Prediction model-based rectal cancer postoperative recurrence risk prediction method

Abstract

The invention relates to the technical field of biomedicine, and discloses a rectal cancer postoperative recurrence risk prediction method based on a prediction model, and the method comprises the steps: obtaining a patient tumor in the in-vivo image data of a rectal cancer patient; extracting imaging features of the tumor of the patient, and identifying recurrence risk features of the rectal cancer patient from the imaging features; analyzing clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathological report; key recurrence influence factors of the rectal cancer patients are identified, recurrence events and follow-up visit time of the rectal cancer patients are identified, risk scoring is carried out on the rectal cancer patients, and a risk scoring result is obtained; based on the risk scoring result, the recurrence risk stratification of the rectal cancer patient is defined, postoperative recurrence risk analysis is conducted on the rectal cancer patient in combination with the recurrence risk stratification, the recurrence risk characteristics and the risk scoring result, and a risk analysis result is obtained. According to the invention, the comprehensiveness and accuracy of the rectal cancer recurrence risk analysis result can be improved.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and in particular to a prediction method and system for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model. Background Art

[0002] The risk of postoperative recurrence of rectal cancer refers to the risk of cancer re-appearing in the patient's body after surgical treatment, including the risk of local recurrence, regional recurrence or distant metastasis. In recent years, with the improvement of people's living standards and changes in lifestyle in my country, the incidence of rectal cancer has increased year by year. Although modern medicine has made significant progress in the treatment of rectal cancer, postoperative recurrence is still an important factor affecting patients' survival rate and quality of life. Therefore, it is particularly important to predict the risk of recurrence of rectal cancer.

[0003] Traditional analysis of the risk of recurrence after rectal cancer surgery mainly predicts the risk of recurrence by analyzing the patient's clinical pathological characteristics, such as tumor staging and lymph node metastasis. Although this method can evaluate the patient's recurrence risk type to a certain extent, a single pathological feature may not be sufficient to comprehensively predict the patient's recurrence risk type, resulting in a lack of comprehensiveness and accuracy in the prediction results. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a method and system for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model, which can improve the comprehensiveness and accuracy of the analysis results of the risk of recurrence of rectal cancer.

[0005] In a first aspect, the present invention provides a method for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model, comprising:

[0006] Acquiring in vivo image data of a rectal cancer patient, identifying a tumor region of the rectal cancer patient based on the in vivo image data, and performing image segmentation processing on the tumor region to obtain the patient's tumor;

[0007] Extracting imaging features of the patient's tumor based on the in vivo imaging data, analyzing the recurrence risk type of the rectal cancer patient, and identifying the recurrence risk features of the rectal cancer patient from the imaging features according to the recurrence risk type;

[0008] Collecting preoperative evaluation data and postoperative pathology reports of the rectal cancer patients, analyzing the clinical pathological characteristics of the rectal cancer patients based on the preoperative evaluation data and the postoperative pathology reports, identifying potential influencing factors of the recurrence risk type, and performing multifactor analysis on the potential influencing factors to obtain multifactor analysis results;

[0009] According to the multivariate analysis result, key recurrence influencing factors of the rectal cancer patient are identified, the key recurrence influencing factors are ranked to obtain graded influencing factors, and based on the graded influencing factors and the clinical pathological characteristics, the recurrence events and follow-up time of the rectal cancer patient are identified, and according to the recurrence events and the follow-up time, risk scoring is performed on the rectal cancer patient to obtain a risk scoring result;

[0010] Based on the risk scoring result, the recurrence risk stratification of the rectal cancer patients is defined, and the postoperative recurrence risk analysis of the rectal cancer patients is performed in combination with the recurrence risk stratification, the risk scoring result and the recurrence risk characteristics to obtain a risk analysis result.

[0011] In a possible implementation of the first aspect, analyzing the recurrence risk type of the rectal cancer patient includes:

[0012] Collecting the preoperative tumor staging report and tumor pathological characteristics of the rectal cancer patients;

[0013] Analyze the tumor margin status and lymph node involvement of the rectal cancer patient according to the preoperative tumor staging report and the tumor pathological characteristics;

[0014] Analyze the postoperative evaluation report of the rectal cancer patient based on the tumor resection margin status and the lymph node involvement status;

[0015] According to the postoperative evaluation report, identifying the tumor invasion range of the rectal cancer patient;

[0016] The recurrence risk type of the rectal cancer patient is analyzed based on the tumor invasion range, the tumor pathological characteristics and the tumor resection margin status.

[0017] In a second aspect, the present invention provides a prediction system for the risk of postoperative recurrence of rectal cancer based on a prediction model, characterized in that the system comprises:

[0018] A data acquisition module, used to obtain in vivo image data of a rectal cancer patient, identify a tumor region of the rectal cancer patient based on the in vivo image data, and perform image segmentation processing on the tumor region to obtain the patient's tumor;

[0019] An imaging feature recognition module, for extracting the imaging features of the patient's tumor based on the in vivo imaging data, analyzing the recurrence risk type of the rectal cancer patient, and identifying the recurrence risk features of the rectal cancer patient from the imaging features according to the recurrence risk type;

[0020] An influencing factor analysis module is used to collect the preoperative evaluation data and postoperative pathology report of the rectal cancer patient, analyze the clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathology report, identify the potential influencing factors of the recurrence risk type, and perform multifactor analysis on the potential influencing factors to obtain a multifactor analysis result;

[0021] a risk scoring module, for identifying key recurrence influencing factors of the rectal cancer patient according to the multi-factor analysis result, ranking the key recurrence influencing factors to obtain graded influencing factors, identifying the recurrence events and follow-up time of the rectal cancer patient based on the graded influencing factors and the clinical pathological characteristics, and performing risk scoring on the rectal cancer patient according to the recurrence events and the follow-up time to obtain a risk scoring result;

[0022] The prognostic analysis module is used to define the recurrence risk stratification of the rectal cancer patients based on the risk scoring results, and to perform postoperative recurrence risk analysis on the rectal cancer patients in combination with the recurrence risk stratification, the risk scoring results and the recurrence risk characteristics to obtain risk analysis results.

