A decision-making support system based on medical record data analysis

Abstract

The present invention discloses an auxiliary decision-making system based on medical record data analysis, and relates to the technical field of data analysis. The auxiliary decision-making system based on medical record data analysis includes a medical record data analysis subsystem, a treatment plan recommendation subsystem, and an expected effect evaluation subsystem. By analyzing the medical record data of patients with pulmonary infection and classifying them according to the severity of the disease, personalized treatment plan recommendations can be provided for each patient, and the auxiliary decision-making data set and treatment plan recommendations can be updated in real time. Evaluation and adjustment are performed according to the actual situation of the patient and changes in the disease. Through the expected effect evaluation subsystem, the effect of the treatment plan can be evaluated, and the situation of poor treatment effect can be discovered in time, so as to adjust the treatment plan and improve the treatment effect and the patient's recovery rate.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular to an auxiliary decision-making system based on medical record data analysis. Background Art

[0002] As people's quality of life improves, they pay more attention to their own health conditions. When people need medication, they hope to get more accurate medication recommendations. At the same time, the medical team also hopes to use the medication accurately. However, since the patient's condition is relatively complicated, it may be difficult to provide accurate treatment.

[0003] The Chinese patent application document with publication number CN113657970A discloses a drug recommendation method, device, equipment and storage medium based on artificial intelligence. The method obtains the health insurance claim information of the user within a preset time period, determines the user's disease information based on the health insurance claim information, and queries the drug ingredient information of the drug corresponding to the disease of the user based on the disease information. Then, the disease information and drug ingredient information are input into a pre-trained drug recommendation model, and multiple drug recommendation plans are output. Finally, a target drug recommendation plan that meets the user's medication time is screened out from the multiple drug recommendation plans, and the target drug recommendation plan is recommended to the user, thereby automatically generating a drug recommendation plan that meets the user's health insurance claim information through artificial intelligence, without the user having to fill in a large amount of information, and the recommendation efficiency is higher; at the same time, by inputting the drug ingredient information in the drug recommendation model, it is ensured that the degree of mutual influence between drugs in each recommended drug recommendation plan is lower than the preset value, so as to avoid the mutual influence between drugs and failure or harm to the user's body, thereby improving the recommendation accuracy; in addition, by screening out the target drug recommendation plan that meets the user's medication time, it is ensured that the recommended target drug recommendation plan meets the user's medication habits, and intelligent and personalized recommendations are achieved.

[0004] Although the existing medication treatment plan recommendation method can achieve intelligent and personalized recommendations, it is inconvenient to update the auxiliary decision-making data set and treatment plan recommendations in real time according to the patient's actual situation and changes in the condition to adapt to the patient's changing needs. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an auxiliary decision-making system based on medical record data analysis, which solves the problem that the existing medication treatment plan recommendation method is inconvenient to update the auxiliary decision-making data set and treatment plan recommendations in real time according to the patient's actual situation and changes in the condition to adapt to the patient's changing needs.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: an auxiliary decision-making system based on medical record data analysis, including a medical record data analysis subsystem, a treatment plan recommendation subsystem and an expected effect evaluation subsystem, wherein: the medical record data analysis subsystem is used to extract the medical record data of patients with lung infection, obtain X-ray imaging data and CT scan imaging data, and obtain an auxiliary decision-making data set based on the X-ray imaging data and CT scan imaging data; the treatment plan recommendation subsystem is used to classify the severity of the condition of patients with lung infection according to the auxiliary decision-making data set obtained by the medical record data analysis subsystem, and recommend treatment plans based on the classified levels, each level corresponding to a treatment plan; the expected effect evaluation subsystem is used to evaluate the condition of patients with lung infection after the patients with lung infection adopt the treatment plan.

[0007] Furthermore, the auxiliary decision data set includes an X-ray imaging feature data set and a CT scan imaging feature data set. The process of grading the severity of the condition of patients with lung infection based on the auxiliary decision data set is as follows: obtaining reference data sets of lung infection of various levels stored in the database, wherein the reference data sets of lung infection of various levels include X-ray imaging reference feature data sets and CT scan imaging reference feature data sets; comparing the auxiliary decision data set with the reference data sets of lung infection of various levels to obtain various condition assessment values; calibrating the level of the lung infection reference data set corresponding to the minimum condition assessment value as the severity of the condition of the patient with lung infection.

[0008] Furthermore, the calculation formula of the condition assessment value is as follows:

[0009]

[0010] In the formula, i is the number of the disease level, i = 1, 2, 3, ..., n, n is the total number of disease levels, Bp i is the disease assessment value obtained by comparing the auxiliary decision-making data set with the i-th level lung infection reference data set, A1 is the X-ray imaging feature data set, is the X-ray imaging feature dataset in the i-th level lung infection dataset, is the calculation function of the X-ray imaging feature data set and the X-ray imaging reference feature data set, which is used to calculate the similarity between the X-ray imaging feature data set and the X-ray imaging reference feature data set. α1 is the calculation function The weight factor of the value, B1 is the CT scan imaging feature data set, The CT scan imaging parameter feature dataset in the i-th level lung infection parameter dataset, is the calculation function of the CT scan imaging feature data set and the CT scan imaging reference feature data set, which is used to calculate the similarity between the CT scan imaging feature data set and the CT scan imaging reference feature data set. α2 is the calculation function The weighting factor of the value of .

