Clinical decision support system for treating pelvic adhesive infertility by combining traditional Chinese medicine and western medicine

Abstract

The invention discloses a clinical decision support system for treating pelvic adhesive infertility through combination of traditional Chinese medicine and western medicine, relates to the technical field of medical health and artificial intelligence crossing, and assists doctors in formulating personalized schemes, covering screening, diagnosis, scheme generation, curative effect evaluation, adaptation to gynecology of hospitals at the second level and above, acquisition of western medicine images, laboratory and gynecology data and diagnosis and treatment of pelvic adhesive infertility. Performing preprocessing filling missing values, marking abnormal values, western medicine disease differentiation, traditional Chinese medicine syndrome differentiation and correlation to form a combined diagnosis report, and marking a basis; matching traditional Chinese and western medicine schemes, and combining individual difference adjustment to generate a scheme containing a period and review time; and review data grading evaluation is carried out, reasons of invalid patients are analyzed, and a scheme is optimized. The system integrates traditional Chinese and western medicine diagnosis and treatment, improves diagnosis accuracy and scheme suitability, identifies drug interaction in advance to guarantee medication safety, dynamically evaluates the curative effect and adjusts the scheme, realizes full-period management, provides reference for doctors, adapts to different medical scenes, and improves the treatment effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the medical and health technology field, and in particular to a clinical decision support system for treating pelvic adhesion infertility in the integrative medicine. BACKGROUND

[0002] As a gynecological difficult disease, the diagnosis and treatment of pelvic adhesion infertility needs to combine the patient's fertility indicators, the precise judgment of anatomical structure and inflammatory state by Western medicine, and the overall conditioning of syndrome constitution by traditional Chinese medicine. There are many limitations in the current clinical diagnosis and treatment process. At the level of Western medicine diagnosis and treatment, the adhesion site and degree are identified by imaging examinations such as pelvic ultrasound, MRI and salpingography, but the interpretation of examination results is easily influenced by the doctor's experience, and different doctors may have different standards for judging "moderate adhesion". At the same time, Western medicine treatment focuses on adhesion release or inflammation control, and lacks dynamic adaptation to individual differences in patients, such as using regular drug doses for patients with liver and kidney dysfunction, which may increase the risk of adverse reactions. In addition, Western medicine lacks long-term management of the disease, and only assesses the efficacy through regular reexamination, which is difficult to adjust the scheme in real time and may lead to prolonged treatment cycle.

[0003] At the level of traditional Chinese medicine diagnosis, pelvic adhesion infertility is often related to blood stasis, cold dampness, damp heat and other syndromes, but the diagnosis process highly depends on the doctor's subjective experience, and there is a lack of quantitative standard for judging the severity of syndromes such as "cold pain in the lower abdomen" and "dark red tongue", which may lead to "same symptom different syndrome" and "same syndrome different treatment". For example, two patients with "lumbosacral pain", one may be diagnosed as qi stagnation and blood stasis, and the other may be diagnosed as cold dampness and stagnation. If the diagnosis is biased, it will lead to the wrong choice of prescription and affect the treatment effect. At the same time, the synergy of traditional Chinese medicine and Western medicine treatment is insufficient, and the interaction between the two is often not evaluated in clinical practice, which may increase the risk of kidney damage and threaten the safety of drug use.

[0004] Most of the existing clinical decision support systems focus on a single medical system (Western medicine or traditional Chinese medicine), and lack the design of deep integration of traditional Chinese and Western medicine. Some western medicine decision systems can only recommend surgery or drugs according to imaging results, and cannot combine traditional Chinese medicine to individualize the adjustment scheme; although the traditional Chinese medicine auxiliary system can provide prescription suggestions, it does not include western medical examination indicators, and it is difficult to balance the single factor of anatomical abnormality and the dual needs of multiple factors and syndrome regulation. In addition, the system data collection dimension is single, and the key factors affecting adhesion progression such as patient reproductive history and living habits are often ignored, resulting in incomplete diagnosis; the efficacy evaluation only focuses on short-term symptom improvement, and is not related to the reproductive outcome, which cannot provide long-term diagnosis and treatment expectations for patients. These problems lead clinicians to spend a lot of time integrating multi-source data and weighing the pros and cons of treatment when developing a combination of traditional Chinese and Western medicine, not only prolonging the diagnosis and treatment time, but also reducing the effectiveness of the scheme due to information omission or experience bias, and there is an urgent need for a clinical decision support system that covers the whole process and deeply integrates traditional Chinese and Western medicine to solve the precision, synergy and safety problems in current diagnosis and treatment. SUMMARY

[0005] The clinical decision support system for treating pelvic adhesion infertility by combining traditional Chinese and Western medicine according to the present application solves the problems mentioned in the prior art.

[0006] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] A clinical decision support system for treating pelvic adhesion infertility by combining traditional Chinese and Western medicine, comprising the following modules:

[0008] A multi-modal data acquisition module is used to acquire multi-dimensional diagnosis and treatment data of patients in traditional Chinese and Western medicine, and supports multi-source data access. The Western medicine data acquisition unit interfaces with the hospital HIS system, imports pelvic imaging examination, laboratory examination, sex hormone, gynecological examination data; the traditional Chinese medicine data acquisition unit inputs traditional Chinese medicine four diagnostic information through a structured questionnaire and a doctor's syndrome input interface, and supports image uploading; the data preprocessing subunit fills in the missing values by using the adjacent case interpolation method, identifies and marks the abnormal values by the 3σ rule, and the data integrity is ≥98%, which provides input for subsequent diagnosis.

[0009] A traditional Chinese and Western medicine dual-dimensional diagnosis module: realizes the joint diagnosis of pelvic adhesion infertility in traditional Chinese and Western medicine. The Western medicine diagnosis unit identifies the adhesion site and adhesion degree based on imaging examination, generates a Western medicine sub-process in combination with inflammatory factor levels and tubal patency; the traditional Chinese medicine syndrome differentiation unit adopts a'syndrome element-syndrome' mapping model, quantitatively scores four diagnostic information, determines the core syndrome by weighted summation, and generates the traditional Chinese medicine syndrome type in combination with the patient's constitution; the diagnosis result integration subunit associates the Western medicine sub-process and the traditional Chinese medicine syndrome type, forms a 'Western medicine sub-process + traditional Chinese medicine syndrome type' joint diagnosis report such as'moderate adhesion + stagnant syndrome of chong and ren for a long time', and marks the diagnosis basis.

[0010] Personalized diagnosis and treatment plan generation module: output of integrative diagnosis and treatment plan. Western medicine scheme generation unit based on western medicine sub-program matching treatment means: mild adhesion recommended drug treatment, moderate adhesion recommended physical therapy + drug assistance, severe adhesion recommended surgical treatment; TCM scheme generation unit based on TCM syndrome type matching principle and prescription: TCM scheme generation unit based on TCM syndrome type and menstrual period matching principle and prescription: select self-made Bai Lian San Jie Tang as the main prescription for the disease, yin long period after menstrual period, treatment should focus on tonifying yin to balance yin and yang; Yin and Yang transformation period, menstrual period and ovulation period, should pay attention to promoting blood circulation and regulating menstruation, which is beneficial to the discharge of menstrual blood and ovum; Yang long period, premenstrual period, treatment should be warm and tonifying yang qi to promote menstruation, select Bai Lian San Jie Tang and Yishen Guchong Tang, and recommend Chinese medicine sealing therapy and Chinese medicine retention enema therapy; The scheme optimization subunit adjusts the scheme according to the individual differences of the patient, such as reducing the drug dose for patients with liver and kidney dysfunction, replacing the sensitizing drug for patients with allergic constitution, and generating a complete scheme including treatment cycle, review time and matters needing attention.

