Clinical prognosis model construction method for medical oncology

Abstract

The invention belongs to the field of medical oncology clinic, and particularly relates to a medical oncology clinic prognosis model construction method which comprises the following steps: S1, constructing a prognosis model construction system framework, the prognosis model construction system framework comprises a tumor data collection module, a tumor data screening module, a prognosis model construction module, a model training module, a model optimization module, a prognosis model evaluation module, an evaluation result analysis module, a storage module, an early warning module, a reason analysis module, a solution suggestion module and an early warning mode setting module; an updating time setting module, an updating reminding module and a prognosis model updating module; and S2, medical oncology clinical data are collected through a tumor data collection module, the collected data are screened through a tumor data screening module, and due to construction of the prognosis model, doctors can better understand the conditions of patients, so that a more effective treatment scheme is formulated.

Description

Technical Field

[0001] This invention relates to the field of clinical oncology technology, and in particular to a method for constructing a prognostic model for clinical use in oncology. Background Technology

[0002] Medical oncology is a sub-department of oncology, belonging to internal medicine. It has a very broad scope of diagnosis and treatment, mainly focusing on the medical treatment of various benign and malignant tumors. Based on the patient's symptoms and condition, it adopts a comprehensive treatment approach, including radiotherapy, chemotherapy, targeted therapy, and immunotherapy, to alleviate the patient's symptoms and control the disease. However, surgery is not performed. The prognosis is based on the prediction of the disease's progression through experience. Prognostic analysis is very practical in clinical practice and has a guiding role in clinical practice. Existing technologies are not conducive to predicting patient prognosis and planning subsequent treatment plans. Developing solutions after problems arise is time-consuming. Therefore, we propose a method for constructing a clinical prognostic model in oncology to address these issues. Summary of the Invention

[0003] The purpose of this invention is to address the shortcomings of existing technologies, such as difficulty in predicting patient prognosis, hindering the planning of subsequent treatment plans, and the time-consuming process of developing solutions after problems arise. Therefore, this invention proposes a method for constructing a prognostic model for clinical use in oncology.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: A method for constructing a prognostic model for clinical use in oncology includes the following steps: S1. Establish a prognostic model construction system framework, which includes a tumor data collection module, a tumor data screening module, a prognostic model construction module, a model training module, a model optimization module, a prognostic model evaluation module, an evaluation result analysis module, a storage module, an early warning module, a cause analysis module, a solution suggestion module, an early warning method setting module, an update time setting module, an update reminder module, and a prognostic model update module. S2. The tumor data collection module collects clinical data from the Department of Oncology, the tumor data filtering module filters the collected data, selects variables related to prognosis, and transmits the selected variables to the prognosis model construction module. The prognosis model construction module constructs a prognosis model based on the received data. S3. The prognostic model is trained through the model training module, optimized through the model optimization module, evaluated through the prognostic model evaluation module, and the performance of the evaluation is analyzed and judged through the evaluation result analysis module. S4. When the evaluated performance fails to meet the requirements, an early warning module is used to issue an early warning, a cause analysis module is used to analyze the reasons for the performance failure, and a solution suggestion module is used to provide solution suggestions.

[0005] Preferably, the warning mode setting module sets the warning mode of the warning module. The warning mode setting module includes an audio-visual setting unit, which is connected to an information warning setting unit, which is connected to a voice warning setting unit, and the voice warning setting unit is connected to a warning frequency setting unit.

[0006] Preferably, the early warning module includes an audible and visual early warning unit, which is connected to an information early warning unit, which is connected to a voice early warning unit. The audible and visual early warning unit is used for audible and visual early warning, the information early warning unit is used to send information to management personnel for early warning, and the voice early warning unit is used to provide voice early warning.

[0007] Preferably, the solution suggestion module includes a cause identification unit, which is connected to a solution matching unit, which is connected to a solution feedback unit, and the solution feedback unit is connected to a feedback recording unit.

[0008] Preferably, the cause identification unit is used to identify the cause of non-compliance, the solution matching unit matches a solution based on the identified cause, and the solution feedback unit feeds back the matched solution to the management personnel.

[0009] Preferably, the update time setting module is used to set the update frequency of the prognostic model, the update reminder module is used to send a reminder to the administrator when an update is required, and the prognostic model update module updates the prognostic model of the prognostic model construction module after the set update time is reached.