[0011] Furthermore, the X-ray imaging feature data set includes lung density and lung field transmittance of patients with lung infection, and the CT scan imaging feature data set includes lung mutation area, patchy shadow area and lung lobe atrophy of patients with lung infection; the X-ray imaging parameter feature data set includes lung parameterized density and lung field parameterized transmittance of patients with lung infection, and the CT scan imaging parameterized feature data set includes lung mutation parameterized area, patchy shadow parameterized area and lung lobe parameterized atrophy of patients with lung infection.

[0012] Furthermore, the process of evaluating the condition of lung infection patients after the lung infection patients adopt the treatment plan is as follows: according to the set treatment cycle, the condition information of the lung infection patients who accept the treatment plan is collected, the post-treatment medical record data is obtained, and the post-treatment condition data is extracted from the post-treatment medical record data; the extracted post-treatment condition data of each treatment cycle is compared and analyzed with the expected post-treatment condition data corresponding to each treatment cycle stored in the database, the condition of the lung infection patients is evaluated, and it is determined whether to continue or stop the treatment based on the condition.

[0013] Furthermore, the process of evaluating the condition of patients with lung infections and determining whether to continue or stop treatment based on the condition is as follows: comparing and analyzing the post-treatment medical record data of each treatment cycle with the expected condition data of the corresponding treatment cycle to obtain a first evaluation index, which is used to represent the difference between the treatment effect of each treatment cycle and the expected effect; judging whether the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold, if the first evaluation index corresponding to the last treatment cycle is not greater than the set first evaluation threshold, stopping the treatment; if the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold, determining whether to continue or stop treatment based on the post-treatment medical record data of each treatment cycle.

[0014] Furthermore, the process of determining whether to continue treatment or stop treatment based on the post-treatment medical record data of each treatment cycle is as follows: obtain the post-treatment medical record data of n treatment cycles before the last treatment cycle and number them; obtain the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle; if the difference index is greater than the set difference threshold, it means that the lung infection patient has an effect after receiving the treatment plan but has not achieved the expected effect of each treatment cycle, and the treatment should continue; if the difference index is not greater than the set difference threshold, it means that the lung infection patient has no effect after receiving the treatment plan and has not achieved the expected effect of each treatment cycle, and the treatment should be stopped and the treatment plan should be changed.

[0015] Furthermore, the post-treatment condition data includes post-treatment X-ray data and post-treatment CT scan data; the expected post-treatment condition data includes expected post-treatment X-ray data and expected post-treatment CT scan data.

[0016] Furthermore, the calculation formula of the first evaluation index is as follows:

[0017]

[0018] Where j is the number of the treatment cycle, j = 1, 2, 3, ..., m, m is the total number of treatment cycles, Pz j is the first evaluation index of the jth treatment cycle, C 1j is the X-ray data after treatment of the jth treatment cycle, C 2j is the expected X-ray data after treatment of the jth treatment cycle, f(C 1j ,C 2j ) is the calculation function of the post-treatment X-ray data and the expected post-treatment X-ray data, D 1h is the CT scan data after treatment of the jth treatment cycle, D 2j is the expected CT scan data after treatment of the jth treatment cycle, f(D 1j ,D 2j ) is the calculation function of the CT scan data after treatment and the expected CT scan data after treatment.

[0019] Furthermore, the calculation formula of the difference index is as follows:

[0020]

[0021] In the formula, Cz (n,n+1) fm is the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle, n is the post-treatment lung density in the post-treatment X-ray data of the nth treatment cycle, fm n+1is the lung density after treatment in the post-treatment X-ray data of the n+1th treatment cycle, ft n is the post-treatment lung field transmittance in the post-treatment X-ray data of the nth treatment cycle, ft n+1 fs is the post-treatment lung field transmittance in the post-treatment X-ray data of the n+1th treatment cycle, n fs is the post-treatment lung mutation area in the post-treatment CT scan data of the nth treatment cycle, n+1 fb is the post-treatment lung mutation area in the post-treatment CT scan data of the n+1th treatment cycle, n fb is the post-treatment patchy shadow area in the post-treatment CT scan data of the nth treatment cycle, n+1 is the patchy shadow area after treatment in the CT scan data after treatment for the n+1th treatment cycle, fz n is the post-treatment lung lobe shrinkage in the post-treatment CT scan data of the nth treatment cycle, fz n+1 It is the amount of lung lobe shrinkage after treatment in the CT scan data after treatment of the n+1th treatment cycle.