[0023] Compared with the prior art, the technical principle and beneficial effects of this solution are:

[0024] The embodiment of the present invention can identify the tumor characteristics of the rectal cancer patient by acquiring the in vivo imaging data of the rectal cancer patient, help evaluate the tumor information of the rectal cancer patient, and provide data support for the analysis of the patient's recurrence risk type; secondly, the embodiment of the present invention can obtain the biological behavior of the tumor by extracting the imaging characteristics of the patient's tumor based on the in vivo imaging data, so as to help doctors diagnose the tumor more accurately and evaluate the patient's recurrence risk type, and identify the recurrence risk characteristics of the rectal cancer patient from the imaging characteristics as the input parameters of the prediction model according to the recurrence risk type, thereby improving the accuracy of the prediction model; thirdly, the embodiment of the present invention can identify the high-risk population of rectal cancer patients by analyzing the clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathology report, and can accurately evaluate the patient's recurrence risk, and at the same time can determine the key factors causing postoperative recurrence of rectal cancer patients, so as to take preventive measures in advance. , to prevent the patient's condition from worsening; further, the embodiment of the present invention identifies the key recurrence influencing factors of the recurrence risk type according to the multi-factor analysis results, and ranks the key recurrence influencing factors to obtain graded influencing factors, which can clarify the pathological characteristics corresponding to the key recurrence influencing factors affecting the recurrence risk type, reduce the interference of confounding factors, and help improve the robustness of the prediction model; finally, combined with the recurrence risk stratification, the risk scoring results and the recurrence risk characteristics, the postoperative recurrence risk analysis of the rectal cancer patients is performed to obtain risk analysis results, which can provide more complex and advanced features for the prognostic model, help rectal cancer patients provide more accurate individualized treatment and risk assessment, and predict the recurrence risk type of rectal cancer patients by combining multiple predictive factors, which can improve the accuracy and comprehensiveness of the prediction model, help the medical and patient groups to find signs of recurrence in advance, treat in time, and improve the patient's survival rate. Therefore, the prediction method for postoperative recurrence risk of rectal cancer based on the prediction model proposed in the present invention can improve the comprehensiveness and accuracy of the results of the recurrence risk analysis of rectal cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0026] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0027] Figure 1 A schematic flow chart of a method for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model provided in one embodiment of the present invention;

[0028] Figure 2 A schematic diagram of modules of a system for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0029] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0030] The embodiment of the present invention provides a prediction method for the risk of postoperative recurrence of rectal cancer based on a prediction model, and the execution subject of the prediction method for the risk of postoperative recurrence of rectal cancer based on a prediction model includes but is not limited to at least one of the electronic devices that can be configured to execute the method provided by the embodiment of the present invention, such as a server, a terminal, etc. In other words, the prediction method for the risk of postoperative recurrence of rectal cancer based on a prediction model can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (Content De li very Network, CDN), and big data and artificial intelligence platforms.

[0031] See also Figure 1 FIG. 1 is a flow chart of a method for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model provided by an embodiment of the present invention. Figure 1 The prediction methods for the risk of postoperative recurrence of rectal cancer based on prediction models described in the literature include:

[0032] S1. Acquire in vivo image data of a rectal cancer patient, identify a tumor region of the rectal cancer patient based on the in vivo image data, and perform image segmentation processing on the tumor region to obtain the patient's tumor.

[0033] The embodiments of the present invention can help evaluate the tumor information of rectal cancer patients and provide data support for the analysis of the patient's recurrence risk type by acquiring the in vivo imaging data of the rectal cancer patients. The rectal cancer patients refer to individuals diagnosed with rectal cancer, and the in vivo imaging data refer to image information about the internal structure and function of the human body acquired by medical imaging equipment.

[0034] Optionally, the in vivo image data of the rectal cancer patient may be acquired by a medical imaging device, such as a magnetic resonance imaging device.

[0035] Furthermore, the embodiment of the present invention identifies the tumor region of the rectal cancer patient based on the in vivo imaging data, and performs image segmentation processing on the tumor region to obtain the patient's tumor, which can more accurately identify the tumor characteristics of the rectal cancer patient, thereby improving the accuracy of recurrence risk prediction. The tumor region refers to the region where the tumor of the rectal cancer patient identified in the in vivo imaging data is located, and the patient's tumor refers to the specific tumor of the rectal cancer patient identified and segmented in the in vivo imaging data.

[0036] Optionally, the tumor region identification of the rectal cancer patient based on the in vivo imaging data can be achieved using ITK-SNAP imaging software, and the image segmentation processing of the tumor region can be obtained through the MITK toolkit.

[0037] S2. Based on the in vivo imaging data, extract the imaging features of the patient's tumor, analyze the recurrence risk type of the rectal cancer patient, and identify the recurrence risk features of the rectal cancer patient from the imaging features according to the recurrence risk type.

[0038] The embodiment of the present invention can obtain the biological behavior of the tumor by extracting the imaging features of the patient's tumor based on the in vivo imaging data, thereby helping doctors to diagnose the tumor more accurately and assess the patient's recurrence risk type. The imaging features refer to imaging features extracted from medical images that can reflect the intrinsic biological characteristics and microenvironment of the tumor, such as DWI (diffusion weighted imaging) features and T2WI (T2 weighted imaging) features.