[0010] Preferably, the tumor data collection module includes a clinical feature collection unit, which is connected to a biomarker collection unit, which is connected to a clinical outcome collection unit, and the clinical outcome collection unit is connected to a classification and transmission unit.

[0011] Preferably, the clinical feature collection unit is used to collect tumor clinical features, the biomarker collection unit is used to collect biomarkers, the clinical outcome collection unit is used to collect clinical outcome data, and the classification and transmission unit is used to classify and transmit the collected data.

[0012] Preferably, in step S3, the prognostic model is trained using a deep learning algorithm in the model training module, the model is optimized using the model optimization module, the model performance is evaluated using the prognostic model evaluation module, and the performance of the evaluation is analyzed and judged using the evaluation result analysis module. When optimizing the model, the feature space is expanded and more variables that affect the development of the disease, such as lifestyle habits and family history, are added to improve the interpretability of the model.

[0013] Preferably, the tumor data collection module is connected to the tumor data screening module, the tumor data screening module is connected to the prognostic model construction module, the update time setting module is connected to the update reminder module, the update reminder module is connected to the prognostic model update module, the prognostic model update module is connected to the prognostic model construction module, the prognostic model construction module is connected to the prognostic model evaluation module and the model training module, the model training module is connected to the model optimization module, the prognostic model evaluation module is connected to the evaluation result analysis module, the evaluation result analysis module is connected to the storage module and the early warning module, the early warning module is connected to the cause analysis module, the cause analysis module is connected to the solution suggestion module, and the early warning method setting module is connected to the early warning module.

[0014] The beneficial effects of the method for constructing a clinical prognostic model for oncology described in this invention are as follows: The tumor data collection module collects clinical data from the Department of Oncology, the tumor data filtering module filters the collected data, selects variables related to prognosis, and transmits the selected variables to the prognosis model building module. The prognosis model building module builds a prognosis model based on the received data. The prognostic model is trained through the model training module, optimized through the model optimization module, evaluated through the prognostic model evaluation module, and analyzed and judged through the evaluation result analysis module. When the evaluation performance fails to meet the requirements, an early warning module issues an early warning, a cause analysis module analyzes the reasons for the performance failure, and a solution suggestion module provides solutions. The prognostic model developed in this invention helps doctors better understand the patient's condition, thereby enabling them to formulate more effective treatment plans. Attached Figure Description

[0015] Figure 1 This is a system framework diagram of a prognostic model construction method for clinical use in oncology proposed in this invention; Figure 2 This is a block diagram of the tumor data collection module in a method for constructing a prognostic model for clinical use in oncology proposed in this invention. Figure 3 This is a block diagram of the early warning module of a method for constructing a prognostic model for clinical use in oncology proposed in this invention; Figure 4 This is a block diagram of the early warning mode setting module of the prognostic model construction method for clinical use in oncology proposed in this invention; Figure 5 This is a block diagram of a proposed solution module for a method of constructing a prognostic model for clinical use in oncology, as suggested in this invention. Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0017] Example 1 Reference Figures 1-5 A method for constructing a prognostic model for clinical use in oncology includes the following steps: S1. Establish a prognostic model construction system framework, which includes a tumor data collection module, a tumor data screening module, a prognostic model construction module, a model training module, a model optimization module, a prognostic model evaluation module, an evaluation result analysis module, a storage module, an early warning module, a cause analysis module, a solution suggestion module, an early warning method setting module, an update time setting module, an update reminder module, and a prognostic model update module. S2. The tumor data collection module collects clinical data from the Department of Oncology, the tumor data filtering module filters the collected data, selects variables related to prognosis, and transmits the selected variables to the prognosis model construction module. The prognosis model construction module constructs a prognosis model based on the received data. S3. The prognostic model is trained through the model training module, optimized through the model optimization module, evaluated through the prognostic model evaluation module, and the performance of the evaluation is analyzed and judged through the evaluation result analysis module. S4. When the evaluated performance fails to meet the requirements, an early warning module is used to issue an early warning, a cause analysis module is used to analyze the reasons for the performance failure, and a solution suggestion module is used to provide solution suggestions.

[0018] In this embodiment, the warning mode setting module sets the warning mode of the warning module. The warning mode setting module includes an audio-visual setting unit, which is connected to an information warning setting unit, which is connected to a voice warning setting unit, and the voice warning setting unit is connected to a warning frequency setting unit.

