CN117524499A - Method and system for evaluating curative effect of immunotherapy of advanced non-small cell lung cancer - Google Patents
Method and system for evaluating curative effect of immunotherapy of advanced non-small cell lung cancer Download PDFInfo
- Publication number
- CN117524499A CN117524499A CN202311486758.6A CN202311486758A CN117524499A CN 117524499 A CN117524499 A CN 117524499A CN 202311486758 A CN202311486758 A CN 202311486758A CN 117524499 A CN117524499 A CN 117524499A
- Authority
- CN
- China
- Prior art keywords
- features
- immunotherapy
- clinical
- model
- label
- 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
- 238000009169 immunotherapy Methods 0.000 title claims abstract description 51
- 208000002154 non-small cell lung carcinoma Diseases 0.000 title claims abstract description 42
- 208000029729 tumor suppressor gene on chromosome 11 Diseases 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 34
- 230000000694 effects Effects 0.000 title claims abstract description 28
- 238000007637 random forest analysis Methods 0.000 claims abstract description 30
- 238000012216 screening Methods 0.000 claims abstract description 21
- 230000004927 fusion Effects 0.000 claims abstract description 20
- 238000004364 calculation method Methods 0.000 claims abstract description 8
- 238000006757 chemical reactions by type Methods 0.000 claims abstract description 8
- 238000002372 labelling Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 21
- 206010028980 Neoplasm Diseases 0.000 claims description 15
- 230000000391 smoking effect Effects 0.000 claims description 13
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000011282 treatment Methods 0.000 claims description 9
- 238000012360 testing method Methods 0.000 claims description 6
- 238000002619 cancer immunotherapy Methods 0.000 claims description 5
- 238000011156 evaluation Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 230000001225 therapeutic effect Effects 0.000 claims 1
- 238000012545 processing Methods 0.000 abstract description 5
- 238000002474 experimental method Methods 0.000 description 8
- 230000004044 response Effects 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000004088 simulation Methods 0.000 description 5
- 230000006907 apoptotic process Effects 0.000 description 4
- 238000004590 computer program Methods 0.000 description 4
- 238000003066 decision tree Methods 0.000 description 4
- 229940045513 CTLA4 antagonist Drugs 0.000 description 3
- 239000000090 biomarker Substances 0.000 description 3
- 229940076838 Immune checkpoint inhibitor Drugs 0.000 description 2
- 101000668058 Infectious salmon anemia virus (isolate Atlantic salmon/Norway/810/9/99) RNA-directed RNA polymerase catalytic subunit Proteins 0.000 description 2
- 102000037984 Inhibitory immune checkpoint proteins Human genes 0.000 description 2
- 108091008026 Inhibitory immune checkpoint proteins Proteins 0.000 description 2
- 230000005911 anti-cytotoxic effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000012274 immune-checkpoint protein inhibitor Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 208000000649 small cell carcinoma Diseases 0.000 description 2
- 102000008096 B7-H1 Antigen Human genes 0.000 description 1
- 108010074708 B7-H1 Antigen Proteins 0.000 description 1
- 102000008203 CTLA-4 Antigen Human genes 0.000 description 1
- 108010021064 CTLA-4 Antigen Proteins 0.000 description 1
- 208000032818 Microsatellite Instability Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000002512 chemotherapy Methods 0.000 description 1
- 238000000546 chi-square test Methods 0.000 description 1
- 238000011284 combination treatment Methods 0.000 description 1
- 238000002591 computed tomography Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000036737 immune function Effects 0.000 description 1
- 230000028993 immune response Effects 0.000 description 1
- 230000007688 immunotoxicity Effects 0.000 description 1
- 231100000386 immunotoxicity Toxicity 0.000 description 1
- 230000005923 long-lasting effect Effects 0.000 description 1
- 210000004072 lung Anatomy 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000869 mutational effect Effects 0.000 description 1
- 238000007500 overflow downdraw method Methods 0.000 description 1
- 230000007170 pathology Effects 0.000 description 1
- 238000013077 scoring method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000004797 therapeutic response Effects 0.000 description 1
- 210000003171 tumor-infiltrating lymphocyte Anatomy 0.000 description 1
- 238000001521 two-tailed test Methods 0.000 description 1
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
- G16H50/70—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/211—Selection of the most significant subset of features
- G06F18/2113—Selection of the most significant subset of features by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
-
- 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
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention provides a method and a system for evaluating the curative effect of immunotherapy of advanced non-small cell lung cancer, which relate to the technical field of image processing and comprise the steps of acquiring original CT image data and clinical information archive data of an advanced non-small cell lung cancer patient before immunotherapy, extracting wavelet features and Gaussian-Laplace features in original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model; classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, outputting a matching result of a prediction label and the label by using a clinical model, fusing a random forest with the prediction result of the clinical model, and outputting a result of predicting the immune therapy curative effect in the fusion model.