[0039] Optionally, the imaging feature extraction of the patient's tumor based on the in vivo imaging data can be achieved using magnetic resonance images.

[0040] Furthermore, the embodiments of the present invention can help identify which clinical and imaging characteristics are key factors in predicting recurrence risk by analyzing the recurrence risk type of the rectal cancer patient. The recurrence risk type refers to the risk type of tumor re-growth at the primary site or other parts of the body after the rectal cancer patient receives surgical treatment, such as the risk of tumor spread.

[0041] As an embodiment of the present invention, the analysis of the recurrence risk type of the rectal cancer patient includes: collecting the preoperative tumor staging report and tumor pathological characteristics of the rectal cancer patient; analyzing the tumor margin status and lymph node involvement of the rectal cancer patient according to the preoperative tumor staging report and the tumor pathological characteristics; analyzing the postoperative evaluation report of the rectal cancer patient based on the tumor margin status and the lymph node involvement; identifying the tumor invasion range of the rectal cancer patient according to the postoperative evaluation report; and analyzing the recurrence risk type of the rectal cancer patient in combination with the tumor invasion range, the tumor pathological characteristics and the tumor margin status.

[0042] Among them, the preoperative tumor staging report refers to the tumor staging results obtained through clinical examination, imaging examination and pathological examination (such as biopsy) before surgery, the tumor pathological characteristics refer to the detailed characteristics of the tumor obtained through pathological examination, the tumor resection margin status refers to the edge status between the resected tumor tissue and the surrounding normal tissue during tumor surgery, the lymph node burden status refers to whether cancer cells have spread to the lymph nodes in cancer patients, the postoperative evaluation report refers to a detailed report obtained by pathological examination of the resected specimens after surgery, and the tumor invasion range refers to the degree of tumor spread in the tissue and the depth of invasion.

[0043] Optionally, the preoperative tumor staging report of the rectal cancer patient can be obtained through an imaging examination instrument, such as a CT scanner, and the collection of the tumor pathological characteristics of the rectal cancer patient can be achieved by biopsy. The preoperative tumor staging report and the tumor pathological characteristics, the analysis of the tumor resection margin status and lymph node burden of the rectal cancer patient can be obtained through pathological examination of the surgical resection specimen, and based on the postoperative evaluation report, the identification of the tumor invasion range of the rectal cancer patient can be achieved by using PET-CT.

[0044] The embodiment of the present invention can more accurately predict the recurrence risk type of rectal cancer patients and improve the accuracy of the prediction model by identifying the recurrence risk characteristics of the rectal cancer patients from the imaging characteristics according to the recurrence risk type. The recurrence risk characteristics refer to specific imaging characteristics related to the recurrence risk of rectal cancer identified in in vivo imaging data, such as pixel value distribution characteristics in CT or MRI images.

[0045] As an embodiment of the present invention, the method of identifying the recurrence risk characteristics of the rectal cancer patient from the imaging characteristics according to the recurrence risk type includes: scheduling the imaging characteristics of the rectal cancer patient; calculating the correlation coefficient between the imaging characteristics and the recurrence risk type; identifying the associated characteristics between the imaging characteristics and the recurrence risk type according to the correlation coefficient; analyzing the feature importance of the associated characteristics to the recurrence risk type; and identifying the recurrence risk characteristics of the rectal cancer patient from the imaging characteristics based on the feature importance.

[0046] Among them, the correlation coefficient refers to a statistic used to measure the strength and direction of the linear relationship between two variables, the association feature refers to an imaging feature that is significantly correlated with the recurrence risk type, and the feature importance refers to the contribution of a certain feature to the recurrence risk type analysis.

[0047] Optionally, the feature importance analysis of the associated features to the recurrence risk type may be implemented using the G ini index variation in a random forest algorithm.

[0048] In an optional embodiment of the present invention, the correlation coefficient between the imaging feature and the recurrence risk type is calculated using the following formula:

[0049]

[0050] Among them, q x,y represents the correlation coefficient between imaging features and recurrence risk type, x i represents the observed value of the i-th imaging feature, y i represents the observed value of the i-th recurrence risk type, represents the sample mean of the imaging feature, represents the sample mean of the recurrence risk type, and m represents the number of samples corresponding to the imaging features and recurrence risk types.

[0051] S3. Collect the preoperative evaluation data and postoperative pathology report of the rectal cancer patient, analyze the clinical pathological characteristics of the rectal cancer patient based on the preoperative evaluation data and the postoperative pathology report, identify the potential influencing factors of the recurrence risk type, perform multifactor analysis on the potential influencing factors, and obtain the multifactor analysis results.

[0052] The embodiment of the present invention can provide comprehensive clinical and pathological information by collecting the preoperative evaluation data and postoperative pathology report of the rectal cancer patient, helping doctors to more accurately evaluate the patient's risk of recurrence and formulate personalized treatment plans. The preoperative evaluation data refers to a series of patient information obtained through various diagnostic methods and examinations before the rectal cancer patient undergoes surgery, and the postoperative pathology report refers to a detailed report issued after pathological examination of the resected tumor tissue after the rectal cancer patient undergoes surgery.

[0053] Optionally, the collection of the preoperative evaluation data and postoperative pathology reports of the rectal cancer patients can be achieved using an electronic medical record system.

[0054] Furthermore, the embodiment of the present invention can identify high-risk groups of rectal cancer patients and accurately assess the risk of recurrence of patients by analyzing the clinical pathological characteristics of the rectal cancer patients based on the preoperative evaluation data and the postoperative pathology report. The clinical pathological characteristics refer to various information obtained through clinical examination and pathological examination that describes the disease status and lesion characteristics of rectal cancer patients, such as the degree of similarity between tumor cells and normal cells.