[0022] The present invention has the following beneficial effects:

[0023] This decision support system based on medical record data analysis analyzes the medical record data of patients with lung infections and classifies them into levels according to the severity of the disease. It can provide personalized treatment plan recommendations for each patient, update the decision support data set and treatment plan recommendations in real time, and evaluate and adjust according to the patient's actual situation and changes in the condition. Through the expected effect evaluation subsystem, the effectiveness of the treatment plan can be evaluated, and situations where the treatment effect is poor can be discovered in a timely manner, so as to adjust the treatment plan and improve the treatment effect and the patient's recovery rate.

[0024] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a flow chart of the auxiliary decision-making system based on medical record data analysis of the present invention. DETAILED DESCRIPTION

[0026] The embodiment of the present application uses an auxiliary decision-making system based on medical record data analysis to achieve real-time updating of auxiliary decision-making data sets and treatment plan recommendations according to the patient's actual situation and changes in the condition to adapt to the patient's changing needs.

[0027] The overall idea of ​​the problem in the embodiment of this application is as follows:

[0028] Extract medical record data of patients with lung infection from hospital information systems or electronic medical record systems, including basic information, clinical test data, imaging data, etc. By extracting X-ray imaging data and CT scan imaging data, image analysis and processing are performed to extract key feature information, such as infection site, lesion range, etc., to form a decision-making support data set.

[0029] The system receives the decision-making support data set from the medical record data analysis subsystem, and classifies the severity of the lung infection patients according to the characteristic information therein. Based on the classification level, the system pre-sets a series of treatment plans for different levels and corresponds them to the level. According to the patient's level, the corresponding treatment plan is recommended to ensure personalized treatment suggestions for patients with different disease severity.

[0030] The patient's condition after adopting the treatment plan is evaluated to verify the treatment effect. The patient's efficacy can be evaluated through various indicators such as clinical scores and imaging changes to help doctors adjust the treatment plan or make further diagnosis and treatment decisions.

[0031] See also Figure 1 The embodiment of the present invention provides a technical solution: an auxiliary decision-making system based on medical record data analysis, including a medical record data analysis subsystem, a treatment plan recommendation subsystem and an expected effect evaluation subsystem, wherein: the medical record data analysis subsystem is used to extract the medical record data of patients with lung infection, obtain X-ray imaging data and CT scan imaging data, and obtain an auxiliary decision-making data set based on the X-ray imaging data and the CT scan imaging data; the treatment plan recommendation subsystem is used to classify the severity of the condition of patients with lung infection according to the auxiliary decision-making data set obtained by the medical record data analysis subsystem, and recommend treatment plans based on the classified levels, each level corresponds to a treatment plan, and whether the treatment plan is used needs to be judged by relevant doctors; the expected effect evaluation subsystem is used to evaluate the condition of patients with lung infection after the patients with lung infection adopt the treatment plan.

[0032] In general, through the analysis and processing of medical records of patients with lung infections, combined with personalized treatment plan recommendations and treatment effect evaluation, a tool for assisting doctors in making decisions is provided, aiming to improve treatment effects and patients' medical experience. At the same time, through the real-time update mechanism, it ensures that the system can adapt to the changing needs of patients in a timely manner and provide timely and effective decision-making support.

[0033] Specifically, the auxiliary decision data set includes an X-ray imaging feature data set and a CT scan imaging feature data set. The process of classifying the severity of the condition of patients with lung infection based on the auxiliary decision data set is as follows: obtain the reference data sets of lung infection of each level stored in the database, and the reference data sets of lung infection of each level include an X-ray imaging reference feature data set and a CT scan imaging reference feature data set; compare the auxiliary decision data set with the reference data sets of lung infection of each level to obtain each condition assessment value; calibrate the level of the lung infection reference data set corresponding to the minimum condition assessment value as the severity of the condition of the lung infection patient.

[0034] The calculation formula of the disease assessment value is as follows:

[0035]

[0036] In the formula, i is the number of the disease level, i = 1, 2, 3, ..., n, n is the total number of disease levels, Bp i is the disease assessment value obtained by comparing the auxiliary decision-making data set with the i-th level lung infection reference data set, A1 is the X-ray imaging feature data set, is the X-ray imaging feature dataset in the i-th level lung infection dataset, is the calculation function of the X-ray imaging feature data set and the X-ray imaging reference feature data set, which is used to calculate the similarity between the X-ray imaging feature data set and the X-ray imaging reference feature data set. α1 is the calculation function The weight factor of the value, B1 is the CT scan imaging feature data set, is the CT scan imaging parameter feature dataset in the i-th level lung infection parameter dataset, is the calculation function of the CT scan imaging feature data set and the CT scan imaging reference feature data set, which is used to calculate the similarity between the CT scan imaging feature data set and the CT scan imaging reference feature data set. α2 is the calculation function The weighting factor of the value of .

