Intelligent neck pressure regulating device based on deep learning

Abstract

The invention provides an intelligent neck pressure regulating device based on deep learning, and the device comprises a learning data obtaining subsystem which is used for obtaining neck lymphatic leakage local treatment pressurization data; the deep learning subsystem is used for learning neck lymphatic leakage local treatment pressurization data based on a deep learning model to obtain a simulation pressurization model; the simulation pressurization parameter determination subsystem is used for acquiring postoperative condition information of a patient wearing the intelligent neck pressure regulator and determining simulation pressurization parameters according to a simulation pressurization model; and the intelligent pressure regulating subsystem is used for triggering a pressure regulating signal of the intelligent neck pressure regulator according to the simulation pressurization parameter. The invention relates to an intelligent neck pressure regulating device based on deep learning. A deep learning technology is utilized to learn neck lymphatic leakage local treatment pressurization data to obtain a simulation pressurization model; the simulation pressurization parameters are determined according to the postoperative condition information of the patient, the pressure regulation control instruction of the neck pressure regulator is triggered, and the local pressurization treatment efficiency of the neck lymphatic leakage is higher and more intelligent.

Description

Technical Field

[0001] The present invention relates to the field of deep learning technology, and in particular to an intelligent neck pressure regulating device based on deep learning. Background Art

[0002] Thyroid cancer is the most common malignant tumor of the endocrine system and a common malignant tumor in head and neck surgery. Surgical radical neck lymph node dissection is one of the best surgical options. Chyle leakage or lymphatic leakage is a complication after lateral neck lymph node dissection. Chyle leakage or lymphatic leakage is usually caused by rupture of the thoracic duct and lymphatic duct branches. It is especially prone to occur when separating the left jugular angle. Chyle leakage or lymphatic leakage mostly occurs on the second day after surgery, and a few occur on the third and fourth days after surgery. The drainage volume is less than 1000ml / d. Patients with chyle leakage or lymphatic leakage can be cured by conservative treatment, but the disadvantage is that the hospitalization time is prolonged.

[0003] At present, the clinical treatment methods for chylous leakage or lymphatic leakage include negative pressure drainage balls, local pressurization, low-fat diet, intravenous administration of growth inhibitors and parenteral nutrition support therapy to accelerate patient recovery and shorten the length of hospital stay. However, the above pressurization operation requires medical staff to perform manually based on personal experience, which has low processing efficiency and is not smart enough.

[0004] In view of this, there is an urgent need for an intelligent neck pressure regulating device based on deep learning to at least solve the above-mentioned shortcomings. Summary of the invention

[0005] One of the purposes of the present invention is to provide an intelligent neck pressure regulating device based on deep learning, which uses deep learning technology to learn the local treatment pressurization data of neck lymph leakage after lateral cervical lymph node dissection of target patients by medical staff to obtain a simulated pressurization model; then automatically determines the simulated pressurization parameters according to the postoperative condition information of the patient wearing the intelligent neck pressure regulator; the simulated pressurization parameters are converted by a pressure regulation signal conversion template to output the pressure regulation control instructions of the neck pressure regulator to control the neck pressure, without the need for manual pressurization by medical staff, and the local pressurization treatment of neck lymph leakage is more efficient and intelligent.

[0006] An embodiment of the present invention provides an intelligent neck pressure regulating device based on deep learning, comprising:

[0007] A learning data acquisition subsystem is used to obtain the local treatment and compression data of neck lymph node leakage after the lateral neck lymph node dissection of the target patient by medical staff;

[0008] A deep learning subsystem, used to learn the local treatment pressurization data of cervical lymphatic leakage based on a deep learning model and obtain a simulated pressurization model;

[0009] A simulated pressurization parameter determination subsystem is used to obtain postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determine the simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model;

[0010] The intelligent pressure regulation subsystem is used to determine the pressure regulation signal and perform intelligent pressure regulation control according to the analog pressurization parameters and the pressure regulation signal conversion template of the intelligent neck pressure regulator.

[0011] Preferably, the learning data acquisition subsystem acquires the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection of the target patient by medical staff, including:

[0012] Connect with medical information sources;

[0013] Through the medical information source, the local treatment and pressurization data of cervical lymph node leakage after lateral cervical lymph node dissection performed by medical staff on the target patient are retrieved; wherein, the local treatment and pressurization data of cervical lymph node leakage include: the postoperative condition information and reference pressure parameters of the target patient.

[0014] Preferably, the deep learning subsystem learns the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model to obtain a simulated pressurization model, including:

[0015] The postoperative condition information of the target patient is used as the input of the Vi T model, and the reference pressure parameters are used as the output of the Vi T model to train the simulated pressurization model.

