CN113555123A - Method for establishing prediction model of survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy - Google Patents
Method for establishing prediction model of survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy Download PDFInfo
- Publication number
- CN113555123A CN113555123A CN202110994379.2A CN202110994379A CN113555123A CN 113555123 A CN113555123 A CN 113555123A CN 202110994379 A CN202110994379 A CN 202110994379A CN 113555123 A CN113555123 A CN 113555123A
- Authority
- CN
- China
- Prior art keywords
- model
- survival
- gallbladder cancer
- analysis
- group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 208000022072 Gallbladder Neoplasms Diseases 0.000 title claims abstract description 57
- 201000010175 gallbladder cancer Diseases 0.000 title claims abstract description 57
- 230000004083 survival effect Effects 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 37
- 230000008901 benefit Effects 0.000 title claims abstract description 16
- 238000001959 radiotherapy Methods 0.000 title claims abstract description 15
- 238000002512 chemotherapy Methods 0.000 title claims abstract description 14
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000012795 verification Methods 0.000 claims abstract description 13
- 238000011282 treatment Methods 0.000 claims abstract description 11
- 238000003745 diagnosis Methods 0.000 claims abstract description 7
- 230000001575 pathological effect Effects 0.000 claims abstract description 6
- 238000009098 adjuvant therapy Methods 0.000 claims description 21
- 230000004069 differentiation Effects 0.000 claims description 14
- 210000001165 lymph node Anatomy 0.000 claims description 14
- 206010028980 Neoplasm Diseases 0.000 claims description 8
- 238000011226 adjuvant chemotherapy Methods 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000002980 postoperative effect Effects 0.000 claims description 4
- 238000011160 research Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000013211 curve analysis Methods 0.000 claims description 3
- 238000000556 factor analysis Methods 0.000 claims description 3
- 238000001325 log-rank test Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000011127 radiochemotherapy Methods 0.000 claims description 2
- 102000007066 Prostate-Specific Antigen Human genes 0.000 description 15
- 108010072866 Prostate-Specific Antigen Proteins 0.000 description 15
- 238000004393 prognosis Methods 0.000 description 9
- 238000001356 surgical procedure Methods 0.000 description 5
- 238000011360 adjunctive therapy Methods 0.000 description 3
- 210000003445 biliary tract Anatomy 0.000 description 3
- 238000011248 postoperative chemotherapy Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000002271 resection Methods 0.000 description 3
- 201000011510 cancer Diseases 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000002560 therapeutic procedure Methods 0.000 description 2
- 238000010200 validation analysis Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 208000002699 Digestive System Neoplasms Diseases 0.000 description 1
- 241000854350 Enicospilus group Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009650 gentamicin protection assay Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000036210 malignancy Effects 0.000 description 1
- 238000012821 model calculation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010837 poor prognosis Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000011371 sixth-line therapy Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/40—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Databases & Information Systems (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Surgery (AREA)
- Urology & Nephrology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention relates to a prediction model establishing method for survival benefit of gallbladder cancer patients after radiotherapy and chemotherapy, which comprises the steps of collecting data, calculating a tendency score PS by adopting a generalized enhanced model GBM, balancing the balance of various groups of variables by adopting an inverse probability weighting method IPTW (IPTW), drawing a Kaplan-Meier survival curve, comparing survival difference of different TNM stages after receiving different auxiliary treatments, and finding that the gallbladder cancer patients in stages III-IV can benefit from the auxiliary treatments; dividing the data of the gallbladder cancer patients in the stages III-IV into a building module and an internal verification group according to the diagnosis time, using SPSS software to successively carry out single-factor and multi-factor COX risk proportion model analysis, and acquiring a final variable to be included in a final nomogram model; a nomogram for predicting the OS is plotted for prediction. And extracting features by collecting conventional clinical pathological indexes of patients. The result is objective and accurate, and the specific individuals are respectively subjected to prediction analysis, so that a credible prediction and analysis result is provided.
Description
Technical Field
The invention relates to a diagnosis and prediction technology, in particular to a method for establishing a prediction model for survival benefit of a gallbladder cancer patient in stage III-IV after radiotherapy and chemotherapy.
