Method and system for predicting postoperative life quality of prostate cancer patient and medium

By constructing a multi-dimensional prediction model based on preoperative data, the problem of inaccurate postoperative quality of life assessment in the prior art in the prior art is solved, and personalized rehabilitation suggestions and improvement of postoperative recovery results are achieved.

CN118942693APending Publication Date: 2024-11-12WEST CHINA HOSPITAL SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410950834.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The prior art lacks a multi-dimensional quality of life prediction system that can comprehensively consider the preoperative clinical baseline characteristics and MRI parameters of prostate cancer patients, which makes it difficult to evaluate and manage postoperative quality of life, and the prediction is not accurate enough, affecting the patient's recovery effect.

Method used

By collecting the preoperative clinical baseline characteristics and MRI parameters of patients, a multidimensional prediction model was constructed using single-factor analysis, multi-factor regression analysis and Lasso regression analysis, predictive scores of postoperative quality of life were generated, and personalized rehabilitation suggestions were provided.

Benefits of technology

Accurate multi-dimensional prediction of the quality of life of prostate cancer patients after surgery is achieved, personalized rehabilitation suggestions are provided, postoperative recovery effect is improved, the impact of complications is reduced, and clinical management and patient satisfaction are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118942693A_ABST
    Figure CN118942693A_ABST
Patent Text Reader

Abstract

The invention provides a method and system for predicting the postoperative life quality of a prostate cancer patient and a medium. The method comprises the following steps: collecting preoperative clinical baseline characteristics (such as age, height, weight, BMI, complication, PSA, TURP medical history, Gleason score, clinical T staging and pathological needle biopsy results) and preoperative MRI parameters (such as membrane urethra length MUL, membrane urethra angle, prostate length, symphysis pubis-prostate tip length PAL, prostate height, prostate length PAL, prostate length PAL and prostate length PAL) of a prostate cancer patient, and collecting the preoperative clinical baseline characteristics (such as age, height, weight, BMI, complication, PSA, TURP medical history, Gleason score, clinical T staging and pathological needle biopsy results) of the prostate cancer patient. Prostate width, left / right obturator inner muscle thickness, levator outer edge spacing, levator inner edge spacing, transverse membrane urethra thickness, anterior and posterior membrane urethra thickness, left / right anal sphincter thickness and urethra wall thickness); inputting the collected data into a prediction model library; based on the prediction model library, performing multi-dimensional prediction on the individual RARP postoperative life quality by using single-factor analysis, multi-factor regression analysis and Lasso regression analysis; and generating a prediction score, and providing postoperative rehabilitation suggestions according to the prediction score.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of postoperative monitoring of prostate diseases, and in particular to a method, system and medium for predicting the postoperative quality of life of prostate cancer patients. Background Art

[0002] Prostate cancer is the second most common malignant tumor among men worldwide, and its mortality rate ranks fifth among all cancers in men. 8.2% of the new cases of prostate cancer in the world each year come from China, and deaths caused by it account for 13.6% of the world's total. Overall, the incidence of prostate cancer in my country is lower than the global average, but it has shown a significant upward trend in recent years, causing a huge health burden on the Chinese population.

[0003] Radical prostatectomy (RP) is an important treatment for prostate cancer, which is treated by surgical removal of the entire prostate, bilateral seminal vesicles and part of the urethra. Robotic-assisted radical prostatectomy (RARP) has become an important surgical method due to its higher precision and better oncological and functional outcomes. However, although RARP can improve surgical outcomes, patients still face many complications after surgery, such as urinary incontinence, erectile dysfunction, urethral irritation and obstructive symptoms, etc. These complications seriously affect the quality of life of patients after surgery.

[0004] Current studies have mostly focused on predicting a single dimension of postoperative quality of life. For example, Jeong et al. developed a model and validated it with a nomogram to predict immediate, short-term, and long-term recovery of urinary continence after surgery; Mulhall et al. developed a prediction model for functional erectile recovery and severe erectile dysfunction 2 years after RP surgery; Collette et al. combined preoperative clinical parameters and surgical experience to construct a urinary continence prediction model; Honda et al. incorporated MRI features to develop a prediction model for short-term postoperative urinary continence. Although these studies provide prediction tools for a single dimension, they lack a multidimensional quality of life prediction system that comprehensively considers preoperative clinical baseline characteristics and MRI parameters, making it difficult to comprehensively evaluate and manage patients' postoperative quality of life in clinical practice.