[0011] Therapeutic effect dynamic evaluation module: real-time monitoring of diagnosis and treatment effect and adjustment of scheme. Evaluation index collection unit regularly obtains patient review data: western medicine index includes changes in adhesion degree, inflammatory factor level, improvement of tubal patency, TCM index includes changes in syndrome score, improvement of tongue and pulse, reproductive related index includes ovulation monitoring and pregnancy situation; Evaluation calculation subunit compares the changes of indexes before and after treatment, and classifies them as "markedly effective (adhesion degree reduced by 1 level + syndrome score decreased by ≥50%), effective (adhesion degree not aggravated + syndrome score decreased by 20%-49%), and ineffective (adhesion degree aggravated or syndrome score decreased by <20%)" ; The scheme adjustment subunit re-calls the diagnosis module to analyze the reasons (such as insufficient drug dose, change of syndrome type) for the ineffective patients, generates an optimized scheme (such as adjusting prescription addition and subtraction), and forms a "diagnosis-treatment-evaluation-adjustment" closed loop.

[0012] Further, it also includes a TCM syndrome quantitative scoring unit, which calculates the total score of TCM syndromes of patients through weighted calculation, provides quantitative basis for TCM syndrome type determination, and reduces subjective syndrome deviation. The total score of TCM syndromes is calculated by the following formula:

[0013] In the formula, S is the total score of TCM syndromes (dimensionless, range 0-60 points, the higher the score, the more significant the syndrome), n is the number of TCM syndrome items included in the score (n=20 in this system, including core syndromes such as cold pain in the lower abdomen and lumbosacral pain), w i is the weight of the ith syndrome item (dimensionless, set according to the contribution of the syndrome to the syndrome type, such as "abdominal pain" in blood stasis syndrome w i =0.8, "white leukorrhea" w i =0.3, the weight sum is 10), s iThe severity score of the ith syndrome item (dimensionless, graded as "none = 0 points, mild = 1 point, moderate = 2 points, severe = 3 points").

[0014] Further, it also includes a diagnosis and treatment scheme adaptability evaluation unit, which filters the optimal scheme by calculating the matching degree of the scheme and the individual characteristics of the patient, and reduces the treatment risk. The formula for calculating the scheme adaptability index is:

[0015] In the formula, M is the scheme adaptability index (dimensionless, range 0-1, M≥0.8 is highly adaptable, 0.6≤M<0.8 is basically adaptable, and M<0.6 is not adaptable), k is the number of evaluation dimensions (in this system, k=5, including age, liver and kidney function, allergy history, fertility demand, and syndrome stability), x j x_j is the actual characteristic value of the jth dimension (dimensionless, such as age x_1=30 years old, normalized to the [0,1] interval as 0.5; normal liver and kidney function x_2=1, abnormal x_2=0), y j y_j is the recommended characteristic value of the jth dimension (dimensionless, such as a surgical scheme recommending an age of 25-35 years, after normalization y1=0.5; requiring normal liver and kidney function y2=1).

[0016] Further, the multi-modal data acquisition module also includes a patient age, disease course, and fertility history and lifestyle acquisition sub-unit, which acquires the patient's past fertility history, such as the number of abortions, ectopic pregnancy history, past assisted reproduction treatment history, and lifestyle habits such as work and rest routines, dietary preferences, exercise frequency, and stress data, through a structured form. The sub-unit supports automatic data classification and labeling, such as marking "abortion frequency ≥2 times" as "high risk factor" and "daily sleep <6 hours" as "unhealthy lifestyle"; at the same time, it interfaces with wearable device interfaces to obtain the patient's sleep duration and exercise step count data for the past 3 months, with a sampling frequency of 1 time / day and a data update cycle of 1 week. The fertility history and lifestyle data supplemented by this sub-unit can assist the diagnosis module in determining the cause and provide lifestyle intervention suggestions for the scheme generation module, improving the comprehensiveness of the diagnosis and treatment scheme.

[0017] Further, the Chinese and Western medicine dual-dimensional diagnosis module also includes an adhesion risk prediction sub-unit, which builds a machine learning prediction model based on the patient's historical data to predict the patient's adhesion aggravation risk. The model input features include: past adhesion degree, inflammatory factor baseline level, TCM syndrome type, and lifestyle; and the output risk level.

[0018] Further, the individualized diagnosis and treatment plan generation module also includes a drug interaction detection subunit, which has a drug interaction database built in, to detect whether there are contraindications or adverse reactions in the plan. The detection process is: search the database, judge whether there is interaction; if there is risk, generate a risk level, mild, moderate, severe, and a replacement suggestion.

[0019] Further, the efficacy dynamic evaluation module also includes a fertility probability prediction subunit, which predicts the patient's natural conception probability within 1 year by analyzing the changes in indicators before and after treatment, providing a reference for treatment plan adjustment. The fertility probability prediction formula is: P = a x A + b x B + g x C

[0020] In the formula, P is the natural conception probability within 1 year (dimensionless, range 0-1, P≥0.6 is high probability, 0.3≤P<0.6 is medium probability, P<0.3 is low probability), a is the adhesion improvement weight (dimensionless, value 0.4, a = 0.4 for 1 level of adhesion reduction, a = 0.2 for no improvement, a = 0 for aggravation), A is the adhesion improvement score (dimensionless, range 0-1, mild improvement A = 0.5, moderate improvement A = 0.8, significant improvement A = 1), b is the ovarian function weight (dimensionless, value 0.35), B is the ovarian function score (dimensionless, range 0-1, normal ovulation + normal E2 level B = 1, irregular ovulation + low E2 B = 0.5, anovulation B = 0), g is the endometrial condition weight (dimensionless, value 0.25), and C is the endometrial score (dimensionless, range 0-1, thickness ≥8mm + rich blood flow C = 1, thickness 6-7mm + general blood flow C = 0.6, thickness <6mm + poor blood flow C = 0.3).

[0021] Further, it also includes a case knowledge base module, which stores clinical cases of treating pelvic adhesion infertility with integrated traditional Chinese and Western medicine, including patient basic information, diagnosis results, diagnosis and treatment plans, and efficacy data. The cases are desensitized, private information such as names and ID numbers is removed, and core features are labeled. The module supports case retrieval function, doctors can filter similar cases by "age range, adhesion degree, TCM syndrome type, treatment method" multidimensional, view the plan execution effect and adjustment process of similar cases; at the same time support case update, maintain case quality. The case knowledge base provides clinical reference for doctors, especially for young doctors, who can improve the accuracy of plan development and shorten the diagnosis and treatment decision-making time by referring to similar cases.

[0022] Further, it also includes a patient follow-up management module, which automatically generates a follow-up plan according to the diagnosis and treatment scheme, such as follow-up after the end of the first, second and third menstrual cycle, follow-up once a month after pregnancy, and follow-up by telephone. The follow-up data acquisition unit supports doctors to enter the symptom changes and examination results during follow-up, automatically compares the pre-treatment data to generate a follow-up report; for pregnant patients, additional gestational indicators are collected to provide advice on tocolysis; for non-pregnant patients, the reasons for non-pregnancy are analyzed, and the treatment effect is evaluated to adjust the scheme. The follow-up management module realizes the whole cycle tracking of patients, avoids the interruption of efficacy monitoring caused by follow-up omission, and improves the success rate of treatment.

[0023] Further, it also includes a permission management and data security module, which adopts a role-based access control mechanism to divide three roles: doctors, department administrators and system administrators. The module supports dual authentication of user identity, patient data transmission adopts TLS1.3 encryption protocol, the database sets access logs according to the requirements of the Personal Information Protection Law and the Medical Data Security Guidelines, and the retention period is greater than or equal to 3 years; at the same time, data backup is supported, and the backup data is stored in a remote server to prevent data loss. This module ensures the privacy and data security of patients, meets the data compliance requirements of the medical industry, and reduces the risk of data leakage.

[0024] Compared with the existing technology, the beneficial effects of the present application are:

[0025] The clinical decision support system for treating pelvic adhesion infertility by combining traditional Chinese and Western medicine significantly improves the limitations of traditional abnormal protection methods and brings many outstanding advantages.

[0026] Through the system design of deep integration of traditional Chinese and Western medicine, the present application breaks through the limitations of current diagnosis and treatment of pelvic adhesion infertility, improves the accuracy, safety and synergy of the diagnosis and treatment scheme, and provides full-process decision support for clinicians. In the data acquisition and diagnosis link, the multi-modal data acquisition module integrates Western imaging, laboratory examination and TCM four diagnostic information, supplements key data such as reproductive history and living habits, and ensures the integrity and reliability of input combined with data preprocessing; the traditional Chinese and Western medicine dual-dimension diagnosis module forms a combined diagnosis of "anatomical abnormalities + syndrome constitution" through the correlation of Western medicine sub-process and TCM syndrome type, avoids the one-sidedness of a single medical system, and introduces adhesion risk prediction to help doctors intervene in the disease progression in advance and reduce missed diagnosis and misdiagnosis.