[0016] Preferably, the simulated pressurization parameter determination subsystem obtains postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determines the simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model, including:

[0017] Obtain clinical data and medical imaging data of patients wearing the intelligent neck pressure regulator;

[0018] Use clinical data and medical imaging data as postoperative condition information for patients wearing the device;

[0019] The postoperative condition information of the patient wearing the device is used as the input of the simulated pressurization model to obtain the simulated pressurization parameters output by the simulated pressurization model.

[0020] Preferably, the learning data acquisition subsystem retrieves the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection of the target patient by medical staff through the medical information source, including:

[0021] Determine the medical item label of the pre-selected patient through the medical information source;

[0022] Retrieve lymph node leakage treatment project tags from medical project tags, and obtain case data associated with the successfully retrieved medical project tags;

[0023] The data types of case data are obtained; the data types include: post-clearance condition data, lymphatic leakage treatment process data, and lymphatic leakage treatment effect data;

[0024] Determine whether the case data meets the data screening criteria;

[0025] If so, the preselected patients corresponding to the case data were selected as target patients and the local treatment compression data for cervical lymphatic leakage were obtained.

[0026] Preferably, the learning data acquisition subsystem determines whether the case data meets the data screening criteria, including:

[0027] Traverse the case sub-data of each data type in the case data in turn, and take the case sub-data currently being traversed as the target case sub-data;

[0028] When the data type is post-cleavage condition data, a standard condition item description scoring template is obtained; the standard condition item description scoring template includes: one-to-one corresponding standard condition items and standard condition item specification description factors;

[0029] Parse the target case sub-data to obtain record condition items and record condition item description factors;

[0030] Obtain the missing condition item of the record condition item corresponding to the standard condition item;

[0031] Get the missing case scoring library;

[0032] Determine the missing situation score based on the missing situation score library and the missing situation items;

[0033] Obtaining a matching situation item of the record situation item corresponding to the standard situation item;

[0034] Calculate the factor cosine value of the record situation item description factor of the matching situation item and the corresponding matching standard situation item and the standard situation item specification description factor;

[0035] Obtain a matching scenario scoring library;

[0036] Determine a matching situation score according to the matching situation score library and the factor cosine value;

[0037] Sum the missing situation score and the matching situation score to obtain the first score value of the target case sub-data;

[0038] When the data type is lymphatic leakage treatment process data, obtain the evaluation records during the treatment process;

[0039] Determine the evaluation type of the evaluation record, which includes active evaluation and passive evaluation;

[0040] Get the evaluation record scoring strategy corresponding to the evaluation type;

[0041] According to the evaluation record scoring strategy, determine the treatment process scores corresponding to the evaluation records of different evaluation types;

[0042] Calculate the second score value of the target case sub-data according to the preset evaluation weight and treatment process score corresponding to the evaluation type;

[0043] When the data type is lymphatic leakage treatment effect data, the effect score of the target case sub-data is obtained and used as the third score value;

[0044] After all case sub-data of each data type in the case data are traversed, the first score value, the second score value and the third score value are cumulatively calculated to obtain the data screening value of the case data;

[0045] If the data screening value is greater than or equal to the preset data screening value threshold, the case data is determined to meet the data screening criteria.

[0046] Preferably, the learning data acquisition subsystem determines the treatment process scores corresponding to the evaluation records of different evaluation types according to the evaluation record scoring strategy, including:

[0047] When the evaluation type is active evaluation, parse the evaluation record and obtain the patient's evaluation words;

[0048] Obtaining a preset discomfort level value determination library;

[0049] Determine the discomfort level value of the evaluation record of the active evaluation according to the patient's evaluation words and the discomfort level value of the library;

[0050] Determine the treatment process score based on the conversion relationship between the preset discomfort level value and the treatment process score;

[0051] When the evaluation type is passive evaluation, parse the evaluation record and obtain the evaluation words of the medical staff;

[0052] Based on the lymphatic leakage treatment knowledge records in the lymphatic leakage treatment knowledge base, a lymphatic leakage treatment rationality analysis model is trained;

[0053] Characterizing the evaluation words of medical staff to obtain evaluation word features, the evaluation word features include: the meaning of the lymphatic leakage treatment process evaluated by the evaluation word, the evaluation word meaning, the reverse evaluation word meaning, and the target word meaning between the evaluation word corresponding to the evaluation word meaning and the evaluation word corresponding to the reverse evaluation word meaning;

[0054] The evaluation word features are input into the lymph node leakage treatment rationality analysis model to obtain the rationality value, which is used as the treatment process score.

[0055] An embodiment of the present invention provides an intelligent neck pressure regulating device based on deep learning, further comprising:

[0056] The offset detection subsystem is used to detect the offset of the pressing position during the operation of the intelligent neck pressure regulator.