Background
Gallbladder cancer (GBC) is the fifth most prevalent of gastrointestinal tumors, with an annual incidence of 2 to 27 per 10 million people. Although relatively rare, GBC is the most aggressive malignancy of the biliary tract, with a 5-year survival rate of about 5%. Surgical resection is the only method of treating GBC. However, the recurrence rate after complete resection is high, often leading to poor prognosis. Therefore, adjuvant treatments such as postoperative chemotherapy, chemotherapy in combination with radiotherapy are applied to GBC patients to improve patient prognosis.
However, because of the low incidence rate of GBC, the difference of the GBC auxiliary treatment indication and the curative effect of the combination of postoperative chemotherapy and chemotherapy radiotherapy is controversial. The experts in diagnosis and treatment of biliary tract tumor diagnosis and treatment in the Chinese society of clinical oncology suggest GBC patients in stages III-IV to receive postoperative adjuvant therapy, and the experts in diagnosis and treatment of biliary tract surgery group suggest GBC patients in stage II and above to receive postoperative adjuvant therapy. In addition, the difference of the curative effects of the postoperative chemotherapy and the chemotherapy combined radiotherapy is lack of deep research, and no better method for predicting the prognosis of the GBC patient receiving the chemotherapy or the chemotherapy combined radiotherapy exists at present.
Disclosure of Invention
Aiming at the problem of later-stage treatment prediction of a gallbladder cancer patient, the establishment method of the prediction model for survival benefit of the gallbladder cancer patient after radiotherapy and chemotherapy is provided, and the prediction model can be effectively applied to guidance of auxiliary treatment of the gallbladder cancer patient in the stage III-IV.
The technical scheme of the invention is as follows: a method for establishing a prediction model of survival benefit of a gallbladder cancer patient after radiotherapy and chemotherapy specifically comprises the following steps:
1) collecting data: collecting clinical and pathological data including sex, age, tumor differentiation degree, TNM staging, operation mode, lymph node cleaning number, postoperative adjuvant therapy mode, and total survival time;
2) and (3) screening data: calculating a tendency score PS by adopting a generalized enhanced model GBM, balancing the sex, age, tumor differentiation degree, operation mode and lymph node clearing number variable of patients without adjuvant therapy, chemotherapy and chemotherapy combined radiotherapy in each group by adopting an inverse probability weighting method IPTW, drawing a Kaplan-Meier survival curve, comparing survival differences after different TNM stages receive different adjuvant therapies, and finding that the gallbladder cancer patients in stages III-IV can benefit from the adjuvant therapy;
3) dividing the data of the stage III-IV gallbladder cancer patient in the step 1) into a building module group and an internal verification group according to the diagnosis time, using SPSS software, firstly analyzing the prediction capability of the building module group on the total survival time OS through a single-factor COX risk proportion model, wherein the prediction capability of the building module group on the sex, the age, the differentiation grade, the T stage, the N stage, the operation mode, the lymph node cleaning quantity and the auxiliary treatment mode variable are considered to be obviously related to the OS in the analysis result and are brought into multi-factor analysis;
4) inputting all factors related to OS obtained by analyzing in the step 3), performing multi-factor COX risk proportion model analysis, and bringing variables with p values smaller than 0.05 of analysis results into a final nomogram model;
5) drawing a nomogram for predicting the OS by using survivval and rms packages of the R software according to the variables obtained by analyzing in the step 4);
6) after the nomogram model is established, the model is verified using the internal verification group and the external verification group data.