[0027] In the diagnosis and treatment scheme generation link, the individualized scheme module matches the traditional Chinese and Western medicine treatment methods based on the combined diagnosis results, adjusts the scheme considering disease treatment and individual tolerance; the drug interaction detection subunit identifies the risk of drug combination in advance to avoid adverse reactions and ensure medication safety; the scheme adaptability evaluation selects the optimal scheme by quantifying the matching degree, reduces the subjective experience deviation of doctors, improves the adaptability of the scheme to patients, and shortens the treatment cycle.

[0028] In the efficacy evaluation and long-term management link, the efficacy dynamic evaluation module monitors the changes of Chinese and Western medicine indicators in real time, evaluates according to the "marked effect-effective-ineffective" classification, automatically adjusts the scheme for ineffective patients, and forms a diagnosis and treatment closed loop; the fertility probability prediction provides the patient with a reference for expected conception, helping doctors and patients to jointly develop a reasonable diagnosis and treatment goal; the follow-up management module realizes the tracking of patients throughout the cycle, avoids the interruption of efficacy monitoring caused by missed follow-up, and improves the success rate of treatment. In addition, the case knowledge base provides similar case references for young doctors, the permission management and data security module guarantees patient privacy, and meets the medical data compliance requirements.

[0029] Overall, the present application promotes the diagnosis and treatment of pelvic adhesion infertility from "experience-driven" to "data-driven", from "single system" to "integration of traditional Chinese and Western medicine", and from "static scheme" to "dynamic adjustment", which not only improves the efficiency of clinical diagnosis and treatment, but also guarantees the safety and effectiveness of treatment, and provides an intelligent solution for gynecological diagnosis and treatment of traditional Chinese and Western medicine. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 The schematic diagram of the clinical decision support system for treating pelvic adhesion infertility by combining traditional Chinese and Western medicine according to the present application;

[0031] Figure 2 The core effect comparison chart of traditional Chinese and Western medicine combined diagnosis and treatment and traditional single diagnosis and treatment mode;

[0032] Figure 3 The trend chart of traditional Chinese medicine syndrome score and Western medicine inflammation factor level change;

[0033] Figure 4 The diagnosis and treatment scheme composition and fertility probability prediction chart of patients with different adhesion degrees. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0035] In the description of the present application, it is understood that the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0036] In addition, the terms "first", "second" are only for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited. In addition, the terms "mounting", "connecting", "connecting" should be broadly understood, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium, or it can be the communication between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances, and the present application will be further described in detail below with reference to the drawings.

[0037] Reference Figures 1 to 4 A clinical decision support system for treating pelvic adhesion infertility by combining traditional Chinese and Western medicine, comprising the following modules:

[0038] Multi-modal data acquisition module: for acquiring multi-dimensional diagnosis and treatment data of patients in traditional Chinese and Western medicine, supporting multi-source data access. The Western medicine data acquisition unit interfaces with the hospital HIS system to automatically import pelvic imaging examination (ultrasound, MRI, layer thickness ≤3mm, resolution ≥512x512 pixels), laboratory examination (blood routine, inflammatory factors, sex hormones FSH, LH, E2, etc.), gynecological examination (uterine position, adnexal tenderness, cervical secretion characteristics) data; The traditional Chinese medicine data acquisition unit inputs traditional Chinese medicine four diagnostic information through a structured questionnaire (such as medical history, disease course, cold pain in the lower abdomen, lumbosacral pain, accompanied by amount and shape of leukorrhea, color, tongue appearance (tongue, tongue fur, tongue body), pulse appearance (such as stringy pulse, dry pulse)) and doctor's syndrome input interface; The data preprocessing subunit fills in the missing values (such as part of the traditional Chinese medicine syndromes not recorded) by using the adjacent case interpolation method, identifies and marks the abnormal values (such as the detection value of inflammatory factors exceeding 3 times the reference range) by 3σ rule, and the data integrity is ≥98%, providing input for subsequent diagnosis.

[0039] The dual-dimension diagnosis module of traditional Chinese and Western medicine: to realize the combined diagnosis of pelvic adhesion infertility in traditional Chinese and Western medicine. The Western medicine diagnosis unit identifies the adhesion site (uterus-tubal adhesion, ovarian-tubal adhesion, pelvic peritoneal adhesion) and adhesion degree (mild: adhesion range <1 / 3 of the pelvic cavity, moderate: 1 / 3-2 / 3 of the pelvic cavity, severe: >2 / 3 of the pelvic cavity) based on imaging examination, combined with the level of inflammatory factors (IL-6 > 10 pg / mL indicating active inflammation), and tubal patency (hysterosalpingography results) to generate a Western medicine sub-process (mild, moderate, moderate). The TCM syndrome differentiation unit adopts a "syndrome element-syndrome type" mapping model, quantitatively scores four diagnostic information (e.g. 3 points for cold pain in the lower abdomen, 2 points for lumbosacral pain), determines the core syndromes (blood stasis, cold-damp stagnation, qi stagnation and blood stasis, damp-heat and blood stasis) through weighted summation, and generates TCM syndrome types combined with patient's constitution (qi deficiency, yang deficiency, phlegm-dampness). The diagnosis result integration sub-unit associates the Western medicine sub-process with the TCM syndrome type to form a combined diagnosis report of "Western medicine sub-process + TCM syndrome type" (e.g. "moderate adhesion + stagnation of Chong and Ren channels syndrome for long-term"), and marks the diagnosis basis.

[0040] The personalized diagnosis and treatment scheme generation module: according to the combined diagnosis results, output the diagnosis and treatment scheme of traditional Chinese and Western medicine. The Western medicine scheme generation unit matches the treatment means based on the Western medicine sub-process: mild adhesion recommends drug treatment, moderate adhesion recommends physical therapy + drug assistance, and severe adhesion recommends surgical treatment. The TCM scheme generation unit matches the therapeutic principle and prescription based on the TCM syndrome type: long-term yin refers to the late menstrual period, the treatment focuses on tonifying yin to balance yin and yang, and selects the modified Bai Lian San Jie Decoction and Zhi He Decoction (Wang Chengrong's experience recipe, composition: Gouqi Zi 20g, Tusizi 20g, Buguzhi 20g, Danggui 10g, Chuanqiong 15g, Jixueteng 30g, Taoren 10g, Guizhi 10g, and adjusted according to the accompanying symptoms). The transformation of yin and yang refers to the menstrual and ovulation periods, which should focus on promoting blood circulation and regulating menstruation to facilitate the discharge of menstrual blood and ovum, and selects the modified Bai Lian San Jie Decoction and Tao Hong Si Wu Decoction (composition: Taoren 10g, Honghua 10g, Shudihuang 20g, Danggui 10g, Baishao 30g, Chuanqiong 15g, Niuxi 10g, Xiangfu 15g, and adjusted according to the accompanying symptoms). The long-term yang refers to the premenstrual period, which should be treated by warming and tonifying yang qi to promote menstruation, and selects the modified Bai Lian San Jie Decoction and Yishen Guchong Decoction (Wang Chengrong's experience recipe, composition: Duzhong 15g, Xuduan 15g, Tusizi 20g, Ruzhizong 20g, Baishao 30g, Ganzao 20g, Dangshen 30g, and adjusted according to the accompanying symptoms), and recommends traditional Chinese medicine sealing therapy and traditional Chinese medicine retention enema therapy. The scheme optimization sub-unit adjusts the scheme according to individual differences (age, reproductive history, drug allergy history, liver and kidney function), such as reducing the drug dose for patients with liver and kidney dysfunction (adjusting according to creatinine clearance rate x 0.8), and replacing the sensitizing drugs for allergic constitution, to generate a complete scheme including the treatment cycle (1-3 menstrual cycles), review time, and precautions.