[0005] Quality of life has become an important tool for evaluating the impact of disease and treatment on patients' daily lives, and its assessment is particularly important for diseases where treatment decisions may be affected by expected quality of life outcomes. The Expanded Prostate Cancer Composite Index is the most suitable specific survey for evaluating urological prostate cancer patients. It consists of 26 items and evaluates the patient's quality of life in five aspects: urinary incontinence, urethral irritation and urethral obstruction symptoms, bowel function, sexual function, and hormone levels.

[0006] Although robot-assisted radical prostatectomy has improved the surgical effect to a certain extent, there is still uncertainty about the recovery of postoperative quality of life. Currently, there is a lack of a multidimensional prediction system that can combine the preoperative clinical baseline characteristics and MRI parameters of prostate cancer patients and can more accurately predict the postoperative quality of life of patients, so as to guide the clinic to better prevent and treat postoperative related complications. To this end, the present invention provides a new prediction system, which aims to provide an effective tool for the prediction and management of postoperative quality of life by comprehensively analyzing the patient's preoperative data.

[0007] This new prediction system can not only help clinicians better understand and predict patients' postoperative quality of life before surgery, but also provide patients with personalized rehabilitation advice, improve their postoperative quality of life, and reduce the impact of postoperative complications.

[0008] Therefore, we urgently need to design a smart lock control method and a smart lock control system to solve the problems pointed out in the background technology. Summary of the invention

[0009] The object of the present invention is to provide a method, system and medium for predicting the quality of life of prostate cancer patients after surgery, so as to solve the problems raised in the above-mentioned background technology.

[0010] To achieve the above object, the present invention provides the following technical solution: a method for predicting the quality of life of patients with prostate cancer after surgery, the method comprising the following steps:

[0011] S1. Collect the preoperative clinical baseline characteristics of prostate cancer patients, including age, height, weight, BMI, comorbidities, prostate-specific antigen (PSA), history of transurethral resection of the prostate (TURP), Gleason score, clinical T stage, and pathological puncture biopsy results;

[0012] S2. Collect preoperative MRI parameters of patients with prostate cancer, including membranous urethra length (MUL), membranous urethra angle, prostate length, symphysis pubis-prostate tip length (PAL), prostate height, prostate width, left / right obturator internus muscle thickness, levator lateral edge distance, levator medial edge distance, transverse membranous urethra thickness, anterior and posterior membranous urethra thickness, left / right anal sphincter thickness, and urethral wall thickness;

[0013] S3, input the collected data into the prediction model library;

[0014] S4. Based on the prediction model library, univariate analysis, multivariate regression analysis and Lasso regression analysis were used to perform multidimensional prediction of individual post-RARP quality of life;

[0015] S5. Generate a prediction score and provide postoperative rehabilitation suggestions based on the prediction score.

[0016] As a preferred technical solution of the present invention, the prediction model library includes multiple prediction models related to postoperative short-term and long-term urinary incontinence, urethral irritation and urethral obstruction, intestinal function, sexual function and hormone levels.

[0017] As a preferred technical solution of the present invention, the construction of the prediction model adopts the least absolute shrinkage and selection operator (Lasso) regression analysis, combined with the prediction factors screened by univariate and multivariate regression analysis.

[0018] As a preferred technical solution of the present invention, the model is evaluated by the receiver operating characteristic (ROC) curve and the area under the curve (AUC), and the AUC values ​​are all greater than 0.7 to ensure the accuracy and reliability of the prediction model.

[0019] As a preferred technical solution of the present invention, the quality of life prediction system includes the following modules:

[0020] Data inspection module, used to inspect and clean the collected clinical baseline characteristics and MRI parameter data;

[0021] A data upload module is used to upload the checked data to the prediction model library;

[0022] The prediction scoring module is used to run the prediction model and calculate the short-term and long-term multi-dimensional quality of life prediction scores after RARP surgery for individuals;

[0023] The prediction scoring module further comprises:

[0024] A data processing unit, used for format conversion and logic checking of uploaded data;

[0025] The model running unit is used to call the corresponding model in the prediction model library for calculation;

[0026] The score generation unit is used to summarize the prediction results of each dimension and generate a comprehensive quality of life prediction score.

[0027] The present invention also provides a system for predicting the quality of life of patients with prostate cancer after surgery, the system comprising:

[0028] The clinical information front end is used to collect preoperative clinical baseline characteristics and preoperative MRI parameters of prostate cancer patients;

[0029] Prediction model library, used to store and manage multi-dimensional prediction models of postoperative quality of life;

[0030] The quality of life prediction system is used to generate a prediction score for the quality of life of an individual after RARP surgery based on the data from the clinical information front end and the prediction model library.