[0057] Preferably, the offset detection subsystem performs pressing position offset detection during the operation of the intelligent neck pressure regulator, including:

[0058] During the operation of the intelligent neck pressure regulator, dynamic information of the intelligent neck pressure regulator is obtained;

[0059] If the dynamic information is successfully acquired, the pressing position offset detection is performed.

[0060] Preferably, if the dynamic information is successfully acquired, the offset detection subsystem performs a pressing position offset detection, including:

[0061] If the dynamic information is successfully obtained, the temperature sensing data of the pressurized area within the preset time of the intelligent neck pressure regulator is obtained;

[0062] Based on the temperature sensing data, draw the temperature change gradient diagram of different pressurized areas;

[0063] Calculate the gradient map similarity of the temperature change gradient maps of different pressurized areas;

[0064] If the gradient map similarity is greater than or equal to the preset gradient map similarity threshold, a deviation reminder is issued to the wearer.

[0065] An embodiment of the present invention provides an intelligent neck pressure regulation method based on deep learning, comprising:

[0066] Step 1: Obtain the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection performed by medical staff on the target patients;

[0067] Step 2: Learning the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model to obtain a simulated pressurization model;

[0068] Step 3: Obtain postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determine simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model;

[0069] Step 4: According to the simulated pressurization parameters and the pressure regulation signal conversion template of the intelligent neck pressure regulator, the pressure regulation signal is determined and intelligent pressure regulation control is performed.

[0070] The beneficial effects of the present invention are:

[0071] The present invention utilizes deep learning technology to learn the local treatment pressurization data of cervical lymph node dissection in the lateral cervical area of ​​the target patient by medical staff to obtain a simulated pressurization model; then automatically determines the simulated pressurization parameters according to the postoperative condition information of the patient wearing the intelligent neck pressure regulator; the simulated pressurization parameters are converted by a pressure regulation signal conversion template to output the pressure regulation control instructions of the neck pressure regulator to control the neck pressure, without the need for manual pressurization by medical staff, and the local pressurization treatment of cervical lymph node leakage is more efficient and intelligent.

[0072] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0073] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0075] Figure 1 Schematic diagram of an intelligent neck pressure regulating device based on deep learning in an embodiment of the present invention.

[0076] Figure 2 Schematic diagram of an intelligent neck pressure regulation method based on deep learning in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0078] The embodiment of the present invention provides an intelligent neck pressure regulating device based on deep learning, such as Figure 1 As shown, including:

[0079] Learning data acquisition subsystem 1, used to obtain the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection of the target patient by medical staff;

[0080] Among them, the learning data acquisition subsystem obtains the local treatment and compression data of cervical lymph node leakage after the target patient's lateral cervical lymph node dissection by medical staff, including:

[0081] Connecting to medical information sources; medical information sources are data source nodes in the medical information system, including hospital information systems, electronic medical record systems, and image archiving and transmission systems. Through these nodes, patients' medical data can be retrieved and obtained;

[0082] Through the medical information source, the local treatment and pressurization data of neck lymph node leakage after the target patient's lateral neck lymph node dissection by medical staff are retrieved; wherein, the local treatment and pressurization data of neck lymph node leakage include: the target patient's postoperative condition information and reference pressure parameters; wherein, the postoperative condition information includes: the target patient's medical imaging data (such as: the target patient's neck ultrasound image) and clinical data (such as: the time of lymph node leakage, the drainage volume, the classification of lymph node leakage), etc.; the reference pressure parameters are: the specific parameters used for the local pressurization treatment of neck lymph node leakage after the target patient's lateral neck lymph node dissection, such as: pressurization position, pressurization time and pressurization pressure, etc.;

[0083] Deep learning subsystem 2, used to learn the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model and obtain a simulated pressurization model;

[0084] Among them, the deep learning subsystem learns the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model to obtain the simulated pressurization model, including:

[0085] The postoperative condition information of the target patient is used as the input of the Vi T model, and the reference pressure parameter of the target patient is used as the output of the Vi T model to train the simulated pressurization model; wherein the Vi T model is a Vi si on Transformer model;

[0086] The simulated pressurization parameter determination subsystem 3 is used to obtain the postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determine the simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model;

[0087] The simulated pressurization parameter determination subsystem obtains the postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determines the simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model, including:

[0088] Obtain clinical data and medical imaging data of patients wearing the intelligent neck pressure regulator;

[0089] Use clinical data and medical imaging data as postoperative condition information for patients wearing the device;

[0090] Using the postoperative condition information of the patient wearing the device as the input of the simulated pressurization model, and obtaining the simulated pressurization parameters output by the simulated pressurization model;

[0091] The intelligent pressure regulating subsystem 4 is used to determine the pressure regulating signal and perform intelligent pressure regulating control according to the simulated pressurization parameters and the pressure regulating signal conversion template of the intelligent neck pressure regulating instrument. Among them, the pressure regulating signal conversion template is: a template for obtaining the pressure regulating control instruction (pressure regulating signal) of the intelligent neck pressure regulating instrument by comparing the pressure parameter that needs to be triggered by the intelligent neck pressure regulating instrument. After the pressure regulating signal is triggered, the intelligent neck pressure regulating instrument will automatically apply the pressure corresponding to the pressure parameter to the patient wearing it through its preset pressurization area.