[0041] The curative effect dynamic evaluation module: real-time monitoring of diagnosis and treatment effect and adjusting the scheme. The evaluation index acquisition unit regularly acquires patient review data: Western medicine indexes (adhesion degree change in imaging, inflammatory factor level, improvement of oviduct patency), traditional Chinese medicine indexes (syndrome score change, tongue and pulse improvement), and fertility-related indexes (ovulation monitoring, endometrial thickness (≥8mm is suitable for implantation thickness)); the evaluation calculation subunit compares the index changes before and after treatment, and classifies them according to “marked effect (adhesion degree reduced by 1 level + syndrome score reduced by ≥50%), effectiveness (adhesion degree not aggravated + syndrome score reduced by 20%-49%), and invalidity (adhesion degree aggravated or syndrome score reduced by <20%)”; the scheme adjustment subunit re-calls the diagnosis module to analyze the reasons (such as insufficient drug dosage, syndrome type change) for the invalid patients, generates an optimized scheme (adjusts the prescription), and forms a “diagnosis-treatment-evaluation-adjustment” closed loop.

[0042] In the present application, a traditional Chinese medicine syndrome quantitative scoring unit is also included, which provides quantitative basis for traditional Chinese medicine syndrome type determination by weighted calculation of the total score of the patient's traditional Chinese medicine syndrome, and reduces subjective syndrome differentiation deviation. The total score of the traditional Chinese medicine syndrome is calculated according to the following formula:

[0043] In the formula, S is the total score of the traditional Chinese medicine syndrome (dimensionless, range 0-60 points, and the higher the score, the more significant the syndrome), n is the number of traditional Chinese medicine syndrome items included in the scoring (n=20 in the present system, including core syndromes such as cold pain in the lower abdomen and lumbosacral pain), w i is the weight of the i-th syndrome item (dimensionless, set according to the contribution of the syndrome to the syndrome type, such as “stabbing pain in the lower abdomen” in blood stasis syndrome w i =0.8, “white leukorrhea” w i =0.3, and the weight sum is 10), s i is the severity score of the i-th syndrome item (dimensionless, classified as “none=0 points, mild=1 point, moderate=2 points, and severe=3 points”), for example, a patient has “stabbing pain in the lower abdomen (severe, s i =3, w i =0.8), lumbosacral pain (moderate, s i =2, w i =0.6), dark red tongue (moderate, s i =2, w i =0.5), and no other syndromes, then S=0.8×3+0.6×2+0.5×2+0 for the remaining 17 items=2.4+1.2+1=4.6 points, combined with other syndrome scores, if the total score is ≥15 points, it indicates a high possibility of blood stasis syndrome, providing objective data support for syndrome type determination.

[0044] In the present application, a diagnosis and treatment scheme adaptability evaluation unit is also included, which filters the optimal scheme by calculating the matching degree of the scheme and the individual characteristics of the patient, and reduces the treatment risk. The scheme adaptability index calculation formula is:

[0045] In the formula, M is a regimen adaptability index (dimensionless, range 0-1, M>0.8 is highly adaptable, 0.6 j is the actual characteristic value of the patient in the jth dimension (dimensionless, such as age x_1=30 years old, normalized to the [0,1] interval is 0.5; normal liver and kidney function x_2=1, abnormal x_2=0), y j is the recommended characteristic value of the regimen in the jth dimension (dimensionless, such as a surgical regimen recommended age 25-35 years old, after normalization y1=0.5; requires normal liver and kidney function y2=1). For example, a 30-year-old patient (x1=0.5), normal liver and kidney function (x2=1), no history of allergies (x3=1), urgent fertility needs (x4=1), stable syndrome type (x5=1), a surgical regimen y1=0.5, y2=1, y3=1, y4=1, y5=1, then M=1-(|0.5-0.5|+|1-1|+|1-1|+|1-1|+|1-1|) / 5=1-0=1 (highly adaptable), the regimen can be recommended; if the patient has abnormal liver and kidney function (x_2=0), then M=1-(0+|0-1|+0+0+0) / 5=0.8 (basically adaptable), the drug dose in the regimen needs to be adjusted.

[0046] In the present application, the multi-modal data acquisition module further includes a patient age, disease course, and fertility history and lifestyle acquisition subunit. The subunit obtains patient past fertility history (such as the number of miscarriages, ectopic pregnancy history, and past assisted reproductive treatment history), lifestyle (such as work and rest regularity (daily sleep duration >7 hours is regular), dietary preference (whether likes to eat cold and spicy), exercise frequency (more than 3 times a week is regular), and mental stress (classified as “none, mild, moderate, and severe”)), and data. The subunit supports automatic data classification and labeling, such as marking “miscarriage frequency >2 times” as “high risk factor” and “daily sleep <6 hours” as “unhealthy lifestyle”. The fertility history and lifestyle data supplemented by the subunit can assist the diagnosis module in determining the cause (such as multiple miscarriages that may lead to increased adhesion) and provide lifestyle intervention suggestions (such as recommending warm diet for those who like to eat cold) for the regimen generation module, improving the comprehensiveness of the diagnosis and treatment regimen.

[0047] In the present application, the double-dimension diagnosis module of traditional Chinese and Western medicine further includes an adhesion risk prediction subunit, which constructs a machine learning prediction model (using a random forest algorithm, accuracy ≥ 92%) based on patient historical data to predict the risk of adhesion aggravation. The model input features include: past adhesion degree, inflammatory factor baseline level, TCM syndrome type (such as stagnation of blood in the lower energizer), and lifestyle (such as prolonged sitting); the output risk level is low risk (< 30%), medium risk (30%-60%), and high risk (> 60%). For example, a patient with past moderate adhesion (feature value 1), IL-6 = 12 pg / mL (feature value 0.8), damp-heat and blood stasis syndrome (feature value 1), and more than 8 hours of sitting per day (feature value 0.9), the model output risk level is high risk (65%), and the diagnosis report simultaneously prompts "high risk of adhesion aggravation, suggest shortening the review interval (from 4 weeks to 2 weeks)", helping doctors to intervene in advance and reduce the risk of disease progression.

[0048] In the present application, the personalized diagnosis and treatment scheme generation module further includes a drug interaction detection subunit, which has a built-in drug interaction database to detect whether there is a contraindication or adverse reaction between the drugs in the scheme. The detection process is: 1. Extract the drugs in the scheme; 2. Search the database to determine whether there is an interaction; 3. If there is a risk, generate a risk level (mild, moderate, severe) and a substitution suggestion to avoid drug adverse reactions and ensure medication safety.

[0049] In the present application, the efficacy dynamic evaluation module further includes a fertility probability prediction subunit, which predicts the natural conception probability of patients within 1 year by analyzing the changes in indicators before and after treatment, providing a reference for treatment scheme adjustment. The fertility probability prediction formula is: P = α × A + β × B + γ × C

[0050] In the formula, P is the probability of natural conception within 1 year (dimensionless, range 0-1, P≥0.6 is high probability, 0.3≤P<0.6 is medium probability, and P<0.3 is low probability), a is the adhesion improvement weight (dimensionless, value 0.4, a=0.4 for 1 level of adhesion reduction, a=0.2 for no improvement, and a=0 for aggravation), A is the adhesion improvement score (dimensionless, range 0-1, A=0.5 for mild improvement, A=0.8 for moderate improvement, and A=1 for significant improvement), β is the ovarian function weight (dimensionless, value 0.35), B is the ovarian function score (dimensionless, range 0-1, B=1 for normal ovulation+E2 level, B=0.5 for irregular ovulation+E2 low, and B=0 for anovulation), γ is the endometrial condition weight (dimensionless, value 0.25), and C is the endometrial score (dimensionless, range 0-1, C=1 for thickness≥8mm+rich blood flow, C=0.6 for thickness 6-7mm+general blood flow, and C=0.3 for thickness<6mm+poor blood flow). For example, for a patient with significant adhesion improvement (A=1, a=0.4), normal ovarian function (B=1, β=0.35), and optimal endometrial condition (C=1, γ=0.25), P=0.4×1+0.35×1+0.25×1=1 (high probability), indicating that the current regimen is effective and can be continued; for a patient with mild adhesion improvement (A=0.5, a=0.4), general ovarian function (B=0.5, β=0.35), and general endometrial condition (C=0.6, γ=0.25), P=0.4×0.5+0.35×0.5+0.25×0.6=0.2+0.175+0.15=0.525 (medium probability), suggesting that the regimen be optimized (e.g., increase the dosage of ovulation-promoting drugs).