[0031] As a preferred technical solution of the present invention, the clinical information front end includes:

[0032] A data input interface, used to input and store the patient's preoperative clinical baseline characteristics and preoperative MRI parameters;

[0033] Data processing module, used to format and standardize input data.

[0034] As a preferred technical solution of the present invention, the prediction model library includes:

[0035] Postoperative urinary incontinence prediction model, used to predict short-term and long-term urinary incontinence of patients after surgery;

[0036] Postoperative urethral irritation and urethral obstruction prediction model, used to predict postoperative urethral-related symptoms in patients;

[0037] Postoperative intestinal function prediction model, used to predict the recovery of patients' postoperative intestinal function;

[0038] Postoperative sexual function prediction model, used to predict the patient's postoperative sexual function recovery;

[0039] The postoperative hormone level prediction model is used to predict the changes in hormone levels of patients after surgery.

[0040] As a preferred technical solution of the present invention, the quality of life prediction system includes: a data checking module for verifying the integrity and accuracy of input data;

[0041] The prediction calculation module is used to call the model in the prediction model library for calculation;

[0042] The result output module is used to generate and display the patient's postoperative quality of life prediction score and corresponding rehabilitation suggestions.

[0043] The present invention also provides a storage medium for storing computer executable instructions, which, when executed, implement the above-mentioned system for predicting the postoperative quality of life of prostate cancer patients.

[0044] Beneficial Effects

[0045] First, by comprehensively collecting and integrating the preoperative clinical baseline characteristics and MRI parameters of prostate cancer patients, this method can provide detailed and accurate preoperative data of patients. These data include age, height, weight, BMI, comorbidities, PSA, TURP history, Gleason score, clinical T stage, membranous urethra length MUL, membranous urethra angle, prostate length and other indicators, which help to fully understand the health status and disease characteristics of patients and provide a solid foundation for subsequent predictive analysis.

[0046] Secondly, using univariate analysis, multivariate regression analysis and Lasso regression analysis, this method can conduct in-depth statistical analysis and model construction on the collected patient data. Through these advanced statistical analysis methods, key factors closely related to postoperative quality of life can be identified, and accurate multidimensional prediction models can be constructed, thereby achieving scientific prediction of postoperative quality of life after RARP. This not only improves the accuracy of the prediction, but also provides personalized prediction results for different patients.

[0047] Finally, based on the constructed prediction model, this method can generate an individualized postoperative quality of life prediction score and provide specific postoperative rehabilitation recommendations based on it. This personalized rehabilitation recommendation helps patients take appropriate rehabilitation measures after surgery, improve their quality of life, reduce the occurrence of complications, and improve the effect of postoperative recovery. In this way, clinicians can better manage the patient's postoperative rehabilitation process and improve the quality of medical services and patient satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 The present invention is a flow chart of a method for predicting the quality of life of patients with prostate cancer after surgery. DETAILED DESCRIPTION

[0049] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0050] I. General Overview

[0051] The present invention provides a method and system for predicting the quality of life of patients with prostate cancer after surgery. By comprehensively analyzing the preoperative clinical baseline characteristics and preoperative MRI parameters of prostate cancer patients, a multi-dimensional quality of life prediction model is constructed. The system consists of a clinical information front end, a prediction model library, and a quality of life prediction system to achieve a comprehensive prediction of the patient's postoperative quality of life.

[0052] 2. System Architecture

[0053] The system mainly includes the following modules:

[0054] Clinical information front end: used to collect and process patients' preoperative data.

[0055] Prediction model library: stores multiple models for predicting quality of life in different dimensions.

[0056] Quality of life prediction system: Based on the collected data and prediction model, it generates a quality of life prediction score and provides personalized rehabilitation recommendations.

[0057] 3. Implementation steps

[0058] Clinical information front-end data collection:

[0059] 1.1 Collection of basic patient information:

[0060] Preoperative clinical baseline characteristics included age, height, weight, BMI, comorbidities, prostate-specific antigen (PSA), history of transurethral resection of the prostate (TURP), Gleason score, clinical T stage, and pathological puncture biopsy results.

[0061] Preoperative MRI parameters: MRI examination was performed before surgery to obtain data such as membranous urethra length (MUL), membranous urethra angle, prostate length, pubic symphysis-prostatic apex length (PAL), prostate height, prostate width, left / right obturator internus muscle thickness, levator outer edge distance, levator inner edge distance, transverse membranous urethra thickness, anterior and posterior membranous urethra thickness, left / right anal sphincter thickness, and urethral wall thickness.