[0037] Obtain the stored lung infection parameter data sets of various levels from the database, which include X-ray imaging parameter feature data sets and CT scan imaging parameter feature data sets. For each disease level, calculate the disease assessment value according to the given calculation formula. This assessment value is obtained by comparing the auxiliary decision data set with the i-th level lung infection parameter data set.

[0038] The level of the lung infection parameter data set corresponding to the smallest one among all condition assessment values ​​is calibrated as the severity of the lung infection patient's condition.

[0039] Specifically, two weight factors are used to calculate the disease assessment value, which represent the importance of X-ray imaging data and CT scan imaging data respectively. and It is used to measure the similarity between the auxiliary decision-making data set and the reference data set, thereby reflecting the degree of match between the patient's condition and each condition level. Through quantitative calculation, the system can quickly and accurately assess the severity of the patient's condition based on the degree of match between the patient's imaging feature data and the auxiliary decision-making data set.

[0040] The weight factor can determine the relative importance of X-ray imaging data and CT scan imaging data in assessing the severity of the disease based on the experience and clinical practice of medical experts. For example, if CT scan imaging data is more important in the diagnosis of lung infection, a larger value can be assigned to α2 to give more importance to the role of CT scan data in the assessment process.

[0041] The X-ray imaging feature dataset includes the lung density and lung field transmittance of patients with lung infection, and the CT scan imaging feature dataset includes the lung mutation area, patchy shadow area and lung lobe atrophy of patients with lung infection; the X-ray imaging parameter feature dataset includes the lung parameterized density and lung field parameterized transmittance of patients with lung infection, and the CT scan imaging parameterized feature dataset includes the lung mutation parameterized area, patchy shadow parameterized area and lung lobe parameterized atrophy of patients with lung infection.

[0042] Lung density and lung field transmittance are obtained through X-ray imaging. On X-ray films, the density and transmittance of the lungs can be observed, and these characteristic data can be extracted through digital image processing methods. Lung mutation area, patchy shadow area and lobar atrophy can all be obtained through CT scanning imaging. CT scanning can provide more detailed and accurate lung structure information, including the area and distribution of lung lesions, as well as lobar atrophy. CT scanning images can be analyzed and computer software can be used to assist in extracting these characteristic data.

[0043] The X-ray imaging parameter feature data set and the CT scan imaging parameter feature data set are established in advance, and contain lung imaging feature parameter data for different disease levels. They are determined and established by medical experts based on clinical experience and research data, and are used as standards or references to assess the severity of patients' diseases.

[0044] The weight factor is used to adjust the relative importance of different features and is set based on the experience of medical experts and clinical practice. For example, if lung density is considered more important than lung mutation area in the diagnosis of lung infection, the weight factor α1 may be assigned a larger value.

[0045] Where gm is the lung density of patients with lung infection, gcm i is the lung parameter density of the lung infection patient in the X-ray imaging parameter feature dataset of the i-th level lung infection parameter dataset, gt is the lung field transmittance of the lung infection patient, gct i The lung field transmittance of the lung infection patient in the X-ray imaging reference feature dataset in the i-th level lung infection reference dataset.

[0046] The X-ray imaging similarity calculation formula is used to calculate the similarity between the X-ray imaging feature data set in the auxiliary decision data set and the X-ray imaging parameter feature data set of each disease level. The ln function in the formula is used to limit the range of similarity to avoid excessively large or small values ​​affecting the final evaluation results. By comparing the relative changes in lung density and lung field transmittance, the degree of match between the patient's lung condition and each disease level is evaluated, and then the severity of the disease is determined.

[0047] Where gb is the lung mutation area of ​​patients with lung infection, gbm i is the lung mutation reference area of ​​the lung infection patient in the CT scan imaging reference feature dataset of the i-th level lung infection reference dataset, gp is the patchy shadow reference area of ​​the lung infection patient, gpt i is the patchy shadow parameter area of ​​the CT scan imaging parameter feature dataset of the ith level of pulmonary infection, gs is the pulmonary lobe parameter atrophy of the pulmonary infection patient, gst i is the lung lobe atrophy value of the lung infection patients in the CT scan imaging parameter feature dataset in the i-th level lung infection parameter dataset.

[0048] The CT scan imaging similarity calculation formula is used to calculate the similarity between the CT scan imaging feature data set in the auxiliary decision-making data set and the CT scan imaging parameter feature data set of each disease level. The ln function is also used to limit the range of similarity to ensure the rationality of the evaluation results. By comparing the relative changes in the lung mutation area, patchy shadow area and lobe atrophy, the degree of match between the patient's lung condition and each disease level is evaluated to determine the severity of the disease.

[0049] By quantitatively evaluating the similarity between the patient's X-ray imaging features and CT scan imaging features and each disease level, the severity of the patient's disease can be objectively evaluated. Such an evaluation can avoid the intervention of subjective factors and provide a more objective and accurate judgment of the disease. Based on the accurate evaluation of the severity of the disease, personalized treatment plans can be provided for patients of different levels, which can more effectively deal with different degrees of disease and improve the treatment effect.