[0051] In the present application, a case knowledge base module is also included, which stores clinical cases of treating pelvic adhesion-induced infertility by combining traditional Chinese and Western medicine (including patient basic information, diagnosis results, diagnosis and treatment regimen, and efficacy data), the cases are desensitized (remove privacy information such as name and ID number) and labeled with core features (such as “32 years old+moderate adhesion+Qi stagnation and blood stasis syndrome+laparoscopic surgery+Bailian Sanjie Decoction”). The module supports case retrieval function, doctors can filter similar cases (similarity≥80% is highly similar) according to “age range, adhesion degree, TCM syndrome type, and treatment method”, and view the regimen execution effect and adjustment process of the similar cases; meanwhile, the module supports case updating, which is stored in the database after expert review to maintain the quality of the cases. The case knowledge base provides clinical reference for doctors, especially young doctors, who can improve the accuracy of regimen formulation and shorten the diagnosis and treatment decision-making time by referring to similar cases.

[0052] In the present application, a patient follow-up management module is also included, which automatically generates a follow-up plan (such as follow-up after the end of the first, second, and third menstrual cycle, and follow-up once a month after pregnancy) according to the diagnosis and treatment plan, and realizes follow-up through telephone. The follow-up data acquisition unit supports doctors to input the changes in symptoms and examination results during follow-up, and automatically compares the pre-treatment data to generate a follow-up report; for pregnant patients, additional gestational indicators (such as blood HCG doubling, B-ultrasound gestational sac position) are collected, and gestational care suggestions (such as avoiding fatigue and supplementing folic acid) are provided; for non-pregnant patients, the reasons for non-pregnancy (such as ovulation disorders and thin endometrium) are analyzed, and the treatment plan is adjusted according to the efficacy evaluation. The follow-up management module realizes the tracking of patients throughout the cycle, avoids the interruption of efficacy monitoring caused by missed follow-up, and improves the success rate of treatment.

[0053] In the present application, a permission management and data security module is also included, which adopts a role-based access control (RBAC) mechanism to divide three types of roles: doctors (who can view their own patient data and develop plans), department administrators (who can view department patient statistical data and audit cases), and system administrators (who are only responsible for system maintenance and have no data access rights). The module supports dual authentication of user identity (account password + mobile phone verification code), patient data transmission adopts TLS1.3 encryption protocol, the database sets access logs (records access users, time, and operation content) according to the requirements of the Personal Information Protection Law and the Medical Data Security Guidelines, and the retention period is ≥3 years; at the same time, data backup (daily incremental backup + weekly full backup) is supported, and backup data is stored in an off-site server to prevent data loss. This module ensures patient privacy and data security, complies with medical industry data compliance requirements, and reduces the risk of data leakage.

[0054] Example 1

[0055] Secondary hospital gynecological clinic (moderate adhesion + stasis in the chong and ren channels)

[0056] This embodiment is applied to a 32-year-old patient with pelvic adhesion infertility (unpregnant for 3 years after marriage, with a history of 1 induced abortion) treated in the gynecological clinic of a certain secondary hospital. The clinical decision support system for treating pelvic adhesion infertility by combining traditional Chinese and Western medicine is used to solve the problem of moderate adhesion (uterine-tubal adhesion, range 1 / 3-2 / 3 pelvic cavity) and damp-heat stasis syndrome. The system data processing accuracy is 99.2%, the plan matching degree is 96%, the diagnosis and treatment suggestion response delay is 2.8 seconds, and it is suitable for the gynecological clinic scene of primary hospitals.

[0057] I. System deployment and modules

[0058] Application of multi-modal data acquisition module: Western medicine data acquisition unit interfaces with hospital HIS system, automatically imports pelvic MRI (layer thickness 2 mm, resolution 512 x 512 pixels, suggests moderate adhesion of uterus-tubal), laboratory examination (IL-6 = 16 pg / mL > 10 pg / mL, TNF-α = 8 pg / mL, sex hormone: FSH = 6.5 mIU / mL, LH = 5.8 mIU / mL, E2 = 350 pg / mL, detection error 4%), gynecological examination (retroverted uterus, light tenderness of adnexa, large amount of yellow cervical mucus); TCM data acquisition unit collects four diagnostic information through structured questionnaire: burning pain in the lower abdomen (heavy, 3 points), lumbosacral pain (moderate, 2 points), yellow and thick leukorrhea (heavy, 3 points), dark red tongue with yellow and greasy fur (moderate, 2 points), stringy pulse (moderate, 2 points), uploads tongue image (sampling frequency 200 Hz); data preprocessing subunit finds that “previous assisted reproduction history” is missing, uses adjacent cases (31-year-old patient with moderate adhesion) to interpolate and fill in “none”, IL-6 = 16 pg / mL does not exceed 3 times the reference range (reference value 0-10 pg / mL, 3 times is 30 pg / mL), no need to mark abnormality, data integrity 98.5%. Western and TCM dual-dimension diagnosis module: Western medicine diagnosis unit combines MRI and laboratory data, determines that the adhesion site is uterus-tubal, the adhesion degree is moderate, IL-6 elevation suggests active inflammation, tubal imaging shows left side unsmooth and right side smooth, generates western medicine sub-process “moderate adhesion”; TCM syndrome differentiation unit uses “syndrome element-syndrome type” mapping model to quantitatively score four diagnostic information, through formula Calculate syndrome total score: burning pain in the lower abdomen (w i = 0.9, s i = 3, count 2.7), lumbosacral pain (w i = 0.7, s i = 2, count 1.4), yellow and thick leukorrhea (w i = 0.8, s i = 3, count 2.4), dark red tongue with yellow and greasy fur (w i = 0.6, s i = 2, count 1.2), stringy pulse (w i = 0.5, s i=2, total 1.0), the remaining 15 symptoms are not present, S = 2.7 + 1.4 + 2.4 + 1.2 + 1.0 = 8.7 points. Combined with the patient's phlegm-dampness constitution, the TCM syndrome type "damp-heat stagnation syndrome" is generated; the diagnosis result integration subunit forms a joint report of "inflammation-dominant type + damp-heat stagnation syndrome", and the basis for the label is "MRI uterine-fallopian tube moderate adhesion + IL-6 = 16pg / mL + lower abdominal burning pain + dark red tongue with yellow greasy coating → moderate adhesion + Chong and Ren stagnation syndrome Yin period"; the adhesion risk prediction subunit inputs the previous moderate adhesion (1), IL-6 = 16pg / mL (0.9), damp-heat stagnation syndrome (1), sitting for 7 hours a day (0.8), and the random forest model outputs the risk level medium risk (52%), suggesting "it is recommended to have a follow-up examination once every 4 weeks". Personalized treatment plan generation module: The Western medicine plan generation unit is based on inflammation-dominant type, recommending physical therapy (traditional Chinese medicine herbal poultice therapy + traditional Chinese medicine retention enema therapy), and suggesting surgical treatment if necessary; The traditional Chinese medicine plan generation unit is based on stagnation of the Chong and Ren meridians, combined with the patient's menstrual cycle, adopting the method of clearing heat, activating blood and removing blood stasis, with the following prescription: Bailian Sanjie Decoction (Hedyotis diffusa 30g, Scutellaria barbata 30g, Gleditsia sinensis 10g, Epimedium brevicornu 15g, Curculigo orchioides 15g, Polyporus umbellatus 20g, Curcuma zedoaria 10g, Eupolyphaga sinensis 10g, Luffa cylindrica 15g, Citrus reticulata 10g), the dosage is adjusted according to the patient's weight of 60kg (60×0.1g=6g, peach kernel and safflower are increased to 10g and 8g respectively), and traditional Chinese medicine herbal poultice therapy is recommended (prescription No. 2, ingredients: Artemisia argyi, Spatholobus suberectus, Clematis armandii, Carthamus tinctorius, Pulsatilla chinensis, Angelica dahurica, Saposhnikovia divaricata, the above herbs total Grind into coarse powder, mix with half a pound each of white wine and white vinegar, heat in an iron pot over low heat, put into a cloth bag, and apply externally to the lower abdomen for 20-30 minutes, once a day, 10 times as one course of treatment) + Traditional Chinese medicine retention enema therapy (Sparganium rhizome, Curcuma rhizome, Red vine, Polygonum cuspidatum, Phellodendron bark, Smilax glabra, Gleditsia sinensis thorn, Salvia miltiorrhiza, decoct the above herbs in water to 100ml, temperature 40 degrees Celsius, retain enema, once a day, 10 times as one course of treatment); The protocol optimization sub-unit considers that the patient has no history of drug allergy and normal liver and kidney function (creatinine clearance rate 85mL / min), so no adjustment of drug dosage is required; The drug interaction detection sub-unit searches for "cefuroxime + peach kernel", no interaction is found, "cefuroxime + licorice" (without licorice) has no risk, the protocol generation cycle is 2 menstrual cycles, and the retest is performed in the 4th week; The protocol suitability assessment unit uses the formula Calculations: Age x1 = 32 years (normalized to 0.55), recommended age range 25-38 years (y1 = 0.55); Liver and kidney function x2 = 1, y2 = 1; Allergy history x3 = 0 (none), y3 = 0; Fertility desire x4 = 1 (urgent), y4 = 1; Syndrome stability x5 = 1, y5 = 1, M = 1 - (0 + 0 + 0 + 0 + 0) / 5 = 1 (highly suitable), this plan is the preferred recommendation.