[0055] As an embodiment of the present invention, the method of analyzing the clinical pathological characteristics of the rectal cancer patient based on the preoperative evaluation data and the postoperative pathology report includes: identifying the clinical symptoms of the rectal cancer patient based on the preoperative evaluation data; extracting the tumor characteristics of the rectal cancer patient based on the clinical symptoms; scheduling the imaging data of the rectal cancer patient based on the postoperative pathology report and the tumor characteristics, and identifying the tumor cells of the rectal cancer patient from the imaging data; analyzing the invasive ability of the tumor cells on paracancerous tissues, and analyzing the genetic heterogeneity of the tumor cells; identifying the lesion characteristics of the tumor cells in combination with the invasive ability and the genetic heterogeneity; and analyzing the clinical pathological characteristics of the rectal cancer patient based on the clinical symptoms and the lesion characteristics.

[0056] Among them, the clinical symptoms refer to the various symptoms and signs exhibited by patients when seeking medical treatment, such as blood in the stool; the tumor characteristics refer to the detailed characteristics of the tumor obtained through imaging and pathological examinations, including size, location, morphology, density, texture, etc.; the tumor cells refer to the malignant cells that constitute the tumor; the paracancerous tissue refers to the normal tissue or slightly abnormal tissue around the tumor; the invasive ability refers to the ability of tumor cells to invade the surrounding normal tissue; the genetic heterogeneity refers to the differences between tumor cells at the genetic level, such as gene mutations; and the lesion characteristics refer to the detailed characteristics of tumor cells obtained through pathological examination and molecular biological analysis.

[0057] Optionally, based on the preoperative evaluation data, the clinical symptom identification of the rectal cancer patient can be achieved by laboratory examination, the analysis of the tumor cell invasion ability to adjacent cancer tissues can be obtained by HE staining, and the genetic heterogeneity analysis of the tumor cells can be achieved by high-throughput sequencing technology, such as whole genome sequencing technology.

[0058] The embodiments of the present invention can identify the key factors causing postoperative recurrence of rectal cancer patients by identifying the potential influencing factors of the recurrence risk type, and take preventive measures in advance to prevent the patient's condition from worsening. The potential influencing factors refer to various factors that may affect the recurrence risk type of rectal cancer patients, such as tumor stage and tumor deterioration degree.

[0059] As an embodiment of the present invention, the identifying of potential influencing factors of the recurrence risk type includes: scheduling patient data of the recurrence risk type, and identifying the recurrence factors of the recurrence risk type based on the patient data; analyzing the recurrence population of the recurrence risk type based on the recurrence factors; extracting individual difference characteristics of the recurrence population, and identifying the in vivo markers of the recurrence factors in the recurrence population based on the individual difference characteristics; analyzing the in vivo state of the recurrence factors in the recurrence population; and identifying the potential influencing factors of the recurrence risk type based on the in vivo markers, the in vivo state and the individual difference characteristics.

[0060] Among them, the patient data refers to all clinical, pathological, imaging and molecular biological information related to the patient, the recurrence factor refers to the lesion factor closely related to the recurrence risk type, the recurrence group refers to the patient group with postoperative recurrence, the individual difference characteristics refer to the characteristic differences between different patients in genetics, pathology, clinical and other aspects, the in vivo marker refers to the biomarker that can reflect the risk of tumor recurrence, and the in vivo state refers to the physiological and pathological state in the patient's body related to the recurrence factor, such as tumor staging.

[0061] Optionally, the recurrence population analysis of the recurrence risk type based on the recurrence factor can be obtained by Kaplan-Meier survival analysis, the extraction of individual difference characteristics of the recurrence population can be achieved by genetic testing, the in vivo marker identification of the recurrence factor in the recurrence population can be obtained by PCR method, and the in vivo status analysis of the recurrence factor in the recurrence population can be achieved by pathological examination.

[0062] Furthermore, the embodiment of the present invention obtains a multifactor analysis result by performing a multifactor analysis on the potential influencing factors, thereby identifying the key recurrence influencing factors that affect the postoperative recurrence of rectal cancer patients and improving the accuracy of the prediction model. The multifactor analysis result refers to a comprehensive analysis result obtained by considering the impact of multiple potential influencing factors on the recurrence risk type of rectal cancer patients.

[0063] Optionally, the multi-factor analysis of the potential influencing factors can be implemented using a multivariate regression analysis method.

[0064] S4. According to the multivariate analysis results, key recurrence influencing factors of the rectal cancer patients are identified, and the key recurrence influencing factors are ranked to obtain graded influencing factors. Based on the graded influencing factors and the clinical pathological characteristics, the recurrence events and follow-up time of the rectal cancer patients are identified. According to the recurrence events and the follow-up time, risk scores are performed on the rectal cancer patients to obtain risk score results.

[0065] The embodiment of the present invention can reduce the interference of confounding factors and improve the accuracy of recurrence risk type analysis by identifying the key recurrence influencing factors of the rectal cancer patient according to the multivariate analysis results. The potential influencing factors refer to influencing variables that have a significant impact on the recurrence of rectal cancer, such as resection margin status, lymph node metastasis and gene mutation.

[0066] Optionally, according to the multivariate analysis results, the identification of key recurrence influencing factors of the rectal cancer patients can be achieved by using statistical significance.

[0067] Furthermore, the embodiment of the present invention obtains graded influencing factors by ranking the key recurrence influencing factors, clarifies the pathological characteristics corresponding to the key recurrence influencing factors that affect the recurrence risk, and helps improve the robustness of the prediction model. The graded influencing factors refer to influencing factors of different levels obtained after ranking the key recurrence influencing factors by importance.