Description
Technical Field
The disclosure relates to the technical field of image processing, in particular to a method and a system for evaluating the curative effect of immunotherapy of advanced non-small cell lung cancer.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Immunotherapy is a milestone treatment of advanced non-small cell cancers (NSCLC, advanced non-small cell cancers) with Immune Checkpoint Inhibitors (ICIs), which have anti-cytotoxic t lymphocyte-associated protein 4 (CTLA-4, anti-cytotoxic t lymphocyte-associated protein 4), apoptosis protein-1 (PD-1, apoptosis protein-1) and apoptosis ligand-1 (PD-L1, apoptosis ligand-1), and play a vital role in response rate and long-lasting disease remission, play a "braking" role in immune function, and effectively eliminate cancer cells.
Currently, immunotherapy is applied to solid tumors, with better efficacy on advanced NSCLC compared to chemotherapy immunotherapy. However, not all patients with advanced NSCLC respond well to immunotherapy and a small proportion of patients experience serious side effects. To avoid immunotoxicity and develop alternative or combination treatment regimens, the study and development of biomarkers for immunotherapy is an important and challenging goal for advanced NSCLC.
Previous studies have evaluated the performance of several biomarkers in the evaluation of the efficacy of immunotherapy, including tumor mutational burden, tumor infiltrating lymphocytes, and microsatellite instability, all acting as biomarkers for predicting the immune response to advanced NSCLC. Although good predictors are obtained, these efficacy predictors need to be diagnosed by invasive pathology or genetic sequencing, so these indicators are not suitable for monitoring tumor response to immunotherapy during treatment.
Disclosure of Invention
In order to solve the problems, the present disclosure provides a method and a system for evaluating the curative effect of immunotherapy of advanced non-small cell lung cancer, which extract obvious features on CT images of the advanced non-small cell lung cancer receiving the immunotherapy, and construct a predictive model of the type of therapeutic response by using the features, thereby realizing early prediction and evaluation of the unknown immunotherapy effect.
According to some embodiments, the present disclosure employs the following technical solutions:
a method for evaluating the efficacy of immunotherapy for advanced non-small cell lung cancer, comprising:
acquiring original CT image data and clinical information archive data of a patient with advanced non-small cell lung cancer before immunotherapy, and preprocessing the original CT image data;
extracting wavelet features and Gaussian-Laplace features from original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model;
classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, outputting a matching result of a prediction label and the label by using a clinical model, fusing a random forest with the prediction result of the clinical model, and outputting a result of predicting the immune therapy curative effect in the fusion model.
According to some embodiments, the present disclosure employs the following technical solutions:
an advanced non-small cell lung cancer immunotherapy efficacy assessment system comprising:
the data acquisition module is used for acquiring original CT image data and clinical information archive data of the advanced non-small cell lung cancer patient before immunotherapy and preprocessing the original CT image data;
the feature screening module is used for extracting wavelet features and Gaussian-Laplace features from original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model;
classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, and outputting matching results of the prediction labels and the label labels by using a clinical model;
and the fusion prediction module is used for fusing the random forest with the prediction result of the clinical model and outputting the result of the immunotherapy curative effect prediction in the fusion model.