Further, the step 2) comprises the following specific steps:
2.1) GBM model calculates PS: the general model is formed by accumulating a plurality of submodels according to corresponding weights, the assumed space of the submodels is not particularly limited and can be any discrete or continuous function, and the GBM can output probability values according to categories by means of an iterative algorithm, namely each group of PS;
2.2) IPTW samples using trend score PS: on the basis of obtaining the tendency score by calculation, a tendency score weighting method gives each research object a corresponding weight for weighting by using the principle of a standardized method through the tendency score value, so that the tendency scores in each group are uniformly distributed, and the purpose of eliminating the influence of confounding factors is achieved, wherein the sample weight w is 2.1) and the reciprocal of PS is obtained by calculation, and w is 1/PS;
2.3) comparing the difference of each group of variables before and after IPTW by using absolute standard difference, wherein when the absolute standard difference is less than 10%, the balance among the groups is considered to be better;
2.4) survival analysis: and (3) carrying out survival analysis on the samples before and after IPTW weighting by a Kaplan-Meier method, comparing the difference of survival curves by adopting log-rank test, and testing the standard alpha to be 0.05.
Further, the model is verified by C-index and consistency curve analysis.
The invention has the beneficial effects that: the method for establishing the prediction model for survival benefit of the gallbladder cancer patient after radiotherapy and chemotherapy has simple and convenient operation process, is visual and is easy to repeat. Can be completed by the inpatient. And extracting features by collecting conventional clinical pathological indexes of patients. The result is objective and accurate, and the specific individuals are respectively subjected to prediction analysis, so that a credible prediction and analysis result is provided.
Drawings
FIG. 1a is a graph showing the difference between baseline indicators before and after NCRT \ CT matching and weighting for patients with stage II GBC in accordance with the present invention;
FIG. 1b is a graph of the change in baseline index difference before and after NCRT \ CRT matching and weighting for stage II GBC patients according to the present invention;
FIG. 1c is a graph of baseline index variation before and after CT \ CRT matching and weighting for a stage II GBC patient according to the present invention;
FIG. 2a is a graph of K-M survival prior to matching for stage II GBC patients according to the invention;
FIG. 2b is a graph of K-M survival after matching for stage II GBC patients according to the invention;
FIG. 3a is a graph showing the difference between the baseline indicators before and after NCRT \ CT matching and weighting for patients with stage III-IV GBC according to the present invention;
FIG. 3b is a graph of the change in baseline index difference before and after NCRT \ CRT matching and weighting for patients with stage III-IV GBC in accordance with the present invention;
FIG. 3c is a graph of the baseline index variation before and after CT \ CRT matching and weighting for patients with GBC of stage III-IV according to the present invention;
FIG. 4a is a graph of K-M survival prior to matching for stage III-IV GBC patients according to the invention;
FIG. 4b is a graph of K-M survival after matching of stage III-IV GBC patients according to the invention;
FIG. 5 is a nomogram of the prognosis of patients with GBC stages III-IV according to the invention;
FIG. 6a is a graph of the consistency of predicted survival rates at 12, 24, 36 months obtained from modeling group data in accordance with the present invention;
FIG. 6b is a graph of the consistency of predicted survival rates at 12, 24, and 36 months obtained from internal validation cohort data in accordance with the present invention;
fig. 6c is a graph of the consistency of predicted survival rates at 12, 24, and 36 months obtained for the external validation cohort data of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
A method for establishing a prediction model for survival benefit of a gallbladder cancer patient after radiotherapy and chemotherapy comprises the following specific steps:
1. collecting data: 2689 GBC patients who received surgery in the SEER database in 2004-. Clinical and pathological data were collected as follows: gender, age, degree of tumor differentiation, TNM staging, mode of surgery, number of lymph nodes cleared, mode of adjuvant treatment after surgery, total survival time (OS).
2. 2689 patients with GBC from SEER database were staged by TNM into stages II and III-IV. Calculating a trend score (PS) by adopting a Generalized enhanced model (GBM), balancing variables such as sex, age, tumor differentiation degree, operation mode, lymph node clearing number and the like of patients without adjuvant therapy, chemotherapy and chemotherapy combined radiotherapy in each group by adopting an inverse probability weighting method (IPTW), drawing a Kaplan-Meier survival curve, comparing survival differences after different adjuvant therapies are received in stages II and III-IV, and finding that the GBC patients in stages III-IV can benefit from the adjuvant therapy. Thereby screening the data validity.