[0019] In this embodiment, the early warning module includes an audible and visual early warning unit, which is connected to an information early warning unit, and the information early warning unit is connected to a voice early warning unit. The audible and visual early warning unit is used for audible and visual early warning, the information early warning unit is used to send information to management personnel for early warning, and the voice early warning unit is used for voice early warning.

[0020] In this embodiment, the solution suggestion module includes a cause identification unit, which is connected to a solution matching unit, which is connected to a solution feedback unit, and the solution feedback unit is connected to a feedback recording unit.

[0021] In this embodiment, the cause identification unit is used to identify the reasons for non-compliance, the solution matching unit matches solutions according to the identified reasons, and the solution feedback unit feeds back the matched solutions to the management personnel.

[0022] In this embodiment, the update time setting module is used to set the update frequency of the prognostic model, the update reminder module is used to send a reminder to the administrator when an update is required, and the prognostic model update module updates the prognostic model of the prognostic model construction module after the set update time is reached.

[0023] In this embodiment, the tumor data collection module includes a clinical feature collection unit, which is connected to a biomarker collection unit, which is connected to a clinical outcome collection unit, and the clinical outcome collection unit is connected to a classification and transmission unit.

[0024] In this embodiment, the clinical feature collection unit is used to collect tumor clinical features, the biomarker collection unit is used to collect biomarkers, the clinical outcome collection unit is used to collect clinical outcome data, and the classification and transmission unit is used to classify and transmit the collected data.

[0025] In this embodiment, in S3, the prognostic model is trained using the deep learning algorithm of the model training module, the model is optimized using the model optimization module, the model performance is evaluated using the prognostic model evaluation module, and the performance of the evaluation is analyzed and judged using the evaluation result analysis module. When optimizing the model, the feature space is expanded and more variables that affect the development of the disease are added, such as lifestyle habits and family history, to improve the interpretability of the model.

[0026] In this embodiment, the tumor data collection module is connected to the tumor data screening module, the tumor data screening module is connected to the prognostic model construction module, the update time setting module is connected to the update reminder module, the update reminder module is connected to the prognostic model update module, the prognostic model update module is connected to the prognostic model construction module, the prognostic model construction module is connected to the prognostic model evaluation module and the model training module, the model training module is connected to the model optimization module, the prognostic model evaluation module is connected to the evaluation result analysis module, the evaluation result analysis module is connected to the storage module and the early warning module, the early warning module is connected to the cause analysis module, the cause analysis module is connected to the solution suggestion module, and the early warning method setting module is connected to the early warning module.

[0027] The tumor data screening module performs univariate regression analysis on each prognostic gene in each combination of prognostic genes to determine the significance level of each prognostic gene: ; ;in, Represents the risk function; Represents the regression coefficient; Represents the baseline risk function; This represents the maximum value of the gene expression data for the q-th prognostic gene; This represents an exponential function.

[0028] Example 2 The difference between this embodiment and Embodiment 1 is that: A method for constructing a prognostic model for clinical use in oncology includes the following steps: S1. Establish a prognostic model construction system framework, which includes a tumor data collection module, a tumor data screening module, a prognostic model construction module, a model training module, a model optimization module, a prognostic model evaluation module, an evaluation result analysis module, a storage module, an early warning module, a cause analysis module, a solution suggestion module, an early warning method setting module, an update time setting module, an update reminder module, and a prognostic model update module. S2. The tumor data collection module collects clinical data from the Department of Oncology, the tumor data filtering module filters the collected data, selects variables related to prognosis, and transmits the selected variables to the prognosis model construction module. The prognosis model construction module constructs a prognosis model based on the received data. S3. The prognostic model is trained through the model training module, optimized through the model optimization module, evaluated through the prognostic model evaluation module, and the performance of the evaluation is analyzed and judged through the evaluation result analysis module. S4. When the evaluated performance fails to meet the requirements, an early warning module is used to issue an early warning, a cause analysis module is used to analyze the reasons for the performance failure, and a solution suggestion module is used to provide solution suggestions. S5. Classify the performance of the model according to the analysis results of the evaluation result analysis module.