[0068] As an embodiment of the present invention, the key recurrence influencing factors are ranked to obtain graded influencing factors, including: identifying the affected objects corresponding to the key recurrence influencing factors; calculating the effect ratio of the key recurrence influencing factors on the affected objects; performing risk assessment on the key recurrence influencing factors to obtain risk assessment conclusions; and ranking the key recurrence influencing factors according to the effect ratio and the risk assessment conclusion to obtain graded influencing factors.

[0069] Among them, the affected object refers to the specific object directly affected by the key recurrence influencing factors, the effect ratio refers to the degree of influence of the key recurrence influencing factors on the affected object, and the risk assessment conclusion refers to the conclusion drawn after evaluating the potential risk of each key recurrence influencing factor.

[0070] Optionally, the risk assessment of the key recurrence influencing factors may be implemented using a fault tree analysis.

[0071] In an optional embodiment of the present invention, the following formula is used to calculate the effect ratio of the key recurrence influencing factor on the influencing object:

[0072]

[0073] Among them, W h Indicates the effect ratio of the hth key recurrence influencing factor on the affected object, 1-E h represents the non-entropy value of the h-th key recurrence influencing factor, 1-E k represents the non-entropy value of all key recurrence influencing factors, n represents the total number of key recurrence influencing factors, and k represents the sum index of key recurrence influencing factors.

[0074] The embodiment of the present invention identifies the recurrence events and follow-up time of the rectal cancer patient based on the grading influencing factors and the clinical pathological characteristics, and can increase the generalization ability of the prognosis model through the characteristics of the recurrence events to help doctors better treat patients. The recurrence event refers to the recurrence of a disease that has been controlled or eliminated after the initial treatment, and the follow-up time refers to the time interval from the completion of the patient's initial treatment (such as surgery, radiotherapy, chemotherapy, etc.) to the next scheduled medical evaluation or examination.

[0075] Optionally, the identification of the follow-up time of the rectal cancer patient based on the grading influencing factors and the clinical pathological characteristics can be determined by the recurrence risk analysis results.

[0076] As an embodiment of the present invention, the recurrence events of the rectal cancer patient are defined based on the grading influencing factors and the clinical pathological characteristics, including: identifying the recurrence type of the rectal cancer patient based on the grading influencing factors; identifying the recurrence location of the rectal cancer patient based on the recurrence type; analyzing the recurrence risk level of the rectal cancer patient based on the recurrence type and the recurrence location; and defining the recurrence events of the rectal cancer patient in combination with the recurrence type, the recurrence location and the recurrence risk level.

[0077] Among them, the recurrence type refers to the specific circumstances and characteristics of the recurrence of the disease after treatment, the recurrence location refers to the place where tumor cells grow again after treatment, and the recurrence risk level refers to the level of impact caused by the re-growth of tumor cells after treatment.

[0078] Optionally, the identification of the recurrence type of the rectal cancer patient based on the multifactor analysis results can be achieved using magnetic resonance technology, and the analysis of the recurrence risk level of the rectal cancer patient based on the recurrence type and the recurrence location can be determined by a risk assessment model.

[0079] The embodiment of the present invention performs risk scoring on the rectal cancer patients according to the recurrence events and the follow-up time to obtain a risk scoring result, thereby identifying high-risk patients and taking more active treatment measures to improve the patient's prognosis. The risk scoring result refers to the score obtained after quantitatively evaluating the risk of rectal cancer recurrence of an individual or a group.

[0080] In an optional embodiment of the present invention, according to the recurrence event and the follow-up time, the following formula is used to perform a risk score on the rectal cancer patient to obtain a risk score result:

[0081]

[0082] Among them, Loss represents the risk score result of rectal cancer patients, M represents the total number of rectal cancer patients, and W g Represents the weight coefficient of the recurrence event, g j Indicator variable indicating whether the jth rectal cancer patient has a recurrence event, v e represents the weight coefficient of follow-up time, e j The jth colorectal cancer patient’s follow-up time, k represents the same total number of recurrence events and follow-up time, l represents the category index of the scoring model corresponding to the risk score result, u lj It represents the score of the jth rectal cancer patient of category l given by the scoring model corresponding to the risk scoring result.

[0083] S5. Based on the risk scoring result, define the recurrence risk stratification of the rectal cancer patients, and perform postoperative recurrence risk analysis on the rectal cancer patients in combination with the recurrence risk stratification, the risk scoring result and the recurrence risk characteristics to obtain a risk analysis result.

[0084] The embodiment of the present invention defines the recurrence risk stratification of the rectal cancer patients based on the risk scoring results, thereby avoiding overtreatment of low-risk patients and ensuring that high-risk patients can receive appropriate treatment, thereby reducing unnecessary side effects and complications while ensuring the treatment effect. The recurrence risk stratification refers to dividing rectal cancer patients into different risk groups according to the patients' pathological characteristics, clinical characteristics and risk scoring results.

[0085] As an embodiment of the present invention, the recurrence risk stratification of the rectal cancer patient is defined based on the risk scoring result, including: identifying the risk prediction factor of the rectal cancer patient based on the risk scoring result; performing multivariate analysis on the risk prediction factor to obtain a multivariate analysis result; determining the independent risk prediction factor of the rectal cancer patient based on the multivariate analysis result; setting the weight ratio of the independent risk prediction factor; generating a recurrence risk scoring formula for the rectal cancer patient based on the weight ratio; identifying the recurrence risk score distribution of the rectal cancer patient based on the recurrence risk scoring formula; and defining the recurrence risk stratification of the rectal cancer patient based on the recurrence risk score distribution.