[0050] Specifically, the process of evaluating the condition of lung infection patients after they adopt a treatment plan is as follows: according to the set treatment cycle, the condition information of the lung infection patients who receive the treatment plan is collected, the post-treatment medical record data is obtained, and the post-treatment condition data is extracted from the post-treatment medical record data; the extracted post-treatment condition data of each treatment cycle is compared and analyzed with the expected post-treatment condition data corresponding to each treatment cycle stored in the database, the condition of the lung infection patient is evaluated, and it is determined whether to continue or stop treatment based on the condition.

[0051] The post-treatment condition data includes post-treatment X-ray data and post-treatment CT scan data; the post-treatment expected condition data includes post-treatment expected X-ray data and post-treatment expected CT scan data.

[0052] In this embodiment, according to the set treatment cycle, the condition information of the lung infection patients receiving the treatment plan is collected regularly, the medical record data after treatment is obtained, and the condition data after treatment is extracted from the medical record data after treatment, including X-ray data and CT scan data after treatment. The extracted condition data after treatment of each treatment cycle is compared and analyzed with the expected condition data after treatment corresponding to each treatment cycle stored in the database. Through the comparison and analysis, the condition of the lung infection patients is evaluated to understand the changes in the condition after treatment. According to the results of the condition assessment, it is decided whether to continue or stop treatment. If the condition has improved or is as expected, the current treatment plan can be continued or adjusted to further improve the condition. If the condition has not improved or adverse reactions occur, it may be necessary to adjust the treatment plan or stop treatment, and consider other treatment methods.

[0053] By regularly collecting and evaluating post-treatment medical data, we can promptly understand changes in the patient's condition and adjust the treatment plan in a timely manner to improve the treatment effect. Based on the post-treatment condition evaluation results, we can develop a personalized treatment plan for each patient to better meet the patient's treatment needs. By evaluating the condition after treatment, we can promptly discover situations where the treatment is ineffective or adverse reactions occur, avoid unnecessary treatment delays, and reduce the patient's pain and burden.

[0054] Specifically, the process of evaluating the condition of patients with lung infections and determining whether to continue or stop treatment based on the condition is as follows: comparing and analyzing the post-treatment medical record data of each treatment cycle with the expected condition data of the corresponding treatment cycle to obtain a first evaluation index, which is used to represent the difference between the treatment effect of each treatment cycle and the expected effect; judging whether the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold, if the first evaluation index corresponding to the last treatment cycle is not greater than the set first evaluation threshold, stopping the treatment; if the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold, determining whether to continue or stop treatment based on the post-treatment medical record data of each treatment cycle.

[0055] Compare and analyze the post-treatment medical record data of each treatment cycle with the corresponding expected condition data. Through the comparison and analysis, the difference between the post-treatment condition and the expected condition is evaluated to reflect the degree of treatment effect. Use the given calculation function to calculate the post-treatment X-ray data and CT scan data with the expected data, and combine the similarity calculation results of the X-ray data and CT scan data to calculate the first evaluation index, which is used to indicate the degree of difference between the treatment effect of the treatment cycle and the expected effect.

[0056] Determine whether the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold. If the first evaluation index corresponding to the last treatment cycle is not greater than the set threshold, it is considered that the treatment effect does not meet expectations and the treatment needs to be stopped. If the first evaluation index corresponding to the last treatment cycle is greater than the set threshold, continue to observe the situation of subsequent treatment cycles and decide whether to continue or stop treatment based on the condition of the disease.

[0057] By calculating the first evaluation index, the difference between the treatment effect and the expected effect can be quantitatively evaluated, providing an objective reference basis to help make reasonable treatment decisions. When the treatment effect does not meet expectations, the treatment can be stopped in time based on the judgment of the first evaluation index to avoid unnecessary treatment delays and reduce the patient's pain and burden.

[0058] The calculation formula of the first evaluation index is as follows:

[0059]

[0060] Where j is the number of the treatment cycle, j = 1, 2, 3, ..., m, m is the total number of treatment cycles, pz j is the first evaluation index of the jth treatment cycle, C 1j is the X-ray data after treatment of the jth treatment cycle, C 2j is the expected X-ray data after treatment of the jth treatment cycle, f(C 1j,C 2j ) is the calculation function of the post-treatment X-ray data and the expected post-treatment X-ray data, D 1j is the CT scan data after treatment of the jth treatment cycle, D 2j is the expected CT scan data after treatment of the jth treatment cycle, f(D 1j ,D 2j ) is the calculation function of the CT scan data after treatment and the expected CT scan data after treatment.