[0092] The working principle and beneficial effects of the above technical solution are:

[0093] The present invention utilizes deep learning technology to learn the local treatment pressurization data of cervical lymph node dissection in the lateral cervical area of ​​the target patient by medical staff to obtain a simulated pressurization model; then automatically determines the simulated pressurization parameters according to the postoperative condition information of the patient wearing the intelligent neck pressure regulator; the simulated pressurization parameters are converted by a pressure regulation signal conversion template to output the pressure regulation control instructions of the neck pressure regulator to control the neck pressure, without the need for manual pressurization by medical staff, and the local pressurization treatment of cervical lymph node leakage is more efficient and intelligent.

[0094] In one embodiment, the learning data acquisition subsystem retrieves the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection of the target patient by medical staff through the medical information source, including:

[0095] Determine the medical project label of the pre-selected patient through the medical information source; wherein the medical information source is: various data sources containing medical related information, such as: the hospital's electronic medical record system, medical literature database and clinical research data;

[0096] Retrieve lymphatic leakage treatment project tags from medical project tags, and obtain case data associated with the successfully retrieved medical project tags; wherein the medical project tags are keywords for classifying and identifying medical projects; the lymphatic leakage treatment project tags are project tags related to lymphatic leakage treatment, such as "lymphatic leakage treatment"; and the case data are data related to lymphatic leakage treatment of patients of lymphatic leakage treatment projects recorded in the medical information source;

[0097] The data types of case data are obtained; the data types include: post-clearance situation data, lymphatic leakage treatment process data, and lymphatic leakage treatment effect data; among them, post-clearance situation data are: the patient's condition after the pre-selected patients have completed the lateral neck lymph node clearance, such as: neck CT scan images, basic information of patients (patient age, gender, weight and historical medical records, etc.); lymphatic leakage treatment process data are: data during the treatment of lymphatic leakage, such as: the location, strength and time of pressure applied when medical staff use elastic bandages to apply pressure to treat cervical lymphatic leakage; lymphatic leakage treatment effect data are: data reflecting the treatment effect of lymphatic leakage, such as: changes in drainage volume after treatment, symptom relief and complication rate, etc.;

[0098] Determine whether the case data meets the data screening criteria;

[0099] If so, the preselected patients corresponding to the case data were selected as target patients and the local treatment and compression data for cervical lymphatic leakage were obtained;

[0100] Among them, the learning data acquisition subsystem determines whether the case data meets the data screening criteria, including:

[0101] Traverse the case sub-data of each data type in the case data in turn, and take the case sub-data currently being traversed as the target case sub-data;

[0102] When the data type is post-operative data of cleaning, obtain a standard condition item description scoring template; the standard condition item description scoring template includes: one-to-one corresponding standard condition items and standard condition item standard description factors; wherein the standard condition item is: the post-operative condition to be described, such as: "whether to include neck CT images", "basic information of the patient"; the standard condition item standard description factor is: a feature description vector of the standard condition item, the feature description is, for example: the neck CT image in the summary list of the post-operative condition of the pre-selected patient is marked as "yes", and the source library of the neck CT image is associated;

[0103] Parse the target case sub-data to obtain the record condition item and the record condition item description factor; wherein the record condition item is: the post-operative condition information item of the pre-selected patient recorded in the target case sub-data; the record condition item description factor is: the feature description vector of the record condition item;

[0104] Obtaining a missing situation item corresponding to the standard situation item of the recorded situation item; wherein the missing situation item is: a standard situation item that does not exist in the recorded situation item;

[0105] Obtaining a missing situation scoring library; wherein the missing situation scoring library includes multiple scoring rules corresponding to missing situations, and the scoring rules corresponding to the missing situations are preset manually, for example, if the neck CT image information item is missing, the score for determining the missing situation is 40, and if the patient's age information item is missing, the score for the corresponding missing situation is 80;

[0106] Determine the missing situation score based on the missing situation score library and the missing situation items;

[0107] Obtaining a matching situation item of the record situation item corresponding to the standard situation item; wherein the matching situation item is: the standard situation item existing in the record situation item;

