecutive verification dates is obtained and defined as the consecutive number. This means that the number of days the target person's sleep is affected by external factors is the consecutive number, and the number of days corresponding to invalid results is the consecutive number.

2. The method for intervening insomnia based on artificial intelligence and cognitive behavior according to claim 1, characterized in that: The insomnia analysis strategy is specifically: Push a number of target questions to the target person; Get the target person's answer to each target question; The keywords in each answer of the target person are extracted respectively, defined as target keywords, and a keyword set is formed.

3. The artificial intelligence and cognitive behavior based insomnia intervention method according to claim 1, characterized in that: The construction of the insomnia knowledge graph is specifically as follows: Get all insomnia types; Obtain all symptoms corresponding to each insomnia type and define them as insomnia symptoms; Obtain all the causes corresponding to each insomnia symptom and determine them as the causes of insomnia; According to the insomnia types and the insomnia symptoms corresponding to each insomnia type, a first-level map is formed; A secondary map is formed based on the insomnia symptoms and the insomnia causes corresponding to each insomnia symptom; Integrate the primary graph and the secondary graph to form an insomnia knowledge graph.

4. The artificial intelligence and cognitive behavior based insomnia intervention method according to claim 1, characterized in that: The data comparison strategy is specifically as follows: Obtain keyword set and insomnia knowledge graph; Map each keyword in the keyword set to the insomnia knowledge graph and mark it; Extract the labeled insomnia knowledge graph and define it as the labeled graph; The data chain in which both the insomnia symptom and the insomnia cause corresponding to the insomnia symptom are marked in the marked atlas is defined as the marked secondary data chain; If there is a marked secondary data chain in the marked atlas, the insomnia symptom corresponding to the marked secondary data chain is extracted and determined as the target symptom; Extract the insomnia type corresponding to the target symptom in the primary spectrum and identify it as the target type; If there is no marked secondary data chain in the marked map, the number of marked elements under each insomnia type is defined as the number of marks; The insomnia types are sorted from large to small according to the numerical value of the number of marks, and the insomnia type corresponding to the largest number of marks is identified as the target type.

5. The method for intervening insomnia based on artificial intelligence and cognitive behavior according to claim 1, characterized in that: The diagnosis and treatment allocation strategy is specifically: Obtain the insomnia type of the target person and set it as the target type; Acquire a diagnosis and treatment database, wherein the diagnosis and treatment database includes diagnosis and treatment doctors and the types of insomnia that each diagnosis and treatment doctor is good at treating; Extract the treating doctors corresponding to the target type and set them as target doctors; The target doctor's diagnosis and treatment end is connected with the target person's diagnosis and treatment end. The target doctor determines the target person's cognition of insomnia based on the target person's insomnia type and the target person's answer to each target question, and determines it as the target cognition. The target person's sleep action before sleep is extracted and determined as the target action. The target doctor implements online intervention diagnosis and treatment by conducting cognitive adjustment and relaxation training on the target person based on the target cognition and target action.

Images