[0061] The calculation function is used to calculate the similarity or difference between the medical record data after treatment and the expected condition data after treatment. In the formula, f(C 1j ,C 2j ) and f(D 1j ,D 2j ) two calculation functions, which are used to calculate the similarity of X-ray data and CT scan data respectively. Among them, the post-treatment medical record data refers to the clinical data of the patient after the treatment cycle, including the X-ray data and CT scan data after the treatment. These data are usually obtained through medical imaging technology, such as X-ray films or CT scan images. According to the patient's treatment situation, imaging examinations will be performed regularly to obtain the condition data after treatment. The expected condition data after treatment refers to the description or standard of the expected condition after each treatment cycle. These data are usually determined based on medical knowledge and clinical experience. For example, for different treatment plans and treatment cycles, the expected condition data will be set according to the patient's condition and treatment plan. The determination of the first evaluation threshold usually requires comprehensive consideration of multiple factors, including clinical practice experience, medical research data, characteristics of the treatment plan, and individual differences of patients. For example, for clinical practice experience and medical research data, doctors set the first evaluation threshold based on their own clinical practice experience and professional knowledge, combined with the changes in the patient's condition during the treatment process; or by analyzing relevant data for specific diseases or treatment plans, such as clinical trial results, epidemiological research results, etc., to obtain statistical information on the treatment effect, thereby determining the first evaluation threshold.

[0062] In the formula, zm j is the lung density after treatment in the post-treatment X-ray data of the jth treatment cycle, zcm j is the expected lung density after treatment in the expected X-ray data after treatment of the jth treatment cycle, zt j is the post-treatment lung field transmittance in the post-treatment X-ray data of the jth treatment cycle, zct i is the expected lung transmittance after treatment in the expected X-ray data after treatment of the jth treatment cycle, and e is a natural constant.

[0063] This formula is used to calculate the similarity between the post-treatment X-ray data and the expected X-ray data. First, the difference between the post-treatment lung density and the post-treatment lung field transmittance and the expected value is calculated, and then the exponential function is used to quantify this difference. The role of the exponential function is to map the difference to a non-negative real number, so that the smaller the difference, the closer the exponent is to 1, and the higher the similarity.

[0064] In the formula, zb j is the post-treatment lung mutation area in the post-treatment CT scan data of the jth treatment cycle, zbm j is the expected lung mutation area after treatment in the expected CT scan data after treatment of the jth treatment cycle, zp j is the patchy shadow area after treatment in the CT scan data after treatment of the jth treatment cycle, zpt j is the expected patchy shadow area after treatment in the expected CT scan data after treatment of the jth treatment cycle, zs j is the post-treatment lung lobe shrinkage in the post-treatment CT scan data of the jth treatment cycle, zst j is the expected amount of lung lobe shrinkage after treatment in the expected CT scan data after treatment of the jth treatment cycle, and e is a natural constant.

[0065] This formula is used to calculate the similarity between the post-treatment CT scan data and the expected CT scan data. Similarly, the difference between the post-treatment lung mutation area, patchy shadow area, and lobar atrophy and the expected value is first calculated, and then the exponential function is used to quantify this difference to calculate the similarity.

[0066] In this implementation scheme, by comparing and analyzing the medical record data after treatment with the expected condition data and calculating the similarity or difference, the relationship between the treatment effect and the expected effect can be quantified, which helps doctors to more accurately evaluate the treatment effect and thus guide subsequent treatment decisions. These formulas are based on mathematical models, objectively consider the differences between different indicators, avoid the influence of subjective factors on the evaluation of treatment effects, and make the evaluation more objective and scientific. By calculating the similarity or difference between the medical record data after treatment and the expected condition data, doctors can better understand whether the treatment effect meets expectations, thereby providing a basis for treatment decisions, adjusting the treatment plan in a timely manner, and improving the treatment effect.

[0067] Specifically, the process of determining whether to continue treatment or stop treatment based on the post-treatment medical record data of each treatment cycle is as follows: obtain the post-treatment medical record data of n treatment cycles before the last treatment cycle and number them; obtain the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle; if the difference index is greater than the set difference threshold, it means that the lung infection patient has an effect after receiving the treatment plan but has not achieved the expected effect of each treatment cycle, and the treatment should continue; if the difference index is not greater than the set difference threshold, it means that the lung infection patient has no effect after receiving the treatment plan and has not achieved the expected effect of each treatment cycle, and the treatment should be stopped and the treatment plan should be changed.

[0068] First, obtain the post-treatment medical record data of n treatment cycles before the last treatment cycle and number them. At the same time, obtain the post-treatment medical record data of the nth treatment cycle and the n+1th treatment cycle. Use the given calculation formula to calculate the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle. This difference index comprehensively considers the differences between multiple indicators in the post-treatment X-ray data and CT scan data. Compare the calculated difference index with the set difference threshold. If the difference index is greater than the set difference threshold, it means that the treatment effect has not met expectations and needs to continue treatment; if the difference index is not greater than the set difference threshold, it means that the treatment effect is good and you can consider stopping treatment. By comparing the differences between the post-treatment medical record data, it is possible to promptly discover whether the treatment effect has met expectations, so as to adjust the treatment plan in a timely manner to avoid worsening of the disease or delay in treatment.