[0108] Calculate the factor cosine value of the record situation item description factor of the matching situation item and the corresponding matching standard situation item and the standard situation item specification description factor; wherein the factor cosine value is: the cosine similarity of the record situation item description factor and the standard situation item specification description factor;

[0109] Obtaining a matching situation scoring library; wherein the matching situation scoring library includes a quantization rule for converting cosine similarities corresponding to different matching situations into matching situation scores, in accordance with the rule that the greater the cosine similarity, the higher the matching situation score, and the specific conversion ratio relationship is manually set according to the type of post-dissection situation information corresponding to the matching situation;

[0110] Determine a matching situation score according to the matching situation score library and the factor cosine value;

[0111] Sum the missing situation score and the matching situation score to obtain the first score value of the target case sub-data;

[0112] When the data type is lymphatic leakage treatment process data, the evaluation record of the treatment process is obtained; wherein the evaluation record includes: the evaluation of the treatment process by the medical staff who performs lymphatic leakage treatment on the pre-selected patient and the feedback evaluation of the pre-selected patient himself;

[0113] Determine the evaluation type of the evaluation record, which includes active evaluation and passive evaluation. Active evaluation refers to the patient's own evaluation of the lymphatic leakage treatment process; passive evaluation refers to the evaluation of the lymphatic leakage treatment process by the medical staff who perform lymphatic leakage treatment on the pre-selected patients.

[0114] Obtain the evaluation record scoring strategy corresponding to the evaluation type; wherein the evaluation record scoring strategy is: a scheme for evaluating the lymphatic leakage treatment process according to the content of the evaluation records of different evaluation types, for example: the active evaluation type determines the patient's discomfort level value according to the discomfort adjectives responded by the patient, and the lower the discomfort level value, the higher the corresponding treatment process score; the passive evaluation type determines the reasonableness level value of the evaluation semantics according to the evaluation semantics of the medical staff and the preset lymphatic leakage treatment knowledge base, and the higher the reasonableness level value, the higher the corresponding treatment process score;

[0115] According to the evaluation record scoring strategy, determine the treatment process scores corresponding to the evaluation records of different evaluation types;

[0116] Calculate the second score value of the target case sub-data according to the preset evaluation weight and treatment process score corresponding to the evaluation type; wherein the second score value is: the result obtained by multiplying the evaluation weight and the corresponding treatment process score and then summing them up, and the evaluation weight corresponding to the active evaluation is less than the evaluation weight corresponding to the passive evaluation;

[0117] When the data type is lymphatic leakage treatment effect data, the effect score of the target case sub-data is obtained and used as the third score value; wherein the effect score is quantitatively obtained based on the effect evaluation semantics and the preset effect evaluation semantics-effect value quantification rule;

[0118] After all case sub-data of each data type in the case data are traversed, the first score value, the second score value and the third score value are cumulatively calculated to obtain the data screening value of the case data;

[0119] If the data screening value is greater than or equal to the preset data screening value threshold, the case data is determined to meet the data screening standard. The preset data screening value threshold is manually pre-set.

[0120] The working principle and beneficial effects of the above technical solution are:

[0121] When obtaining learning data for deep learning models through medical information sources, not all treatment and compression data of patients with local treatment of cervical lymphatic leakage are available, such as: patients’ postoperative information is not detailed, compression strategy is not appropriate, and the effect of local treatment of lymphatic leakage is not good. Therefore, it is necessary to screen the data obtained from medical information sources;

[0122] First, determine the medical project label, search for the lymphatic leakage treatment project label in the medical project label, and obtain the case data under the label; determine the data type of the case data, including: post-operative data of the cleaning, lymphatic leakage treatment process data, and lymphatic leakage treatment effect data, and judge whether the case data meets the data screening criteria, among which,

[0123] Determining whether the case data meets the data screening criteria includes: traversing the case sub-data of each data type in the case data in turn, obtaining the target case sub-data currently being traversed, and when the data type is post-operative situation data after cleaning, introducing a standard situation item description scoring template; the standard situation item description scoring template includes the post-operative situation (standard situation item) to be described and its corresponding standard description feature description vector (standard situation item standard description factor); comparing the record situation item and the standard situation item obtained in the target case sub-data to determine the missing situation item and the matching situation item, and determining the missing situation score for the missing situation item based on the missing situation scoring library; the matching situation score is determined according to the calculated record situation item description factor of the matching situation item and the factor cosine value of the standard situation item standard description factor of the corresponding matching standard situation item and the matching situation scoring library, and the missing situation score and the matching situation score are summed to obtain the first score value of the target case sub-data;