[0086] Among them, the risk prediction factor refers to the clinical and pathological characteristics associated with the recurrence risk type of rectal cancer patients, the multivariate analysis result refers to the result obtained by simultaneously analyzing multiple risk prediction factors through statistical methods, the independent risk prediction factor refers to the factor that can significantly predict the recurrence risk in the multivariate analysis even after controlling the influence of other variables, the weight ratio refers to the numerical value of the relative importance or contribution assigned to each independent risk prediction factor, the recurrence risk scoring formula refers to the mathematical formula for calculating the total recurrence risk score for each patient based on the weight ratio of the independent risk prediction factor, and the recurrence risk score distribution refers to the distribution of the total recurrence risk score calculated according to the recurrence risk scoring formula in a specific patient population.

[0087] Optionally, based on the risk scoring results, the identification of risk prediction factors for rectal cancer patients can be achieved using a machine learning model, the multivariate analysis of the risk prediction factors can be obtained through Cox regression analysis, and the weight ratio of the independent risk prediction factors can be determined using the regression coefficients in the multivariate analysis.

[0088] In an optional embodiment of the present invention, the independent risk prediction factor of the rectal cancer patient is determined according to the multivariate analysis result using the following formula:

[0089]

[0090] Among them, N1, N2, ..., N zrepresents the independent risk predictor of colorectal cancer patients, H(t) represents the risk function of colorectal cancer patients at time t, H0(t) represents the baseline risk function of colorectal cancer patients at time t, β1, β2, …, β z represents the regression coefficient corresponding to the independent risk predictor.

[0091] Furthermore, the embodiment of the present invention combines the recurrence risk stratification, the risk scoring results and the recurrence risk characteristics to perform postoperative recurrence risk analysis on the rectal cancer patients to obtain risk analysis results, which can provide more complex and advanced features for the prognostic model, help rectal cancer patients provide more accurate individualized treatment and risk assessment, and predict the recurrence risk type of rectal cancer patients by combining multiple predictive factors, which can improve the accuracy and comprehensiveness of the prediction model, help the medical and patient groups to detect signs of recurrence in advance, provide timely treatment, and improve patient survival rate. The risk analysis result refers to the conclusion of evaluating and predicting the recurrence risk of rectal cancer patients by combining the recurrence risk stratification, recurrence risk characteristics and risk scoring results, such as recurrence time and recurrence grade.

[0092] As an embodiment of the present invention, the postoperative recurrence risk analysis is performed on the rectal cancer patient in combination with the recurrence risk stratification, the risk scoring result and the recurrence risk characteristic to obtain a risk analysis result, including: scheduling the clinical pathological data and genomic expression data of the rectal cancer patient based on the recurrence risk stratification; analyzing the risk recurrence factors of the rectal cancer patient based on the clinical pathological data; analyzing the survival period of the rectal cancer patient based on the risk recurrence factors and the risk scoring result; identifying the lesion characteristics of the rectal cancer patient based on the genomic expression data; analyzing the cancer cell metastasis risk and gene mutation status of the rectal cancer patient based on the lesion characteristics; performing a postoperative recurrence risk analysis on the rectal cancer patient in combination with the survival period, the cancer cell metastasis risk and the gene mutation status to obtain a risk analysis result.

[0093] Among them, the clinical pathological data refers to the comprehensive data related to the patient's condition collected during the diagnosis and treatment process, the genomic expression data refers to the data on the activity of specific genes collected from tumor samples of rectal cancer patients, the risk recurrence factors refer to factors that can increase the risk of recurrence in rectal cancer patients, such as lymphovascular invasion and cancer nodules, etc., the survival period refers to the time from the patient's treatment to the first recurrence of the disease or the patient's death due to any reason, the lesion characteristics refer to the detailed information of the tumor obtained in the imaging examination, such as the size of the tumor, the risk of cancer cell metastasis refers to the possible risk of cancer cells spreading from the primary tumor site to other parts of the body, and the gene mutation status refers to the specific gene mutations in the patient's tumor cells.

[0094] Optionally, the analysis of risk recurrence factors of rectal cancer patients based on the clinical pathological data can be achieved by multivariate analysis, the analysis of survival of rectal cancer patients based on the risk recurrence factors and the risk score results can be determined by Kaplan-Meier survival analysis, and the identification of gene mutation status of rectal cancer patients based on the lesion characteristics can be achieved by high-resolution melting curve analysis technology.

[0095] It can be seen that the embodiment of the present invention can identify the tumor characteristics of the rectal cancer patient by acquiring the in vivo imaging data of the rectal cancer patient, help evaluate the tumor information of the rectal cancer patient, and provide data support for the analysis of the patient's recurrence risk type; secondly, the embodiment of the present invention can obtain the biological behavior of the tumor by extracting the imaging characteristics of the patient's tumor based on the in vivo imaging data, so as to help doctors diagnose the tumor more accurately and evaluate the patient's recurrence risk type, and identify the recurrence risk characteristics of the rectal cancer patient from the imaging characteristics as the input parameters of the prediction model according to the recurrence risk type, thereby improving the accuracy of the prediction model; thirdly, the embodiment of the present invention can identify the high-risk population of rectal cancer patients by analyzing the clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathology report, and can accurately evaluate the patient's recurrence risk, and at the same time can determine the key factors causing postoperative recurrence of rectal cancer patients, so as to make preventive measures in advance. measures to prevent the patient's condition from worsening; further, the embodiment of the present invention identifies the key recurrence influencing factors of the recurrence risk type according to the multi-factor analysis results, and ranks the key recurrence influencing factors to obtain graded influencing factors, which can clarify the pathological characteristics corresponding to the key recurrence influencing factors affecting the recurrence risk type, reduce the interference of confounding factors, and help improve the robustness of the prediction model; finally, combined with the recurrence risk stratification, the risk scoring results and the recurrence risk characteristics, the postoperative recurrence risk analysis of the rectal cancer patients is performed to obtain risk analysis results, which can provide more complex and advanced features for the prognostic model, help rectal cancer patients provide more accurate individualized treatment and risk assessment, and predict the recurrence risk type of rectal cancer patients by combining multiple predictive factors, which can improve the accuracy and comprehensiveness of the prediction model, help the medical and patient groups to find signs of recurrence in advance, treat in time, and improve the patient's survival rate. Therefore, the prediction method for postoperative recurrence risk of rectal cancer based on the prediction model proposed in the present invention can improve the comprehensiveness and accuracy of the analysis results of the recurrence risk of rectal cancer. Therefore, the prediction method for the risk of postoperative recurrence of rectal cancer based on the prediction model proposed in the present invention can improve the comprehensiveness and accuracy of the analysis results of the risk of recurrence of rectal cancer.