[0069] The calculation formula of the difference index is as follows:

[0070]

[0071] In the formula, Cz (n,n+1) fm is the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle, n is the post-treatment lung density in the post-treatment X-ray data of the nth treatment cycle, fm n+1 is the lung density after treatment in the post-treatment X-ray data of the n+1th treatment cycle, ft n is the lung field transmittance in the post-treatment X-ray data of the nth treatment cycle, ft n+1 fs is the lung field transmittance in the post-treatment X-ray data of the n+1th treatment cycle, n fs is the lung mutation area in the CT scan data after treatment in the nth treatment cycle, n+1 fb is the lung mutation area in the CT scan data after treatment in the n+1th treatment cycle, nfb is the patchy shadow area in the CT scan data after treatment in the nth treatment cycle, n+1 is the patchy shadow area in the CT scan data after treatment for the n+1th treatment cycle, fz n is the amount of lung lobe shrinkage in the CT scan data after treatment in the nth treatment cycle, fz n+1 is the amount of lung lobe shrinkage in the CT scan data after treatment in the n+1th treatment cycle.

[0072] In this implementation, the formula takes into account multiple indicators such as lung density, lung field transmittance, lung mutation area, patchy shadow area and lobar atrophy in the X-ray data and CT scan data after treatment, and comprehensively evaluates the changes in the condition after treatment. Through the weight distribution in the formula, the importance of different indicators can be reasonably weighed, so as to more accurately reflect the overall situation of the treatment effect. By comparing the difference index with the set difference threshold, it can provide doctors with a standardized decision-making basis to help them determine whether it is necessary to adjust the treatment plan or stop treatment.

[0073] In summary, this application has at least the following effects:

[0074] By analyzing the medical records of patients with lung infections and classifying them into levels according to the severity of their illness, we can provide each patient with personalized treatment recommendations. We can update the decision-making support data set and treatment plan recommendations in real time, and evaluate and adjust them according to the patient's actual situation and changes in their condition. Through the expected effect evaluation subsystem, the system can evaluate the effectiveness of the treatment plan and promptly identify situations where the treatment is not effective, thereby adjusting the treatment plan to improve the treatment effect and the patient's recovery rate.

[0075] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0076] The present invention is described with reference to flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0077] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0078] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0079] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0080] 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. A decision support system based on medical record data analysis, characterized in that: It includes medical record data analysis subsystem, treatment plan recommendation subsystem and expected effect evaluation subsystem, among which: The medical record data analysis subsystem is used to extract the medical record data of patients with lung infection, obtain X-ray imaging data and CT scanning imaging data, and obtain an auxiliary decision-making data set based on the X-ray imaging data and CT scanning imaging data; The treatment plan recommendation subsystem is used to classify the severity of the condition of the lung infection patient according to the auxiliary decision-making data set obtained by the medical record data analysis subsystem, and recommend treatment plans based on the classified levels, each level corresponding to a treatment plan; The auxiliary decision data set includes an X-ray imaging feature data set and a CT scan imaging feature data set. The process of grading the severity of the condition of the lung infection patient based on the auxiliary decision data set is as follows: Acquire the reference data sets of each level of lung infection stored in the database, wherein the reference data sets of each level of lung infection include X-ray imaging reference feature data sets and CT scanning imaging reference feature data sets; Compare the decision-making support data set with the reference data sets of each level of lung infection to obtain the assessment values ​​of each condition; The level of the lung infection parameter data set corresponding to the minimum condition assessment value is calibrated as the severity of the lung infection patient's condition; The calculation formula of the disease assessment value is as follows: In the formula, i is the number of the disease level, i = 1, 2, 3, ..., n, n is the total number of disease levels, Bp i is the disease assessment value obtained by comparing the auxiliary decision-making data set with the i-th level lung infection reference data set, A1 is the X-ray imaging feature data set, is the X-ray imaging feature dataset in the i-th level lung infection dataset, is the calculation function of the X-ray imaging feature data set and the X-ray imaging reference feature data set, which is used to calculate the similarity between the X-ray imaging feature data set and the X-ray imaging reference feature data set. α1 is the calculation function The weight factor of the value, B1 is the CT scan imaging feature data set, The CT scan imaging parameter feature dataset in the i-th level lung infection parameter dataset, is the calculation function of the CT scan imaging feature data set and the CT scan imaging reference feature data set, which is used to calculate the similarity between the CT scan imaging feature data set and the CT scan imaging reference feature data set. α2 is the calculation function The weight factor of the value of ; The expected effect evaluation subsystem is used to evaluate the condition of the lung infection patient after the lung infection patient adopts the treatment plan.