2.1 GBM model calculation sample PS
Classical PS estimates are mostly derived by using parametric linear logistic regression models or variable selection techniques to screen out interactive terms or non-linear terms, which may miss important confounding variables or misassigning functional relationships between covariates and process choices in the model. The invention is based on a GBM model, which is a model that combines multiple regression trees to obtain an iterative process of complex nonlinear relationships between intervention allocation and preprocessing covariates without overfitting data. The total model is formed by accumulating a plurality of submodels according to corresponding weights, the assumed space of the submodels is not particularly limited and can be any discrete or continuous function. The GBM, by means of an iterative algorithm, outputs probability values, i.e. groups of PS, in accordance with the categories.
For GBC patients in stage II, their age, sex, differentiation grade, mode of operation, number of lymph node cleanings, etc. were taken into account and their PS calculated. And (4) calculating PS (prostate specific antigen) for the GBC patients in stages III-IV by taking the GBC patients in stages III-IV into the age, sex, differentiation grade, T stage, N stage, operation mode, lymph node clearing number and other variables. This is achieved by the Tweng package of the R software (V4.0.2).
2.2 IPTW samples Using Trend scores
Assuming no unidentified confounders exist, the weighting adjustment adjusts the data based on two probabilities of counterfactual events under certain conditions: namely, the two cases of processing factors and not processing are assumed to be received by each observation object. The estimated weight of the tendency score is used for weighting each observation unit to generate a virtual standard population (the generation of the virtual population refers to the number of persons who copy the weight), and in the virtual population, two groups of mixed factors tend to be consistent and are similar to a certain pre-selected standard population distribution. The tendency score weighting method is based on calculating the tendency score, and based on the principle of a standardized method, each research object is given a corresponding weight for weighting through the tendency score value, so that the tendency scores in each group are distributed consistently, and the purpose of eliminating the influence of confounding factors is achieved. Thus the weighted method of predisposition score is a standardized method based on individuals. IPTW is a weighting method that adjusts for all observed objects as a "standard population". The sample weight w is the reciprocal of PS calculated in 2.1, w is 1/PS;
2.3 use of absolute Standard Difference(Absolute Standard differentiated difference) comparison of IPTW Prep (P)treat) After (P)control) The variables in each group are different by d.
Absolute standard deviation
It is generally accepted that the balance between groups is better when the absolute standard deviation is less than 10%. It can be seen from figures 1a, 1b,1c that the GBC stage II patient variables receiving different adjunctive therapies after IPTW weighting tend to balance. It can be seen from figures 3a, 3b,3c that the GBC stage III-IV patient variables weighted with IPTW and receiving different adjunctive therapies tend to be balanced.
2.4 survival assay
And (3) carrying out survival analysis on the samples before and after IPTW weighting by a Kaplan-Meier method, comparing the difference of survival curves by adopting log-rank test, and testing the standard alpha to be 0.05. FIGS. 2a and 2b show the survival of the GBC patients before and after IPTW weighting in II stage, respectively, with no significant difference in prognosis for the NCRT, CT and CRT patients. FIGS. 4a and 4b show the survival of patients with GBC in stages III-IV before and after IPTW as well as different adjuvant therapy regimens, showing that patients with GBC in stages III-IV receive CT therapy with superior prognosis of CRT therapy and have better prognosis than NCRT. Therefore, patients with stage III-IV GBC benefit from CT, CRT adjunctive therapy.
3. The method comprises the steps of dividing 1496 GBC patients subjected to operations in stages III-IV in the SEER database into a building group (2004 + 2012 confirmed) and an internal verification group (2013 + 2015 confirmed), and taking 88 GBC patients subjected to operations in the Zhongshan hospital affiliated to the university of Fudan in 2016 + 2018 as an external verification group.
4. Using SPSS software (V19.0 statistical software), the predictive power of variables such as gender, age, differentiation grade, T stage, N stage, surgical procedure, number of lymph node cleanings, adjuvant therapy procedure, etc. in the modeled group on OS was first analyzed by a one-way COX risk ratio model, and statistically significant (p <0.05) variables in the analysis results were considered to be significantly associated with OS and included in the multi-factor analysis. Wherein variables such as sex, age, grade of differentiation, stage T, stage N lymph node clearing number, adjuvant therapy modality, etc. are significantly related to the patient OS in the single-factor COX risk ratio model.