[0096] like Figure 2Shown is a system functional module diagram of the present invention for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model.

[0097] The prediction system 200 for the risk of postoperative recurrence of rectal cancer based on the prediction model of the present invention can be installed in an electronic device. According to the functions implemented, the prediction system for the risk of postoperative recurrence of rectal cancer based on the prediction model can include a data acquisition module 201, an image feature recognition module 202, an influencing factor analysis module 203, a risk scoring module 204, and a prognosis analysis module 205. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0098] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0099] The data acquisition module 201 is used to obtain in vivo image data of a rectal cancer patient, identify the tumor region of the rectal cancer patient based on the in vivo image data, and perform image segmentation processing on the tumor region to obtain the patient's tumor;

[0100] An image feature recognition module 202 is used to extract the imaging features of the patient's tumor based on the in vivo image data, analyze the recurrence risk type of the rectal cancer patient, and identify the recurrence risk features of the rectal cancer patient from the imaging features according to the recurrence risk type;

[0101] The influencing factor analysis module 203 is used to collect the preoperative evaluation data and postoperative pathology report of the rectal cancer patient, analyze the clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathology report, identify the potential influencing factors of the recurrence risk type, and perform multifactor analysis on the potential influencing factors to obtain a multifactor analysis result;

[0102] The risk scoring module 204 is used to identify the key recurrence influencing factors of the rectal cancer patient according to the multi-factor analysis result, rank the key recurrence influencing factors to obtain graded influencing factors, identify the recurrence events and follow-up time of the rectal cancer patient based on the graded influencing factors and the clinical pathological characteristics, and perform risk scoring on the rectal cancer patient according to the recurrence events and the follow-up time to obtain a risk scoring result;

[0103] The prognostic analysis module 205 is used to define the recurrence risk stratification of the rectal cancer patients based on the risk scoring results, and to perform postoperative recurrence risk analysis on the rectal cancer patients in combination with the recurrence risk stratification, the risk scoring results and the recurrence risk characteristics to obtain risk analysis results.

[0104] In detail, the modules in the prediction system 200 for rectal cancer postoperative recurrence risk based on the prediction model in the embodiment of the present invention are used in the same manner as described above. Figure 1 The same technical means are used as the method for predicting the risk of postoperative recurrence of rectal cancer based on the prediction model described in , and can produce the same technical effects, so they will not be repeated here.

[0105] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0106] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0107] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0108] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0109] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for predicting the risk of postoperative recurrence of rectal cancer based on a prediction model, characterized in that: The method comprises: Acquiring in vivo image data of a rectal cancer patient, identifying a tumor region of the rectal cancer patient based on the in vivo image data, and performing image segmentation processing on the tumor region to obtain the patient's tumor; Extracting imaging features of the patient's tumor based on the in vivo imaging data, analyzing the recurrence risk type of the rectal cancer patient, and identifying the recurrence risk features of the rectal cancer patient from the imaging features according to the recurrence risk type; Collecting preoperative evaluation data and postoperative pathology reports of the rectal cancer patients, analyzing the clinical pathological characteristics of the rectal cancer patients based on the preoperative evaluation data and the postoperative pathology reports, identifying potential influencing factors of the recurrence risk type, and performing multifactor analysis on the potential influencing factors to obtain multifactor analysis results; According to the multivariate analysis result, key recurrence influencing factors of the rectal cancer patient are identified, the key recurrence influencing factors are ranked to obtain graded influencing factors, and based on the graded influencing factors and the clinical pathological characteristics, the recurrence events and follow-up time of the rectal cancer patient are identified, and according to the recurrence events and the follow-up time, risk scoring is performed on the rectal cancer patient to obtain a risk scoring result; Based on the risk scoring result, the recurrence risk stratification of the rectal cancer patients is defined, and the postoperative recurrence risk analysis of the rectal cancer patients is performed in combination with the recurrence risk stratification, the risk scoring result and the recurrence risk characteristics to obtain a risk analysis result.

2. The method according to claim 1, characterized in that The analysis of the recurrence risk type of the rectal cancer patient includes: Collecting the preoperative tumor staging report and tumor pathological characteristics of the rectal cancer patients; Analyze the tumor margin status and lymph node involvement of the rectal cancer patient according to the preoperative tumor staging report and the tumor pathological characteristics; Analyze the postoperative evaluation report of the rectal cancer patient based on the tumor resection margin status and the lymph node involvement status; According to the postoperative evaluation report, identifying the tumor invasion range of the rectal cancer patient; The recurrence risk type of the rectal cancer patient is analyzed based on the tumor invasion range, the tumor pathological characteristics and the tumor resection margin status.

3. The method according to claim 1, characterized in that The step of identifying the recurrence risk feature of the rectal cancer patient from the imaging features according to the recurrence risk type includes: Dispatching imaging features of the rectal cancer patient, and identifying DWI features and T2WI features among the imaging features; According to the correlation coefficient, identifying the correlation features between the DWI feature and the T2WI feature and the recurrence risk type; analyzing the feature importance of the associated features to the recurrence risk type; Based on the feature importance, the recurrence risk feature of the rectal cancer patient is identified from the imaging features.