[0059] II. Dynamic evaluation of efficacy and follow-up efficacy evaluation implementation: After 2 menstrual cycles of treatment, the evaluation index collection unit obtains review data: Western medicine indicators (MRI shows that the adhesion range is reduced to <1 / 3 of the pelvic cavity, IL-6 = 8 pg / mL, bilateral patency is shown by hysterosalpingography), TCM indicators (burning pain in the lower abdomen (mild, 1 point), lumbosacral pain (mild, 1 point), and white (mild, 1 point), thin white tongue), fertility-related indicators (ovulation monitoring shows that the dominant follicle develops normally, and the endometrial thickness is 9 mm); the evaluation calculation subunit compares before and after treatment: the adhesion degree is reduced by 1 level, and the syndrome score is reduced from 8.7 to 2.3 (a decrease of 73.5% ≥ 50%), which is determined as “markedly effective”; the fertility probability prediction subunit calculates according to the formula P = 0.4 × A + 0.35 × B + 0.25 × C: adhesion is significantly improved (A = 1, a = 0.4, 0.4), ovary function is normal (B = 1, b = 0.35, 0.35), endometrial condition is optimal (C = 1, g = 0.25, 0.25), P = 0.4 + 0.35 + 0.25 = 1 (high probability), which indicates that the current scheme is effective, and can continue to be executed for 1 cycle. Follow-up management: the patient follow-up management module generates a follow-up plan: follow-up after the end of the 3rd menstrual cycle of treatment, follow-up every month after pregnancy; inform by phone “please come to the hospital for reexamination with MRI report and sex hormone results on X day of X month of 2025”; record the symptom changes (no lower abdominal pain, normal leukorrhea) during follow-up, and automatically compare the data before treatment to generate a follow-up report; the patient is pregnant after 3 months, additional collection of blood HCG = 5000 IU / L (normal doubling), B-ultrasound shows intrauterine gestational sac, and health advice of “avoiding overwork and daily supplementation of 400 pg folic acid” is given.

[0060] III. Performance verification table

[0061]

[0062]

[0063] Traditional artificial diagnosis and treatment needs 15-20 minutes to integrate multi-source data, the scheme depends on the experience of doctors, the subjective score of adaptability is only 75 points or less, the recognition rate of drug interactions is less than 65%, the proportion of patients with marked effect is 60%, and the follow-up compliance rate is less than 70%. The system of the present application reduces the diagnosis time to 5 minutes, the adaptability score of the scheme is 90 points or more, the recognition rate of drug interactions is 98%, the proportion of patients with marked effect is 85%, and the follow-up compliance rate is 90%. For example, in the traditional mode, 2 patients had mild kidney damage due to the failure to identify the “cefotaxime + licorice” interaction, and the present application system gave early warning and replaced the drug, without adverse reactions; among 10 patients with marked effect in traditional diagnosis and treatment, the present application system had 25 patients with marked effect, fully embodying the advantages of the system in diagnosis and treatment efficiency, safety and effectiveness, and meeting the needs of gynecological outpatient service in primary hospitals.

[0064] Example 2

[0065] Tertiary hospital reproductive specialist (severe adhesion + chongren stasis) application

[0066] This example is for a 35-year-old pelvic adhesion infertility patient (5 years of marriage without pregnancy, 2 times of artificial abortion history, 1 time of laparoscopic adhesion lysis), who was admitted to the reproductive specialist of a certain tertiary hospital. The system was used to solve the complex diagnosis and treatment problems of severe adhesion (ovary-tubal + pelvic peritoneal adhesion, range > 2 / 3 of the pelvic cavity) and qi stagnation and blood stasis syndrome. The system data processing accuracy is 99.5%, the scheme matching degree is 97%, and it is suitable for high-difficulty and complex case scenarios.

[0067] I. System deployment and module application

[0068] Multi-modal data acquisition module: The western medicine data acquisition unit interfaces with PACS to obtain pelvic MRI (slice thickness 1.5 mm, resolution 1024 x 1024 pixels, indicating severe adhesion of ovary-tube + pelvic peritoneum), laboratory examination (IL-6 = 12 pg / mL, TNF-α = 6 pg / mL, sex hormone: FSH = 8.2 mIU / mL, LH = 7.5 mIU / mL, E2 = 280 pg / mL), gynecological examination (uterus anterior, obvious adnexal tenderness, little amount of white cervical mucus); The traditional Chinese medicine data acquisition unit collects four diagnostic information: lower abdominal stabbing pain (heavy, 3 points), chest and hypochondrium distending pain (moderate, 2 points), many blood clots during menstruation (heavy, 3 points), purple tongue with ecchymosis (heavy, 3 points), and pulse (heavy, 3 points), upload tongue and pulse instrument data; Age, disease duration, and reproductive history and lifestyle acquisition sub-unit: 2 times of abortion (marked as high-risk factor), past history of laparoscopic surgery (2023), 6.5 hours of sleep per day (unhealthy lifestyle), like spicy food, exercise once a week (irregular), moderate mental stress; The data preprocessing sub-unit has no missing values, IL-6 = 12 pg / mL is not abnormal, and the data integrity is 99%.

[0069] Western and traditional Chinese medicine dual-dimension diagnosis module: Western medicine diagnosis unit determines the adhesion site ovary-tube + pelvic peritoneum, degree severe, bilateral oviducts not passable by hysterosalpingography, generates western medicine sub-process "severe adhesion"; The traditional Chinese medicine syndrome differentiation unit calculates the total score of syndromes: lower abdominal stabbing pain (w i = 0.9, s i = 3, 2.7), chest and hypochondrium distending pain (w i = 0.6, s i = 2, 1.2), many blood clots during menstruation (w i = 0.8, s i = 3, 2.4), purple tongue with ecchymosis (w i = 0.7, s i = 3, 2.1), and pulse (w i = 0.8, s i=3, total 2.4), S = 2.7 + 1.2 + 2.4 + 2.1 + 2.4 = 10.8 points. Combined with Qi deficiency constitution, the TCM syndrome type "Qi stagnation and blood stasis syndrome" is generated. The adhesion risk prediction subunit inputs previous severe adhesion (1), IL-6 = 12 pg / mL (0.7), Qi stagnation and blood stasis syndrome (0.9), and sitting for 9 hours a day (0.95). The model outputs high risk (68%) and suggests "re-examination once every 2 weeks".