5. Inputting all factors related to the OS obtained by analyzing in the step 4, carrying out multi-factor COX risk proportion model analysis, and bringing variables with the analysis result p value smaller than 0.05 into the final nomogram model. Age, differentiation grade, T stage, N stage, number of lymph node sweeps, adjuvant treatment modality were significantly associated with OS in the multifactorial COX risk ratio model.
6. The variables obtained from step 5 were analyzed and nomograms for predicting OS were plotted using survivval and rms packages of R software (V4.0.2). In the alignment chart of FIG. 5, the first line is the scale of scores, ranging from 0 to 10. Age of the second line, differentiation grade of the third line, T stage of the fourth line, N stage of the fifth line, lymph node clearing number of the sixth line and adjuvant therapy mode of the seventh line. The eighth row is the total score, with scores ranging from 0-35. The ninth behavior is 1-year survival rate, the tenth behavior is 2-year survival rate, the eleventh behavior is 3-year survival rate, and the twelfth is median survival time.
7. After the nomogram model is established, the model is verified using the internal verification group and the external verification group data. The model was verified by C-index and consistency curve analysis. The C-index indexes of the modeling group, the internal verification group and the external verification group are respectively 0.673 (95% CI:0.654-0.692),0.707 (95% CI: 0.677-0. 739) and 0.729 (95% CI: 0.688-0. 775). The corresponding consistency curves are shown in fig. 6a to 6 c.
For any GBC patient undergoing surgical resection, the actual status of each clinical and pathological variable in columns 2 to 7 of the nomogram (fig. 5) is scored, and the specific score scale for each variable is the score of the position corresponding to column 1. The scores of the 6 variables are summed to determine the total score for the patient, and the corresponding score position is found in column 8, and then the predicted probability of survival of the patient in months 12, 24, 36, and its median OS time can be derived from the corresponding positions of the total score in columns 10-12.
There are confounding factors that are difficult to control for different components in retrospective studies, so bias is often controlled using bias score matching. However, the conventional tendency score matching can only match between two groups, and the GBM model and IPTW method can weight multiple groups. Therefore, the invention firstly adopts the GBM model and the IPTW method to compare the prognosis difference of different stage GBC patients receiving different auxiliary treatments.
The nomogram is a machine learning model which is rich in content, visual and understandable and can accurately evaluate the occurrence probability of the ending event, and is widely applied to prediction of prognosis of malignant tumors. However, there is no comprehensive and accurate set of histogram models for GBC patients that can assess prognosis with different adjuvant therapies.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (3)
1. A method for establishing a prediction model for survival benefit of a gallbladder cancer patient after radiotherapy and chemotherapy is characterized by comprising the following steps:
1) collecting data: collecting clinical and pathological data including sex, age, tumor differentiation degree, TNM staging, operation mode, lymph node cleaning number, postoperative adjuvant therapy mode, and total survival time;
2) and (3) screening data: calculating a tendency score PS by adopting a generalized enhanced model GBM, balancing the sex, age, tumor differentiation degree, operation mode and lymph node clearing number variable of patients without adjuvant therapy, chemotherapy and chemotherapy combined radiotherapy in each group by adopting an inverse probability weighting method IPTW, drawing a Kaplan-Meier survival curve, comparing survival differences after different TNM stages receive different adjuvant therapies, and finding that the gallbladder cancer patients in stages III-IV can benefit from the adjuvant therapy;
3) dividing the data of the stage III-IV gallbladder cancer patient in the step 1) into a building module group and an internal verification group according to the diagnosis time, using SPSS software, firstly analyzing the prediction capability of the building module group on the total survival time OS through a single-factor COX risk proportion model, wherein the prediction capability of the building module group on the sex, the age, the differentiation grade, the T stage, the N stage, the operation mode, the lymph node cleaning quantity and the auxiliary treatment mode variable are considered to be obviously related to the OS in the analysis result and are brought into multi-factor analysis;
4) inputting all factors related to OS obtained by analyzing in the step 3), performing multi-factor COX risk proportion model analysis, and bringing variables with p values smaller than 0.05 of analysis results into a final nomogram model;
5) drawing a nomogram for predicting the OS by using survivval and rms packages of the R software according to the variables obtained by analyzing in the step 4);
6) after the nomogram model is established, the model is verified using the internal verification group and the external verification group data.