According to some embodiments, the present disclosure employs the following technical solutions:
a computer readable storage medium having stored therein a plurality of instructions adapted to be loaded by a processor of a terminal device and to perform the method of evaluating the efficacy of immunotherapy of advanced non-small cell lung cancer.
According to some embodiments, the present disclosure employs the following technical solutions:
a terminal device comprising a processor and a computer readable storage medium, the processor configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the method of evaluating the efficacy of immunotherapy of advanced non-small cell lung cancer.
Compared with the prior art, the beneficial effects of the present disclosure are:
the method comprises the steps of extracting and analyzing image histology characteristics and clinical characteristics of an immune-treated advanced non-small cell lung cancer patient, respectively establishing random forest and clinical model based on the image histology characteristics and the clinical characteristics, and fusing the random forest and the clinical model to obtain a fusion model. Experimental simulation verifies that the fusion model further improves the accuracy of predicting the curative effect of the immunotherapy by using the random forest model only. Meanwhile, on CT image data obtained by using different data center scanning protocols, the method disclosed by the invention can still obtain stable results for predicting the curative effect of immunotherapy.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate and explain the exemplary embodiments of the disclosure and together with the description serve to explain the disclosure, and do not constitute an undue limitation on the disclosure.
Fig. 1 is a flow chart of a method for evaluating the efficacy of an advanced non-small cell lung cancer immunotherapy based on CT images according to an embodiment of the disclosure.
FIG. 2 is a diagram of a simulation experiment of an embodiment of the present disclosure;
fig. 3 is a schematic representation of an experimental subject working curve-ROC for an embodiment of the present disclosure.
The specific embodiment is as follows:
the disclosure is further described below with reference to the drawings and examples.
It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments in accordance with the present disclosure. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
Example 1
In one embodiment of the present disclosure, a method for evaluating the efficacy of immunotherapy for advanced non-small cell lung cancer is provided, comprising the steps of:
step one: acquiring original CT image data and clinical information archive data of a patient with advanced non-small cell lung cancer before immunotherapy, and preprocessing the original CT image data;
step two: extracting wavelet features and Gaussian-Laplace features from original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model;
step three: classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, outputting a matching result of a prediction label and the label by using a clinical model, fusing a random forest with the prediction result of the clinical model, and outputting a result of predicting the immune therapy curative effect in the fusion model.
As an embodiment, the object of the present disclosure may be specifically achieved by the following embodiment steps:
step (1): reading original CT image data and clinical information files of advanced non-small cell lung cancer patients in different data centers before immunotherapy;
step (2): preprocessing an original CT image;
step (3): extracting and screening image histology characteristics of the CT image;
step (4): establishing a random forest model to predict curative effect of training set data, and using the model obtained by training for performance evaluation of a test set;
step (5): screening clinical characteristics in the clinical information file;
step (6): predicting curative effect of the training set data by using a clinical model, and using the model obtained by training for performance evaluation of a test set;
step (7): model fusion is carried out on the random forest model and the clinical model, and the performance of the fusion model is evaluated by utilizing a test set;
step (8): and obtaining the accuracy of the fusion model on the result of the immunotherapy effect prediction.
In one embodiment, in step (2), the imaging differences between the CT images are eliminated before extracting the image histology features, so that the original CT images are preprocessed, and the preprocessing step includes windowing and gray value normalization. Firstly, the CT value of an original CT image is adjusted to a special observation value of the lung, namely, the window level is set to 300HU, and the window width is set to 1400HU; since the gray scale range of a natural image that can be recognized by the human eye is 0-255, and extraneous tissue in the tumor image can interfere with the field of view of the observer, the gray scale value of the CT tumor image is normalized to 0-255.