[0070] Personalized treatment plan generation module: The Western medicine plan generation unit recommends laparoscopic adhesiolysis combined with hysteroscopy, followed by postoperative drug treatment (cephalosporin sodium 2g IV drip). (bid, for 2 consecutive days); The TCM treatment plan generation unit adopts the method of promoting qi circulation, activating blood and removing blood stasis. The prescription is: Bailian Sanjie Decoction (Hedyotis diffusa 30g, Scutellaria barbata 30g, Gleditsia sinensis 10g, Epimedium brevicornu 15g, Curculigo orchioides 15g, Polyporus umbellatus 20g, Curcuma zedoaria 10g, Eupolyphaga sinensis 10g, Luffa cylindrica 15g, Citrus reticulata 10g), the dosage is adjusted according to the body weight of 65kg (65×0.1g=6.5g). The TCM herbal poultice therapy (Formula No. 2: raw Artemisia argyi, Spatholobus suberectus, Clematis armandii, Carthamus tinctorius, Pulsatilla chinensis, Angelica dahurica, Saposhnikovia divaricata, the above herbs are ground into coarse powder, mixed with half a catty each of white wine and white vinegar, put into an iron pot and heat over low heat, put into a cloth bag, apply to the lower abdomen for 20-30 minutes, once a day). 10 treatments constitute one course of treatment) + Traditional Chinese Medicine retention enema therapy (Sparganium rhizome, Curcuma rhizome, Red vine, Polygonum cuspidatum, Phellodendron bark, Smilax glabra, Gleditsia sinensis thorns, Salvia miltiorrhiza, decoct the above herbs in water to 100ml, temperature 40 degrees Celsius, retain enema, once a day, 10 treatments constitute one course of treatment); Drug interaction detection sub-unit search is risk-free, generated protocol cycle is 3 menstrual cycles; Protocol suitability assessment: Age x_1 = 35 years old (normalized 0.65), surgical protocol recommended 30-40 years old (y1 = 0.65); Liver and kidney function x2 = 0.8 (mildly abnormal), y2 = 1, M = 1 - (0 + |0.8 - 1| + 0 + 0 + 0) / 5 = 1 - 0.04 = 0.96 (highly suitable).

[0071] II. Dynamic evaluation and follow-up of treatment efficacy

[0072] Efficacy evaluation implementation: After 3 cycles of treatment, review the data: MRI shows that the adhesion range is reduced to 1 / 3-2 / 3 of the pelvic cavity (moderate), IL-6 = 8 pg / mL, hysterosalpingography shows that the left side is unobstructed and the right side is obstructed; TCM syndrome score: lower abdominal stabbing pain (mild, 1 point), chest and breast pain (none, 0 points), S = 0.9 x 1 = 0.9 points; the evaluation calculation sub-unit determines that it is "effective" (the degree of adhesion does not increase, and the syndrome score decreases by 92%> 49%); fertility probability prediction: moderate improvement of adhesion (A = 0.8, a = 0.4, 0.32), general ovarian function (B = 0.6, β = 0.35, 0.21), endometrial thickness 8mm (C = 1, γ = 0.25, 0.25), P = 0.32 + 0.21 + 0.25 = 0.78 (high probability), it is recommended to continue the traditional Chinese medicine sealing bag therapy + traditional Chinese medicine retention enema therapy; the case knowledge base module retrieves similar cases of "35 years old + severe adhesion + chongren stasis syndrome + hysteroscopy + bai lian sanjie decoction" (similarity 85%), which shows that the effective rate of this scheme is 82%, providing a reference for doctors.

[0073] Follow-up management: The follow-up plan is set to follow up at 1, 2 and 3 months after surgery, and telephone follow-up; the patient is not pregnant at the 2nd month follow-up, and the reason is analyzed as "general endometrial blood flow", the treatment effect is evaluated and the scheme is adjusted, and kidney-nourishing endometrial traditional Chinese medicine is added; the patient is pregnant at the 4th month, and the follow-up management module collects blood HCG = 8000 IU / L, and the treatment for preventing miscarriage is carried out.

[0074] Four, performance verification table

[0075] Evaluation index Traditional artificial diagnosis and treatment mode The system of the present application Promotion effect Complex case diagnosis accuracy (%) 75～80 95～98 Significant improvement Operation scheme optimization rate (%) 60～65 90～95 Greatly improved Fertility probability prediction coincidence rate (%) 70～75 92～95 Significant improvement Adverse event occurrence rate (%) 15～20 3～5 Greatly reduced Doctor diagnosis and treatment satisfaction (%) 70～75 90～95 Significant improvement

[0076] The diagnosis accuracy rate of traditional artificial diagnosis and treatment of complex cases is 80% or less, the operation scheme optimization rate is 65%, the fertility probability prediction coincidence rate is 75%, the incidence of adverse events (such as drug adverse reactions and surgical complications) is 20%, and the doctor's satisfaction is 75%. The diagnosis accuracy rate of the system is 98%, the operation scheme optimization rate is 95%, the fertility probability prediction coincidence rate is 95%, the incidence of adverse events is 5%, and the doctor's satisfaction is 95%. For example, in the traditional mode, 3 patients relapsed after surgery due to unoptimized surgical scheme, and the system combined with the adhesion site and constitution to adjust the surgical path, only 1 relapsed; 8 cases of traditional diagnosis and treatment have a prediction deviation of more than 20%, and the system has a prediction coincidence rate of 95%, helping patients to establish reasonable expectations, fully adapting to the diagnosis and treatment needs of complex cases in third-level hospitals, and improving the diagnosis and treatment effect and doctor's work efficiency of high-difficulty cases.

[0077] Figure 2Intuitively embody the comprehensive advantages of integrated diagnosis and treatment of traditional Chinese and Western medicine. Traditional Western medicine diagnosis and treatment rely on imaging and drugs, with a diagnostic accuracy rate of less than 85%, a regimen adaptability of 75 points, and a drug interaction recognition rate of only 70% due to the neglect of TCM syndrome types and individual constitutions. The proportion of patients with obvious effects is 62%, and postoperative adhesion recurrence (adverse events) is easy to occur (15%); traditional Chinese medicine diagnosis and treatment has advantages in syndrome regulation, but lacks anatomical assessment of Western medicine, with a diagnostic accuracy rate of less than 82%, a drug interaction recognition rate of 60%, and limited efficacy for severe adhesions. Through the dual-diagnosis of traditional Chinese and Western medicine and the generation of personalized programs, the diagnostic accuracy rate is improved to 98%, the regimen adaptability is 95 points, the drug interaction recognition rate is 98%, the proportion of patients with obvious effects is 85%, and the adverse events are reduced to 5%. The follow-up compliance rate is 90%, which fully solves the problems of "poor precision, low safety, and limited efficacy" of traditional single diagnosis and treatment, and adapts to different difficulty levels of pelvic adhesion infertility diagnosis and treatment scenes.

[0078] Figure 3 Quantitative display of the dynamic effect of integrated treatment of traditional Chinese and Western medicine. Before treatment, the total score of the patient's syndrome was 10.8 points, IL-6 = 16 pg / mL (inflammation active), and adhesion severe; after the first cycle, the TCM Qihu Huoxue Yuyang prescription (Bailian Sanjie Decoction) reduced the total score of the syndrome to 7.2 points (improved by 33.3%), IL-6 to 12.5 pg / mL (controlled by 21.9%), and adhesion still severe; after the third cycle, the total score of the syndrome was 4.5 points (improved by 58.3%), IL-6 was reduced to 9.8 pg / mL (within the normal range), and adhesion was reduced to moderate; after the fourth cycle, the total score of the syndrome was 0.8 points (basically no symptoms), IL-6 = 6.5 pg / mL, and adhesion was mild. The syndrome improvement rate was 92.6%, the inflammation control rate was 59.4%, and the chart clearly embodied the synergistic relationship of "syndrome improvement-inflammation control-adhesion reduction", avoiding the limitations of traditional diagnosis and treatment which only focus on a single dimension, and providing complete data support for efficacy evaluation in the combination of traditional Chinese and Western medicine.