2. The method for establishing the prediction model of survival benefit of the gallbladder cancer patient after chemoradiotherapy according to claim 1, wherein the step 2) is realized by the following specific steps:
2.1) GBM model calculates PS: the general model is formed by accumulating a plurality of submodels according to corresponding weights, the assumed space of the submodels is not particularly limited and can be any discrete or continuous function, and the GBM can output probability values according to categories by means of an iterative algorithm, namely each group of PS;
2.2) IPTW samples using trend score PS: on the basis of obtaining the tendency score by calculation, a tendency score weighting method gives each research object a corresponding weight for weighting by using the principle of a standardized method through the tendency score value, so that the tendency scores in each group are uniformly distributed, and the purpose of eliminating the influence of confounding factors is achieved, wherein the sample weight w is 2.1) and the reciprocal of PS is obtained by calculation, and w is 1/PS;
2.3) comparing the difference of each group of variables before and after IPTW by using absolute standard difference, wherein when the absolute standard difference is less than 10%, the balance among the groups is considered to be better;
2.4) survival analysis: and (3) carrying out survival analysis on the samples before and after IPTW weighting by a Kaplan-Meier method, comparing the difference of survival curves by adopting log-rank test, and testing the standard alpha to be 0.05.
3. The method for establishing a predictive model of survival benefit of a gallbladder cancer patient after chemotherapy according to claim 1 or 2, wherein the model is verified by C-index and consistency curve analysis.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110994379.2A CN113555123A (en) | 2021-08-27 | 2021-08-27 | Method for establishing prediction model of survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110994379.2A CN113555123A (en) | 2021-08-27 | 2021-08-27 | Method for establishing prediction model of survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113555123A true CN113555123A (en) | 2021-10-26 |
Family
ID=78106107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110994379.2A Pending CN113555123A (en) | 2021-08-27 | 2021-08-27 | Method for establishing prediction model of survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113555123A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016094330A2 (en) * | 2014-12-08 | 2016-06-16 | 20/20 Genesystems, Inc | Methods and machine learning systems for predicting the liklihood or risk of having cancer |
| EP3310260A1 (en) * | 2015-06-22 | 2018-04-25 | Sunnybrook Research Institute | Systems and methods for prediction of tumor response to chemotherapy using pre-treatment quantitative ultrasound parameters |
| US20180374583A1 (en) * | 2017-05-16 | 2018-12-27 | Abraxis Bioscience, Llc | Nomogram and survival predictions for pancreatic cancer |
| CN111542260A (en) * | 2018-04-10 | 2020-08-14 | 希尔-罗姆服务公司 | Patient risk assessment based on data from multiple sources in a healthcare facility |
| EP3758026A1 (en) * | 2019-06-28 | 2020-12-30 | Hill-Rom Services, Inc. | Patient risk assessment based on data from multiple sources in a healthcare facility |
| CN112992274A (en) * | 2021-03-31 | 2021-06-18 | 青岛泱深生物医药有限公司 | Method and system for constructing disease risk prediction model based on sequencing and machine learning |
| CN113571189A (en) * | 2021-08-23 | 2021-10-29 | 复旦大学附属中山医院 | Establishment method of prediction model for survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy |
-
2021
- 2021-08-27 CN CN202110994379.2A patent/CN113555123A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016094330A2 (en) * | 2014-12-08 | 2016-06-16 | 20/20 Genesystems, Inc | Methods and machine learning systems for predicting the liklihood or risk of having cancer |
| CN109036571A (en) * | 2014-12-08 | 2018-12-18 | 20/20基因系统股份有限公司 | The method and machine learning system of a possibility that for predicting with cancer or risk |
| EP3310260A1 (en) * | 2015-06-22 | 2018-04-25 | Sunnybrook Research Institute | Systems and methods for prediction of tumor response to chemotherapy using pre-treatment quantitative ultrasound parameters |