Specific:
step (21): window adjustment: windowing is carried out on the original CT image data, the window level of the CT value of the image is set to be 300HU, and the window width is set to be 1400HU;
step (22): gray scale normalization: and normalizing the gray level value of the image after the window adjustment to be between 0 and 255 to obtain a CT preprocessing image, which is hereinafter referred to as a CT image.
In step (3), the standardization of the image histology features is to eliminate the dimension difference between the features, and provide more accurate and real contrast data for the subsequent feature extraction step.
LASSO regression is characterized by variable screening and complexity adjustment while fitting a generalized linear model, therefore, a LASSO regression algorithm can be used to perform feature screening. Complexity adjustment in an algorithm refers to controlling the complexity of the model through a series of parameters, thereby avoiding overfitting. The degree of complexity adjustment of LASSO regression is controlled by a parameter lambda, and the larger lambda is, and the penalty force on the linear model with more variables is larger, so that a variable with fewer variables is finally obtained, and representative variable combinations are compared.
Based on the above principle, the step (3) specifically comprises:
step (31), feature extraction: three types of image histology features are extracted on the CT image. The original features are directly extracted from the CT image, and the wavelet features and the Gaussian-Laplace features are extracted from the CT image subjected to wavelet filtering and Gaussian-Laplace filtering. Wherein each class of features includes shape features, texture features, histogram features;
step (32), feature normalization: the Z-score is standardized for the image histology characteristics, and the specific calculation formula is as follows:
wherein X is std For a normalized feature, u is the mean of the feature, σ is the standard deviation of the feature;
step (33) feature screening: the existing LASSO algorithm is used to calculate the feature weighted ranking to select the optimal features in the training set.
Furthermore, in the step (4), the earliest model of the random forest classifier using the Bagging algorithm combines a plurality of decision trees, each time the data set is randomly selected with a return, and part of the characteristics are randomly selected as input, so that the robustness of the model is greatly improved, and the relevance between single models is reduced.
The structural characteristics of the combination of the decision trees enable the random forest to process a large number of input variables; the importance of the variables can be assessed in deciding the category; when a forest is built, an unbiased estimate can be generated internally for generalized errors; the inclusion of a good method can estimate the lost material and maintain accuracy if a significant portion of the material is lost; for unbalanced classified datasets, errors can be balanced; can be extended to be applied to unlabeled data, such as data that is typically clustered using unsupervised clustering, and also to detect deviations and view data; the learning process of the model is fast.
Thus, specifically, step (4) is specifically:
step (41), constructing a random forest classifier, wherein a decision tree is 20, the maximum depth of the tree is 4, and leaf nodes are 5, and the decision tree is used as model training parameters;
marking training set data according to the reaction type, wherein PD is marked as 1, SD is marked as 1, and PR is marked as 0;
step (43), training label prediction on a training set by using a random forest model to obtain a training model;
step (44), using the obtained training model for a test set to judge the label;
and (45) matching the obtained prediction label with the label to obtain the accuracy of the treatment effect prediction of the immunotherapy.
Further, in step (5), because clinical features involve both classification variables and continuous variables, the use of strict feature screening algorithms results in the loss of clinical features with classification potential, and statistical methods are used to analyze their relevance to classification. In particular, the method comprises the steps of,
step (51) clinical feature classification: the clinical characteristics in the clinical information data comprise gender, smoking history, TNM stage, tumor histology and age, and the characteristics are divided into classification variables and continuous variables;
step (52) adopts chi-square test to analyze classification variables in clinical characteristics, specifically comprising gender, smoking history, TNM stage and tumor histology type;
step (53) analyzing continuous variables in clinical characteristics, particularly age, by using a Mann-Whitney two-tailed test;
step (54) obtains clinical features with efficacy prediction potential, namely gender and smoking history respectively through screening.
Furthermore, in order to simulate the prediction principle of the random forest, the clinical model uses a scoring method to replace the prediction score output by the machine learning model, so that the prediction result is determined, and the random forest and the clinical model are further convenient to fuse better.