[0062] 1.2 Data standardization and processing:

[0063] The collected clinical baseline characteristics and MRI parameter data were standardized to ensure the consistency and accuracy of the data.

[0064] Prediction model library construction:

[0065] 2.1 Univariate analysis:

[0066] Univariate analysis was performed on the correlation between each variable and each dimension of postoperative quality of life to preliminarily screen out potential predictive factors.

[0067] 2.2 Multivariate regression analysis:

[0068] The potential predictive factors screened out by univariate analysis were subjected to multivariate regression analysis to further identify factors that had a significant impact on postoperative quality of life.

[0069] 2.3Lasso regression analysis:

[0070] The least absolute shrinkage and selection operator (Lasso) regression analysis methods were used to further screen the results of the multivariate regression analysis and ultimately determine the key factors for building a prediction model.

[0071] 2.4 Prediction model construction:

[0072] Based on the selected key factors, logistic regression, linear regression and other methods were used to construct multiple models for predicting different dimensions of postoperative quality of life, including short-term and long-term urinary incontinence, urethral irritation and urethral obstruction, intestinal function, sexual function and hormone levels.

[0073] 2.5 Model Evaluation:

[0074] The receiver operating characteristic (ROC) curve and area under the curve (AUC) were used to evaluate the prediction model to ensure that the model had high prediction accuracy and reliability. AUC values ​​greater than 0.7 indicated that the model had good discrimination ability.

[0075] Quality of life prediction system operation:

[0076] 3.1 Data inspection module:

[0077] Check the input data for completeness and accuracy to ensure that the data complies with the input format.

[0078] 3.2 Data upload module:

[0079] Upload the checked data to the prediction model library and record the front-end individual data number.

[0080] 3.3 Prediction calculation module:

[0081] The corresponding models in the prediction model library were run to calculate the short-term and long-term multidimensional quality of life prediction scores after RARP surgery.

[0082] 3.4 Result output module:

[0083] Based on the prediction score, a detailed prediction report is generated, including the quality of life score in each dimension and the comprehensive score.

[0084] Provide personalized postoperative rehabilitation advice to help patients develop reasonable rehabilitation plans and improve their quality of life after surgery.

[0085] IV. Application Examples

[0086] Example 1: Urinary incontinence prediction:

[0087] The patients' preoperative age, BMI, Charlson score, MUL, PAL and other parameters were collected.

[0088] Input data into the prediction model library and run the urinary incontinence prediction model.

[0089] Generate a prediction score for short-term and long-term urinary incontinence after surgery and provide corresponding rehabilitation suggestions, such as pelvic floor muscle exercises, dietary adjustments, etc.

[0090] Example 2: Sexual function prediction:

[0091] The patients' information including age, BMI, TURP surgery history, clinical T stage, and Gleason score were collected.

[0092] Input data into the prediction model library and run the sexual function prediction model.

[0093] Generate a predictive score for the patient's postoperative sexual function recovery and provide corresponding rehabilitation suggestions, such as drug treatment, psychological counseling, etc.

[0094] Example 3: Comprehensive quality of life prediction:

[0095] Preoperative clinical baseline characteristics and preoperative MRI parameters were comprehensively collected.

[0096] Input data into the prediction model library and run the quality of life prediction model for each dimension separately.

[0097] Generate a comprehensive predictive score for the patient's multi-dimensional quality of life after surgery, and provide an overall rehabilitation plan, including advice on diet, exercise, psychological counseling, and other aspects.

[0098] 5. System advantages

[0099] Multi-dimensional prediction: The system of the present invention can comprehensively predict multiple dimensions of the patient's postoperative quality of life, not limited to a single indicator, and provide a more comprehensive evaluation result.

[0100] Personalized advice: Based on the predicted results, personalized postoperative rehabilitation advice is provided to help patients recover better and improve their quality of life.

[0101] High Accuracy: Advanced data analysis and regression methods are used to ensure that the prediction model has high accuracy and reliability.

[0102] Ease of operation: The system is simple in design and easy to operate, allowing medical staff to quickly master and apply it in clinical practice.

[0103] VI. Conclusion

[0104] The present invention provides an innovative method and system for predicting the postoperative quality of life of prostate cancer patients. By combining preoperative clinical baseline characteristics and MRI parameters and utilizing advanced data analysis and prediction modeling techniques, it can accurately and comprehensively predict the multi-dimensional quality of life after surgery, thus providing strong support for clinical decision-making and patient rehabilitation, and helping to improve the patient's postoperative quality of life.