[0124] When the data type is lymphatic leakage treatment process data, the evaluation records in the treatment process are introduced, and the evaluation records include the patient's own active evaluation of the lymphatic leakage treatment process and the medical staff's passive evaluation of the lymphatic leakage treatment process; according to different evaluation types, the corresponding evaluation record scoring strategy is obtained and the treatment process score is scored for the corresponding evaluation record; the preset evaluation weight corresponding to the evaluation type and the treatment process score are multiplied and then summed to obtain the second score value;

[0125] Finally, when the data type is lymphatic leakage treatment effect data, the effect evaluation semantics are quantified to obtain the effect score. The better the treatment effect represented by the effect evaluation semantics, the higher the third score value;

[0126] The first score value, the second score value and the third score value associated with the case data are summed to obtain the data screening value, and the case data that meets the data screening criteria and has a data screening value greater than or equal to a preset data screening value threshold is screened out, thereby improving the screening quality of the learning data.

[0127] In one embodiment, the learning data acquisition subsystem determines the treatment process scores corresponding to the evaluation records of different evaluation types according to the evaluation record scoring strategy, including:

[0128] When the evaluation type is active evaluation, the evaluation record is parsed to obtain the patient evaluation words; wherein the patient evaluation words are obtained by performing word segmentation processing on the patient evaluation record;

[0129] Obtaining a preset discomfort level value determination library; wherein the discomfort level value determination library presets a plurality of one-to-one corresponding target words and discomfort level values, which are pre-configured manually;

[0130] Determine the discomfort level value of the evaluation record of the active evaluation according to the patient's evaluation words and the discomfort level value of the library;

[0131] Based on the conversion relationship between the preset discomfort level value and the treatment process score, the treatment process score is determined; wherein the conversion relationship conforms to an inverse relationship, and the specific proportional relationship is manually set according to the needs;

[0132] When the evaluation type is passive evaluation, the evaluation record is parsed to obtain the evaluation words of the medical staff; wherein the evaluation words of the medical staff are obtained by performing word segmentation processing on the evaluation record of the medical staff;

[0133] Based on the lymphatic leakage treatment knowledge records in the lymphatic leakage treatment knowledge base, a lymphatic leakage treatment rationality analysis model is trained; wherein the lymphatic leakage treatment rationality analysis model is an AI model that uses a neural network model to train the lymphatic leakage treatment knowledge records and outputs a knowledge description rationality value based on the input knowledge description;

[0134] The medical staff evaluation words are characterized to obtain evaluation word features, and the evaluation word features include: the meaning of the lymphatic leakage treatment process evaluated by the evaluation word, the evaluation word meaning, the reverse evaluation word meaning, and the target word meaning between the evaluation word corresponding to the evaluation word meaning and the evaluation word corresponding to the reverse evaluation word meaning; wherein, the lymphatic leakage treatment process meaning is: the meaning of the lymphatic leakage treatment process description word in the medical staff evaluation word; the evaluation word meaning is: the meaning of the medical staff evaluation word that represents the good or bad evaluation, such as: "good", "bad", "right", "wrong", etc.; the reverse evaluation word meaning is: the medical staff evaluation word that represents the good or bad evaluation and is associated with the medical staff evaluation word corresponding to the evaluation word meaning, and the word meaning is opposite to the associated evaluation word meaning, and the association means that there are no other medical staff evaluation words that represent the good or bad evaluation between the two medical staff evaluation words, and the number of words between the two medical staff evaluation words is less than the preset word number threshold; the target word meaning is: the meaning of the medical staff evaluation word between the evaluation word corresponding to the evaluation word meaning and the evaluation word corresponding to the reverse evaluation word meaning;

[0135] The evaluation word features are input into the lymph node leakage treatment rationality analysis model to obtain the rationality value, which is used as the treatment process score.

[0136] The working principle and beneficial effects of the above technical solution are:

[0137] The present invention introduces an evaluation type. When the evaluation type is active evaluation, a discomfort level determination library corresponding to the patient evaluation is introduced. According to the patient evaluation words and the discomfort level determination library, the discomfort level value of the evaluation record of the active evaluation is determined, and the treatment process score is determined according to the inverse conversion relationship between the discomfort level value and the treatment process score. When the evaluation type is passive evaluation, the lymphatic leakage treatment knowledge record in the lymphatic leakage treatment knowledge base is introduced to train the lymphatic leakage treatment rationality analysis model, and the evaluation words of the medical staff are characterized to obtain the lymphatic leakage treatment process word meaning, the evaluation word meaning corresponding to the lymphatic leakage treatment process word meaning, the reverse evaluation word meaning associated with the evaluation word meaning, and the evaluation word corresponding to the evaluation word meaning to the reverse evaluation word meaning. The target word meaning between the evaluation words corresponding to the valence word meaning, this feature extraction setting takes into account the treatment situation in which the medical staff makes negative evaluations and then corrects the treatment when the lymph leakage treatment process is not suitable. It should be noted that individual physiological differences are large, and it is a common situation to make adjustments according to process feedback during pressurization. The present invention takes into account the adjustment situation, and extracts the target word meaning (adjustment situation word meaning) between the evaluation words corresponding to the evaluation word meaning and the evaluation words corresponding to the reverse evaluation word meaning, and jointly inputs them into the lymph leakage treatment rationality analysis model, thereby avoiding the unreasonable treatment process scoring of the lymph leakage treatment process after the negative evaluation words are identified, which leads to the blind elimination of learning data, and improves the learning efficiency.