| US20180374583A1 (en) * | 2017-05-16 | 2018-12-27 | Abraxis Bioscience, Llc | Nomogram and survival predictions for pancreatic cancer |
| CN111542260A (en) * | 2018-04-10 | 2020-08-14 | 希尔-罗姆服务公司 | Patient risk assessment based on data from multiple sources in a healthcare facility |
| EP3758026A1 (en) * | 2019-06-28 | 2020-12-30 | Hill-Rom Services, Inc. | Patient risk assessment based on data from multiple sources in a healthcare facility |
| CN112992274A (en) * | 2021-03-31 | 2021-06-18 | 青岛泱深生物医药有限公司 | Method and system for constructing disease risk prediction model based on sequencing and machine learning |
| CN113571189A (en) * | 2021-08-23 | 2021-10-29 | 复旦大学附属中山医院 | Establishment method of prediction model for survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Leslie et al. | Comparison between various fracture risk assessment tools | |
| Penaloza et al. | Comparison of the Wells score with the simplified revised Geneva score for assessing pretest probability of pulmonary embolism | |
| Venetz et al. | A comparison of the original and simplified Pulmonary Embolism Severity Index | |
| Zhu et al. | Application of machine learning algorithms to predict central lymph node metastasis in T1-T2, non-invasive, and clinically node negative papillary thyroid carcinoma | |
| CN112289455A (en) | Artificial intelligence neural network learning model construction system and construction method | |
| Daskivich et al. | An age adjusted comorbidity index to predict long-term, other cause mortality in men with prostate cancer | |
| Auble et al. | Comparison of four clinical prediction rules for estimating risk in heart failure | |
| Steuer et al. | Predictors and outcomes of venous thromboembolism in hospitalized lung cancer patients: A Nationwide Inpatient Sample database analysis | |
| CN113571189A (en) | Establishment method of prediction model for survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy | |
| Lin et al. | A prognostic model to predict mortality among non–small-cell lung cancer patients in the US Military Health System | |
| Valdes et al. | Salvage HDR brachytherapy: multiple hypothesis testing versus machine learning analysis | |
| Li et al. | Prognostic nomogram for overall survival in extranodal natural killer/T-cell lymphoma patients | |
| Jia et al. | Development and validation of prognostic nomogram in ependymoma: A retrospective analysis of the SEER database | |
| Wang et al. | Comparison and screening of different risk assessment models for deep vein thrombosis in patients with solid tumors | |
| CN113345592B (en) | Construction and diagnosis equipment for acute myeloid leukemia prognosis risk model | |
| Han et al. | Development and validation of an artificial intelligence mobile application for predicting 30-day mortality in critically ill patients with orthopaedic trauma | |
| Lee et al. | Primary anatomical site as a prognostic factor for pleomorphic liposarcoma | |
| CN113555123A (en) | Method for establishing prediction model of survival benefit of gallbladder cancer patient after radiotherapy and chemotherapy | |
| Zhang et al. | Development and validation of a prognostic nomogram for the pre-treatment prediction of early metachronous metastasis in endemic nasopharyngeal carcinoma: a big-data intelligence platform-based analysis | |
| Medrano et al. | Understanding race-based medicine and its impact on radiology | |
| Morillo et al. | Prognostic scores for acute pulmonary embolism | |
| Geraghty et al. | Use of temporally validated machine learning models to predict outcomes of percutaneous nephrolithotomy using data from the British association of urological surgeons percutaneous nephrolithotomy audit | |
| Hwang et al. | Administrative data is as good as medical chart review for comorbidity ascertainment in patients with infections in Singapore | |
| CN112259231A (en) | High-risk gastrointestinal stromal tumor patient postoperative recurrence risk assessment method and system | |
| Roy et al. | Frailty Indices in Metastatic Spine Tumor Surgery: A Narrative Review |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20211026 |
|
| WD01 | Invention patent application deemed withdrawn after publication |