[0105] 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 method for predicting the quality of life of patients with prostate cancer after surgery. It is characterized in that The method comprises the following steps: S1. Collect the preoperative clinical baseline characteristics of prostate cancer patients, including age, height, weight, BMI, comorbidities, prostate-specific antigen (PSA), history of transurethral resection of the prostate (TURP), Gleason score, clinical T stage, and pathological puncture biopsy results; S2. Collect preoperative MRI parameters of patients with prostate cancer, including membranous urethra length (MUL), membranous urethra angle, prostate length, symphysis pubis-prostate tip length (PAL), prostate height, prostate width, left / right obturator internus muscle thickness, levator lateral edge distance, levator medial edge distance, transverse membranous urethra thickness, anterior and posterior membranous urethra thickness, left / right anal sphincter thickness, and urethral wall thickness; S3, input the collected data into the prediction model library; S4. Based on the prediction model library, univariate analysis, multivariate regression analysis and Lasso regression analysis were used to perform multidimensional prediction of individual post-RARP quality of life; S5. Generate a prediction score and provide postoperative rehabilitation suggestions based on the prediction score.

2. The method for predicting the quality of life of patients with prostate cancer after surgery according to claim 1, It is characterized in that The prediction model library includes multiple prediction models related to postoperative short-term and long-term urinary incontinence, urethral irritation and urethral obstruction, intestinal function, sexual function and hormone levels.

3. The method for predicting the quality of life of patients with prostate cancer after surgery according to claim 2, It is characterized in that The prediction model was constructed using least absolute shrinkage and selection operator Lasso regression analysis, combined with predictive factors screened by univariate and multivariate regression analysis.

4. The method for predicting the quality of life of patients with prostate cancer after surgery according to claim 3, It is characterized in that The model was evaluated by receiver operating characteristic (ROC) curve and area under the curve (AUC), and the AUC values ​​were all greater than 0.7 to ensure the accuracy and reliability of the prediction model.

5. The method for predicting the quality of life of patients with prostate cancer after surgery according to claim 1, It is characterized in that The quality of life prediction system includes the following modules: Data inspection module, used to inspect and clean the collected clinical baseline characteristics and MRI parameter data; A data upload module is used to upload the checked data to the prediction model library; The prediction scoring module is used to run the prediction model and calculate the short-term and long-term multi-dimensional quality of life prediction scores after RARP surgery for individuals; The prediction scoring module further comprises: A data processing unit, used for format conversion and logic checking of uploaded data; The model running unit is used to call the corresponding model in the prediction model library for calculation; The score generation unit is used to summarize the prediction results of each dimension and generate a comprehensive quality of life prediction score.

6. A system for predicting the quality of life of patients with prostate cancer after surgery. It is characterized in that The system includes: The clinical information front end is used to collect preoperative clinical baseline characteristics and preoperative MRI parameters of prostate cancer patients; Prediction model library, used to store and manage multi-dimensional prediction models of postoperative quality of life; The quality of life prediction system is used to generate a prediction score for the quality of life of an individual after RARP surgery based on the data from the clinical information front end and the prediction model library.

7. The system for predicting the quality of life of patients with prostate cancer after surgery according to claim 6, It is characterized in that The clinical information front end includes: A data input interface, used to input and store the patient's preoperative clinical baseline characteristics and preoperative MRI parameters; Data processing module, used to format and standardize input data.

8. The system for predicting the quality of life of patients with prostate cancer after surgery according to claim 6, It is characterized in that The prediction model library includes: Postoperative urinary incontinence prediction model, used to predict short-term and long-term urinary incontinence of patients after surgery; Postoperative urethral irritation and urethral obstruction prediction model, used to predict postoperative urethral-related symptoms in patients; Postoperative intestinal function prediction model, used to predict the recovery of patients' postoperative intestinal function; Postoperative sexual function prediction model, used to predict the patient's postoperative sexual function recovery; The postoperative hormone level prediction model is used to predict the changes in hormone levels of patients after surgery.

9. The system for predicting the quality of life of patients with prostate cancer after surgery according to claim 6, It is characterized in that The quality of life prediction system includes: a data checking module for verifying the integrity and accuracy of input data; The prediction calculation module is used to call the model in the prediction model library for calculation; The result output module is used to generate and display the patient's postoperative quality of life prediction score and corresponding rehabilitation suggestions.

10. A storage medium for storing computer executable instructions, Features: When executed, the computer executable instructions implement the system for predicting the postoperative quality of life of prostate cancer patients according to any one of claims 6 to 9.