[0079] Figure 4Embodies the individualization of diagnosis and treatment scheme. Patients with mild adhesion have mild inflammation and anatomical abnormalities, 45% of them are treated with western medicine, 30% of them are treated with Chinese medicine + Chinese medicine sealing therapy + Chinese medicine retention enema therapy, and 25% of them are treated with combined scheme, and the fertility probability is high (about 0.8), the treatment cycle is 1-2, and the reexamination interval is 4 weeks; Severe adhesion needs surgical lysis (40% of western medicine scheme), 40% of them are treated with combined postoperative drug and traditional Chinese medicine conditioning, the fertility probability is medium (about 0.6), the treatment cycle is 3-4, and the reexamination interval is shortened to 2 weeks; Severe adhesion accounts for 50%, needs 6 cycles of treatment, and the fertility probability is low-medium (about 0.5); The proportion of traditional Chinese medicine conditioning for recurrent adhesion increases to 35%, and the risk of re-adhesion is reduced by promoting blood circulation to remove blood stasis. The chart clearly shows the correlation of "adhesion degree-scheme composition-fertility expectation", helps doctors to quickly match the scheme, avoids the traditional "one-size-fits-all" diagnosis and treatment mode, and provides clear treatment cycle and fertility expectation for patients, improves the confidence and compliance of diagnosis and treatment.

[0080] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can make equivalent replacement or change according to the technical scheme and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A clinical decision support system for treating pelvic adhesion infertility in the combination of traditional Chinese and Western medicine, characterized by comprising: a multi-modal data acquisition module for acquiring multi-dimensional diagnosis and treatment data of patients in traditional Chinese and Western medicine, supporting multi-source data access; a Western medicine data acquisition unit connected to a hospital HIS system to import pelvic imaging examination, laboratory examination, gynecological examination data; a traditional Chinese medicine data acquisition unit inputs traditional Chinese medicine four diagnostic information through a structured questionnaire and a doctor's syndrome input interface, supporting image uploading; a data preprocessing subunit fills in missing values using the adjacent case interpolation method, and identifies and marks abnormal values through the 3σ rule; a traditional Chinese and Western medicine dual-dimensional diagnosis module for realizing the combined diagnosis of pelvic adhesion infertility in traditional Chinese and Western medicine; a Western medicine diagnosis unit identifies the adhesion site and adhesion degree based on laboratory tests and imaging examinations, generates a Western medicine staging in combination with inflammatory factor levels and tubal patency; a traditional Chinese medicine syndrome differentiation unit adopts a "syndrome element-syndrome type" mapping model to quantitatively score four diagnostic information, determines the core syndrome through weighted summation, and generates a traditional Chinese medicine syndrome type in combination with the patient's constitution; a diagnosis result integration subunit associates the Western medicine staging and the traditional Chinese medicine syndrome type to form a combined diagnosis report of Western medicine staging + traditional Chinese medicine syndrome type, and marks the diagnosis basis; a personalized diagnosis and treatment plan generation module for outputting a combination of traditional Chinese and Western medicine diagnosis and treatment plan according to the combined diagnosis result; a Western medicine plan generation unit matches treatment means based on the Western medicine staging: mild adhesion recommends drug treatment, moderate adhesion recommends physical therapy + drug assistance, and severe adhesion recommends surgical treatment; a traditional Chinese medicine plan generation unit matches principles and prescriptions based on the traditional Chinese medicine syndrome type and the menstrual period: selects the self-made prescription Bailian Sanjie Decoction as the main prescription for the disease, treats the late menstrual period as the long-term of yin, focuses on tonifying yin to balance yin and yang; the menstrual period and ovulation period should be treated as the transformation period of yin and yang, and should be treated with blood-activating and menstruation-regulating products to facilitate the discharge of menstrual blood and ovum; the premenstrual period is treated as the long-term of yang, and the treatment should be warm and tonifying yang qi to promote the arrival of menstruation, and the prescription Bailian Sanjie Decoction combined with Yishen Guchong Decoction is selected, and the traditional Chinese medicine sealing bag therapy and traditional Chinese medicine retention enema therapy are recommended; a scheme optimization subunit adjusts the scheme in combination with individual differences of patients to generate a complete scheme including treatment cycle, review time and matters needing attention; a curative effect dynamic evaluation module for real-time monitoring of diagnosis and treatment effect and adjustment of scheme; an evaluation index acquisition unit regularly acquires patient review data: Western medicine indexes, traditional Chinese medicine indexes and fertility-related indexes; an evaluation calculation subunit compares the changes of indexes before and after treatment, and classifies them as "markedly effective, effective and ineffective"; a scheme adjustment subunit re-calls the diagnosis module to analyze the reasons for the ineffective patients, generates an optimized scheme, and forms a "diagnosis-treatment-evaluation-adjustment" closed loop.

2. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 1, further characterized in that The TCM syndrome quantification scoring unit includes a TCM syndrome total score calculation formula: In the formula, S is the total score of TCM syndromes, n is the number of TCM syndrome items included in the score, w i is the weight of the i th syndrome item, s i is the severity score of the i th syndrome item.

3. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 1, further characterized by The diagnosis and treatment scheme adaptability evaluation unit calculates the matching degree of the scheme and the individual characteristics of the patient, screens the optimal scheme, and reduces the treatment risk; the scheme adaptability index calculation formula is: In the formula, M is a regimen adaptability index, k is the number of evaluation dimensions, x j is the actual characteristic value of the patient in the jth dimension, y j is the regimen recommended characteristic value in the jth dimension.

4. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 1, characterized in that, The multi-modal data acquisition module further includes a patient age, disease course and fertility history and life habit acquisition subunit, which acquires patient's past fertility history, life habit, dietary preference, mental stress data through a structured form; the subunit supports automatic data classification and labeling.

5. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 1, characterized in that, The double-dimension diagnosis module of traditional Chinese and Western medicine further includes an adhesion risk prediction subunit, which constructs a machine learning prediction model based on patient historical data to predict the risk of adhesion aggravation of the patient; the model input features include: previous adhesion degree, inflammatory factor baseline level, TCM syndrome type, and living habits; and the output risk level.

6. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 1, characterized in that, The personalized diagnosis and treatment scheme generation module further includes a drug interaction detection subunit, which is internally provided with a traditional Chinese and Western medicine drug interaction database to detect whether there is any contraindication or adverse reaction between traditional Chinese and Western medicine drugs in the scheme.

7. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 2, characterized in that, The efficacy dynamic evaluation module further includes a fertility probability prediction subunit, which predicts the natural conception probability of the patient within one year by analyzing the changes in indexes before and after treatment, thereby providing a reference for treatment scheme adjustment; the fertility probability prediction formula is: P=α×A+β×B+γ×C In the formula, P is the natural conception probability within one year, α is the adhesion improvement weight, A is the adhesion improvement score, β is the ovarian function weight, B is the ovarian function score, γ is the endometrial condition weight, and C is the endometrial score.

8. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 3, The features further include a case knowledge base module, which stores clinical cases of treating pelvic adhesion-induced infertility by combining traditional Chinese and Western medicine; the cases are desensitized and marked with core features; The module supports case retrieval function, and doctors can filter similar cases according to "age range, adhesion degree, TCM syndrome type, and treatment mode" in multiple dimensions, and view the scheme execution effect and adjustment process of similar cases; meanwhile, the module supports case updating and storage after expert review.

9. The clinical decision support system for treating pelvic adhesion-induced infertility in a combination of traditional Chinese and Western medicine according to claim 5, further characterized by The module includes a patient follow-up management module, which generates a follow-up plan according to the diagnosis and treatment scheme; a follow-up data acquisition unit supports doctors to input symptoms and examination results during follow-up, and automatically compares the pre-treatment data to generate a follow-up report; For pregnant patients, the module supports follow-up and miscarriage prevention; for non-pregnant patients, the module analyzes the reasons for non-pregnancy and re-adjusts the treatment scheme based on the efficacy evaluation.

10. The clinical decision support system for treating pelvic adhesion-induced infertility by combining traditional Chinese and Western medicine according to claim 4, characterized in that the system further includes a permission management and data security module, which adopts a role-based access control mechanism to divide three types of roles: doctors, department administrators, and system administrators; the module supports double authentication of user identity, uses TLS1.3 encryption protocol for patient data transmission, sets access logs for the database, and supports data backup.

Images