[0138] The embodiment of the present invention provides an intelligent neck pressure regulating device based on deep learning, further comprising:

[0139] The offset detection subsystem is used to detect the offset of the pressing position during the operation of the intelligent neck pressure regulator;

[0140] Among them, the offset detection subsystem performs pressing position offset detection during the operation of the intelligent neck pressure regulator, including:

[0141] During the operation of the intelligent neck pressure regulator, dynamic information of the intelligent neck pressure regulator is obtained; wherein the dynamic information is obtained according to a gyroscope inside the intelligent neck pressure regulator, and when a sudden change occurs in the gyroscope data, it is determined that the dynamic information is obtained successfully;

[0142] If the dynamic information is successfully acquired, the pressing position offset detection is performed;

[0143] If the dynamic information is successfully acquired, the offset detection subsystem performs a pressing position offset detection, including:

[0144] If the dynamic information is successfully acquired, the temperature sensing data of the pressurized area within the preset time of the intelligent neck pressure regulator is acquired; wherein the preset time is: the preset time before the dynamic information is successfully acquired and the preset time after the dynamic information is successfully acquired, and the preset time is manually preset, for example: 10 seconds;

[0145] Draw a temperature change gradient diagram of different pressurized areas according to the temperature sensing data; wherein the temperature change gradient diagram is: a visual diagram showing the temperature distribution change of the pressurized area within a preset time period;

[0146] Calculate the gradient map similarity of the temperature change gradient map of different pressurized areas; wherein the gradient map similarity is: the similarity of image features;

[0147] If the gradient map similarity is greater than or equal to a preset gradient map similarity threshold, a deviation reminder is issued to the wearer, wherein the preset gradient map similarity threshold is manually preset.

[0148] The working principle and beneficial effects of the above technical solution are:

[0149] The present invention detects dynamic information of the intelligent neck pressure regulator. When the dynamic information is successfully obtained, the temperature sensing data of the pressurized area within a preset time period is introduced, and the temperature sensing data is visualized to obtain a temperature change gradient diagram. Since a general intelligent neck pressure regulator is offset in one direction and the temperature of the pressurized skin area before the offset is higher than the temperature of the pressurized skin area after the offset, when relative movement of the intelligent neck pressure regulator is detected and the temperature change characteristics of different pressurized areas are similar, it indicates that the instrument has offset, and the wearer is promptly reminded to adjust to avoid delays in local pressurization therapy.

[0150] The embodiment of the present invention provides an intelligent neck pressure regulation method based on deep learning, such as Figure 2 As shown, including:

[0151] Step 1: Obtain the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection performed by medical staff on the target patients;

[0152] Step 2: Learning the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model to obtain a simulated pressurization model;

[0153] Step 3: Obtain postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determine simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model;

[0154] Step 4: According to the simulated pressurization parameters and the pressure regulation signal conversion template of the intelligent neck pressure regulator, the pressure regulation signal is determined and intelligent pressure regulation control is performed.

[0155] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An intelligent neck pressure regulating device based on deep learning, characterized in that: include: A learning data acquisition subsystem is used to obtain the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection of target patients by medical staff; A deep learning subsystem, used to learn the local treatment pressurization data of cervical lymphatic leakage based on a deep learning model and obtain a simulated pressurization model; A simulated pressurization parameter determination subsystem is used to obtain postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determine the simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model; The intelligent pressure regulation subsystem is used to determine the pressure regulation signal and perform intelligent pressure regulation control according to the simulated pressurization parameters and the pressure regulation signal conversion template of the intelligent neck pressure regulator.

2. The intelligent neck pressure regulating device based on deep learning according to claim 1, characterized in that: The learning data acquisition subsystem acquires the local treatment and compression data of neck lymph node leakage after lateral neck lymph node dissection of the target patient by medical staff, including: Connect with medical information sources; Through the medical information source, the local treatment and pressurization data of cervical lymph node leakage after lateral cervical lymph node dissection performed by medical staff on the target patient are retrieved; wherein, the local treatment and pressurization data of cervical lymph node leakage include: the postoperative condition information and reference pressure parameters of the target patient.