[0029] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for constructing a prognostic model for clinical use in oncology, characterized in that, Includes the following steps: S1. Establish a prognostic model construction system framework, which includes a tumor data collection module, a tumor data screening module, a prognostic model construction module, a model training module, a model optimization module, a prognostic model evaluation module, an evaluation result analysis module, a storage module, an early warning module, a cause analysis module, a solution suggestion module, an early warning method setting module, an update time setting module, an update reminder module, and a prognostic model update module. S2. The tumor data collection module collects clinical data from the Department of Oncology, the tumor data filtering module filters the collected data, selects variables related to prognosis, and transmits the selected variables to the prognosis model construction module. The prognosis model construction module constructs a prognosis model based on the received data. S3. The prognostic model is trained through the model training module, optimized through the model optimization module, evaluated through the prognostic model evaluation module, and the performance of the evaluation is analyzed and judged through the evaluation result analysis module. S4. When the evaluated performance fails to meet the requirements, an early warning module is used to issue an early warning, a cause analysis module is used to analyze the reasons for the performance failure, and a solution suggestion module is used to provide solution suggestions.

2. The method of claim 1, wherein the method is used for constructing a prognostic model for clinical use in oncology. The warning mode setting module sets the warning mode of the warning module. The warning mode setting module includes an audio-visual setting unit, which is connected to an information warning setting unit, which is connected to a voice warning setting unit, and the voice warning setting unit is connected to a warning frequency setting unit.

3. The method of claim 2, wherein the method is a method of constructing a prognostic model for clinical use in oncology, characterized in that, The early warning module includes an audible and visual early warning unit, which is connected to an information early warning unit, which is connected to a voice early warning unit. The audible and visual early warning unit is used for audible and visual early warning, the information early warning unit is used to send information to management personnel for early warning, and the voice early warning unit is used for voice early warning.

4. The method of claim 3, wherein the method is characterized by, The solution suggestion module includes a cause identification unit, which is connected to a solution matching unit, which is connected to a solution feedback unit, and the solution feedback unit is connected to a feedback recording unit.

5. The method of claim 4, wherein the method is used for constructing a prognostic model for clinical use in oncology. The cause identification unit is used to identify the reasons for non-compliance, the solution matching unit matches solutions based on the identified reasons, and the solution feedback unit feeds back the matched solutions to the management personnel.

6. The method of claim 5, wherein the method is used for constructing a prognostic model for clinical use in oncology. The update time setting module is used to set the update frequency of the prognostic model, the update reminder module is used to send a reminder to the administrator when an update is required, and the prognostic model update module updates the prognostic model in the prognostic model construction module after the set update time is reached.

7. The method of claim 6, wherein the method is used for constructing a prognostic model for clinical use in oncology. The tumor data collection module includes a clinical feature collection unit, which is connected to a biomarker collection unit, which is connected to a clinical outcome collection unit, and the clinical outcome collection unit is connected to a classification and transmission unit.

8. The method of claim 7, wherein the method is used for constructing a prognostic model for clinical use in oncology. The clinical feature collection unit is used to collect tumor clinical features, the biomarker collection unit is used to collect biomarkers, the clinical outcome collection unit is used to collect clinical outcome data, and the classification and transmission unit is used to classify and transmit the collected data.

9. The method for constructing a clinical prognostic model for oncology according to claim 8, characterized in that, In S3, the prognostic model is trained using a deep learning algorithm in the model training module, optimized using a model optimization module, evaluated using a prognostic model evaluation module, and analyzed and judged using an evaluation result analysis module. During model optimization, the feature space is expanded and more variables that affect disease development, such as lifestyle habits and family history, are added to improve the interpretability of the model.

10. A method for constructing a clinical prognostic model for oncology according to claim 9, characterized in that, The tumor data collection module is connected to the tumor data screening module, the tumor data screening module is connected to the prognostic model construction module, the update time setting module is connected to the update reminder module, the update reminder module is connected to the prognostic model update module, the prognostic model update module is connected to the prognostic model construction module, the prognostic model construction module is connected to the prognostic model evaluation module and the model training module, the model training module is connected to the model optimization module, the prognostic model evaluation module is connected to the evaluation result analysis module, the evaluation result analysis module is connected to the storage module and the early warning module, the early warning module is connected to the cause analysis module, the cause analysis module is connected to the solution suggestion module, and the early warning method setting module is connected to the early warning module.

Images