4. The method according to claim 1, characterized in that The step of analyzing the clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathology report includes: identifying clinical symptoms of the rectal cancer patient based on the preoperative evaluation data; Based on the clinical symptoms, extracting tumor characteristics of the rectal cancer patient; dispatching imaging data of the rectal cancer patient according to the postoperative pathology report and the tumor characteristics, and identifying tumor cells of the rectal cancer patient from the imaging data; Analyzing the invasion ability of the tumor cells to adjacent paracancerous tissues, and analyzing the genetic heterogeneity of the tumor cells; Combining the invasion ability and the gene heterogeneity, identifying the pathological characteristics of the tumor cells; Based on the clinical symptoms and the lesion characteristics, the clinical pathological characteristics of the rectal cancer patients are analyzed.

5. The method according to claim 1, characterized in that The potential influencing factors for identifying the recurrence risk type include: scheduling patient data of the recurrence risk type, and identifying a recurrence factor of the recurrence risk type based on the patient data; Based on the recurrence factors, analyzing the recurrence population of the recurrence risk type; Extracting individual difference characteristics of the relapse group, and identifying in vivo markers of the relapse factor in the relapse group according to the individual difference characteristics; Analyzing the in vivo status of the relapse factor in the relapse population; Potential influencing factors of the recurrence risk type are identified based on the in vivo markers, the in vivo status and the individual difference characteristics.

6. The method according to claim 1, characterized in that The key recurrence influencing factors are ranked to obtain graded influencing factors, including: Identify the affected objects corresponding to the key recurrence influencing factors; Calculating the effect ratio of the key relapse influencing factors on the affected object; Conducting risk assessment on the key recurrence influencing factors to obtain risk assessment conclusions; According to the effect ratio and the risk assessment conclusion, the key recurrence influencing factors are ranked to obtain graded influencing factors.

7. The method according to claim 1, characterized in that Defining the recurrence event of the rectal cancer patient based on the grading influencing factors and the clinical pathological characteristics and the clinical pathological characteristics includes: identifying the recurrence type of the rectal cancer patient according to the grading influencing factors; Based on the recurrence type, identifying the recurrence location of the rectal cancer patient; Analyzing the recurrence risk level of the rectal cancer patient according to the recurrence type and the recurrence location; The recurrence event of the rectal cancer patient is defined in combination with the recurrence type, the recurrence location and the recurrence risk level.

8. The method according to claim 1, characterized in that Defining the recurrence risk stratification of the rectal cancer patient based on the risk score result includes: Based on the risk score result, identifying risk predictors for the rectal cancer patient; Performing multivariate analysis on the risk prediction factors to obtain multivariate analysis results; Determining independent risk predictors for the rectal cancer patient based on the multivariate analysis results; Setting the weight ratio of the independent risk prediction factor; Based on the weight ratio, generating a recurrence risk scoring formula for the rectal cancer patient; According to the recurrence risk score formula, identifying the recurrence risk score distribution of the rectal cancer patient; Based on the recurrence risk score distribution, the recurrence risk stratification of the rectal cancer patients is defined.

9. The method according to claim 1, characterized in that: The step of combining the recurrence risk stratification, the risk score result and the recurrence risk characteristics to perform postoperative recurrence risk analysis on the rectal cancer patient to obtain a risk analysis result includes: Based on the recurrence risk stratification, scheduling the clinical pathological data and genomic expression data of the rectal cancer patient; Analyzing the risk factors for recurrence of the rectal cancer patient based on the clinical pathological data; Analyzing the survival of the rectal cancer patient based on the risk recurrence factors and the risk score results; identifying the lesion characteristics of the rectal cancer patient according to the genome expression data; Based on the lesion characteristics, analyzing the cancer cell metastasis risk and gene mutation status of the rectal cancer patient; Based on the survival period, the risk of cancer cell metastasis and the gene mutation status, a postoperative recurrence risk analysis is performed on the rectal cancer patients to obtain a risk analysis result.

10. A prediction system for the risk of postoperative recurrence of rectal cancer based on a prediction model, the system implementing the method according to claim 1, characterized in that: The system comprises: A data acquisition module, used to obtain in vivo image data of a rectal cancer patient, identify a tumor region of the rectal cancer patient based on the in vivo image data, and perform image segmentation processing on the tumor region to obtain the patient's tumor; An imaging feature recognition module, for extracting the imaging features of the patient's tumor based on the in vivo imaging data, analyzing the recurrence risk type of the rectal cancer patient, and identifying the recurrence risk features of the rectal cancer patient from the imaging features according to the recurrence risk type; An influencing factor analysis module is used to collect the preoperative evaluation data and postoperative pathology report of the rectal cancer patient, analyze the clinical pathological characteristics of the rectal cancer patient according to the preoperative evaluation data and the postoperative pathology report, identify the potential influencing factors of the recurrence risk type, and perform multifactor analysis on the potential influencing factors to obtain a multifactor analysis result; a risk scoring module, for identifying key recurrence influencing factors of the rectal cancer patient according to the multi-factor analysis result, ranking the key recurrence influencing factors to obtain graded influencing factors, identifying the recurrence events and follow-up time of the rectal cancer patient based on the graded influencing factors and the clinical pathological characteristics, and performing risk scoring on the rectal cancer patient according to the recurrence events and the follow-up time to obtain a risk scoring result; The prognostic analysis module is used to define the recurrence risk stratification of the rectal cancer patients based on the risk scoring results, and to perform postoperative recurrence risk analysis on the rectal cancer patients in combination with the recurrence risk stratification, the risk scoring results and the recurrence risk characteristics to obtain risk analysis results.

Images