3. The intelligent neck pressure regulating device based on deep learning according to claim 1, characterized in that: The deep learning subsystem learns the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model to obtain a simulated pressurization model, including: The postoperative condition information of the target patient is used as the input of the ViT model, and the reference pressure parameters are used as the output of the ViT model to train the simulated pressurization model.

4. The intelligent neck pressure regulating device based on deep learning according to claim 1, characterized in that: The simulated pressurization parameter determination subsystem obtains the postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determines the simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model, including: Obtain clinical data and medical imaging data of patients wearing the intelligent neck pressure regulator; Use clinical data and medical imaging data as postoperative condition information for patients wearing the device; The postoperative condition information of the patient wearing the device is used as the input of the simulated pressurization model to obtain the simulated pressurization parameters output by the simulated pressurization model.

5. The intelligent neck pressure regulating device based on deep learning as claimed in claim 2, characterized in that: The learning data acquisition subsystem retrieves the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection of the target patient by medical staff through medical information sources, including: Determine the medical item label of the pre-selected patient through the medical information source; Retrieve lymph node leakage treatment project tags from medical project tags, and obtain case data associated with the successfully retrieved medical project tags; The data types of case data are obtained; the data types include: post-clearance condition data, lymphatic leakage treatment process data, and lymphatic leakage treatment effect data; Determine whether the case data meets the data screening criteria; If so, the preselected patients corresponding to the case data were selected as target patients and the local treatment compression data for cervical lymphatic leakage were obtained.

6. The intelligent neck pressure regulating device based on deep learning according to claim 5, characterized in that: The learning data acquisition subsystem determines whether the case data meets the data screening criteria, including: Traverse the case sub-data of each data type in the case data in turn, and take the case sub-data currently being traversed as the target case sub-data; When the data type is post-operative condition data after cleaning, obtain the standard condition item description scoring template; According to the standard case item description scoring template, obtain the first scoring value of the target case sub-data; When the data type is lymphatic leakage treatment process data, obtain the evaluation records during the treatment process; Determine the evaluation type of the evaluation record, which includes active evaluation and passive evaluation; Get the evaluation record scoring strategy corresponding to the evaluation type; According to the evaluation record scoring strategy, determine the treatment process scores corresponding to the evaluation records of different evaluation types; Calculate the second score value of the target case sub-data according to the preset evaluation weight and treatment process score corresponding to the evaluation type; When the data type is lymphatic leakage treatment effect data, the effect score of the target case sub-data is obtained and used as the third score value; After all case sub-data of each data type in the case data are traversed, the first score value, the second score value and the third score value are cumulatively calculated to obtain the data screening value of the case data; If the data screening value is greater than or equal to the preset data screening value threshold, the case data is determined to meet the data screening criteria.

7. The intelligent neck pressure regulating device based on deep learning according to claim 1, characterized in that: Also includes: The offset detection subsystem is used to detect the offset of the pressing position during the operation of the intelligent neck pressure regulator.

8. The intelligent neck pressure regulating device based on deep learning according to claim 7, characterized in that: The offset detection subsystem performs the pressure position offset detection during the operation of the intelligent neck pressure regulator, including: During the operation of the intelligent neck pressure regulator, dynamic information of the intelligent neck pressure regulator is obtained; If the dynamic information is successfully acquired, the pressing position offset detection is performed.

9. The intelligent neck pressure regulating device based on deep learning according to claim 8, characterized in that: If the dynamic information is successfully acquired, the offset detection subsystem performs a press position offset detection, including: If the dynamic information is successfully obtained, the temperature sensing data of the pressurized area within the preset time of the intelligent neck pressure regulator is obtained; Based on the temperature sensing data, draw the temperature change gradient diagram of different pressurized areas; Calculate the gradient map similarity of the temperature change gradient maps of different pressurized areas; If the gradient map similarity is greater than or equal to the preset gradient map similarity threshold, a deviation reminder is issued to the wearer.

10. An intelligent neck pressure regulation method based on deep learning, characterized in that: include: Step 1: Obtain the local treatment and compression data of cervical lymph node leakage after lateral cervical lymph node dissection performed by medical staff on the target patients; Step 2: Learning the local treatment pressurization data of cervical lymphatic leakage based on the deep learning model to obtain a simulated pressurization model; Step 3: Obtain postoperative condition information of the patient wearing the intelligent neck pressure regulator, and determine simulated pressurization parameters according to the postoperative condition information and the simulated pressurization model; Step 4: According to the simulated pressurization parameters and the pressure regulation signal conversion template of the intelligent neck pressure regulator, the pressure regulation signal is determined and intelligent pressure regulation control is performed.

Images