2. The auxiliary decision-making system based on medical record data analysis according to claim 1, characterized in that: The X-ray imaging feature data set includes lung density and lung field transmittance of patients with lung infection, and the CT scan imaging feature data set includes lung mutation area, patchy shadow area and lung lobe atrophy of patients with lung infection; The X-ray imaging parameter feature data set includes the lung parameter density and lung field parameter transmittance of patients with lung infection, and the CT scan imaging parameter feature data set includes the lung mutation parameter area, patchy shadow parameter area and lung lobe parameter atrophy of patients with lung infection.

3. The auxiliary decision-making system based on medical record data analysis according to claim 1, characterized in that: The process for evaluating a patient with a lung infection after the patient has been treated is as follows: According to the set treatment cycle, the condition information of the lung infection patients who received the treatment plan was collected, the post-treatment medical record data was obtained, and the post-treatment condition data was extracted from the post-treatment medical record data; The extracted post-treatment condition data of each treatment cycle are compared and analyzed with the expected post-treatment condition data corresponding to each treatment cycle stored in the database to evaluate the condition of patients with lung infection and determine whether to continue or stop treatment based on the condition.

4. The auxiliary decision-making system based on medical record data analysis according to claim 2, characterized in that: The process of assessing the condition of a patient with a lung infection and determining whether to continue or stop treatment based on the condition is as follows: Compare and analyze the post-treatment medical record data of each treatment cycle with the corresponding expected condition data of each treatment cycle to obtain a first evaluation index, wherein the first evaluation index is used to indicate the difference between the treatment effect of each treatment cycle and the expected effect; Determine whether the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold, and if the first evaluation index corresponding to the last treatment cycle is not greater than the set first evaluation threshold, stop the treatment; If the first evaluation index corresponding to the last treatment cycle is greater than the set first evaluation threshold, it is determined whether to continue the treatment or stop the treatment based on the post-treatment medical record data of each treatment cycle.

5. The auxiliary decision-making system based on medical record data analysis according to claim 3 is characterized in that: The process of determining whether to continue or stop treatment based on the medical record data after each treatment cycle is as follows: Obtain the post-treatment medical record data of n treatment cycles before the last treatment cycle and number them; Obtain the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle; If the difference index is greater than the set difference threshold, it means that the lung infection patient has an effect after receiving the treatment plan but has not achieved the expected effect of each treatment cycle, and the treatment should continue; If the difference index is not greater than the set difference threshold, it means that the lung infection patient has not produced any effect after receiving the treatment plan and has not achieved the expected effect of each treatment cycle. In this case, the treatment will be stopped and the treatment plan will be changed.

6. The auxiliary decision-making system based on medical record data analysis according to claim 5, characterized in that: The post-treatment condition data includes post-treatment X-ray data and post-treatment CT scan data; The expected condition data after treatment includes expected X-ray data after treatment and expected CT scan data after treatment.

7. The auxiliary decision-making system based on medical record data analysis according to claim 6, characterized in that: The calculation formula of the first evaluation index is as follows: Where j is the number of the treatment cycle, j = 1, 2, 3, ..., m, m is the total number of treatment cycles, Pz j is the first evaluation index of the jth treatment cycle, C 1j is the X-ray data after treatment of the jth treatment cycle, C 2j is the expected X-ray data after treatment of the jth treatment cycle, f(C 1j ,C 2j ) is the calculation function of the post-treatment X-ray data and the expected post-treatment X-ray data, D 1j is the CT scan data after treatment of the jth treatment cycle, D 2j is the expected CT scan data after treatment of the jth treatment cycle, f(D 1j ,D 2j ) is the calculation function of the CT scan data after treatment and the expected CT scan data after treatment.

8. The auxiliary decision-making system based on medical record data analysis according to claim 6, characterized in that: The calculation formula of the difference index is as follows: In the formula, Cz (n,n+1) fm is the difference index between the post-treatment medical record data of the nth treatment cycle and the post-treatment medical record data of the n+1th treatment cycle, n is the post-treatment lung density in the post-treatment X-ray data of the nth treatment cycle, fm n+1 is the lung density after treatment in the post-treatment X-ray data of the n+1th treatment cycle, ft n is the lung field transmittance in the post-treatment X-ray data of the nth treatment cycle, ft n+1 fs is the lung field transmittance in the post-treatment X-ray data of the n+1th treatment cycle, n fs is the lung mutation area in the CT scan data after treatment in the nth treatment cycle, n+1 fb is the lung mutation area in the CT scan data after treatment in the n+1th treatment cycle, n fb is the patchy shadow area in the CT scan data after treatment in the nth treatment cycle, n+1 is the patchy shadow area in the CT scan data after treatment for the n+1th treatment cycle, fz n is the amount of lung lobe shrinkage in the CT scan data after treatment in the nth treatment cycle, fz n+1 is the amount of lung lobe shrinkage in the CT scan data after treatment in the n+1th treatment cycle.

Images