Therefore, the prediction result given 0,0.5,1 based on the clinical characteristics obtained by the screening was specifically 1 for female smoking, 0.5 for male smoking, 0.5 for female non-smoking, and 0 for male non-smoking.
According to the above, the step (6) specifically comprises:
step (61), establishing a clinical model: predicting based on the clinical characteristics obtained by screening, specifically, predicting female smoking to be 1, predicting male smoking to be 0.5, predicting female non-smoking to be 0.5, predicting male non-smoking to be 0, and taking the above results as prediction results of a clinical model;
step (62), the training set is subjected to label judgment by utilizing the clinical model, so that a reliable clinical model is obtained;
step (63), taking the obtained training model as a parameter of a test set to carry out label judgment;
and (64) matching the obtained prediction label with the label to obtain the accuracy of the treatment effect prediction of the immunotherapy.
The principle of the step (7) is as follows:
in order to improve the performance of the model, a fusion method is applied to fuse the random forest with the prediction result of the clinical model. Defining the prediction result of the random forest as M R The prediction result of the clinical model is defined as M C Then by changing M r And M C The weight of (2) gets the result M of the fusion model F ：
M F ＝ω r ·M R +ω c ·M C
Wherein omega r ，ω c Respectively M R And M C Weights of 0.8 and 0.2, respectively.
Thus, according to the principle described above, step (7) is specifically:
step (71), fusing the random forest with the predicted result of the clinical model: defining the prediction result of the random forest as M R The prediction result of the clinical model is defined as M C And by changing M R And M C The weight of (2) gets the result M of the fusion model F ：
M F ＝ω r ·M R +ω c ·M C
Wherein omega r ，ω c Respectively M R And M C Weights of (2);
and (72) matching the obtained prediction label with the label to obtain the accuracy of the treatment effect prediction of the immunotherapy.
Training the known treatment response label by using the fusion model, using the obtained training model as a parameter for label prediction of unknown treatment response, and comparing the obtained predicted value with the labeled label, thereby obtaining the accuracy of the fusion model on the immune treatment effect prediction result.
The effect of the method for evaluating the curative effect of the advanced non-small cell lung cancer immunotherapy based on CT images can be further illustrated by the following experiments.
1. Simulation experiment:
as shown in fig. 2, the experiment first reads a CT scan image, and then preprocesses the CT image by a pre-programmed program. And extracting features on the preprocessed image to obtain a feature matrix of different tumor immunotherapy reactions.
2. Simulation experiment results
Fig. 3 shows AUC prediction performance curves obtained from the random forest model MR, the clinical model MC, and the fusion model MF, respectively, based on the prediction results.
Tables 1 and 2 show the number of data and the label of advanced non-small cell lung cancer acquired by two central multi-spiral CT scanners used in the two experiments 1 and 2, respectively.
Table 1 data number and signature of advanced non-small cell lung cancer of experiment 1
| Type of tumor response | PD | SD | PR |
| Number of training sets | 32 | 8 | 11 |
| Marking label | 1 | 1 | 0 |
TABLE 2 data number and tag label for advanced non-small cell lung cancer of experiment 2
| Type of tumor response | PD | SD | PR |
| Number of training sets | 6 | 13 | 11 |
| Marking label | 1 | 1 | 0 |
Tables 3 and 4 show the number of predicted data, the number of predicted tags and the number of predicted tags, and it can be seen from both tables that the accuracy of 91% can be achieved in both experiments, and the AUC values were respectively 0.80 or more.
Table 3 number of predicted data, predicted tag, and number of correctly predicted tags in experiment 1
| Type of tumor response | PD | SD | PR |
| Number of training sets | 32 | 8 | 11 |
| Predictive label | 1 | 1 | 0 |
| Correct number of | 31 | 7 | 8 |
Table 4 number of predicted data, predicted tag, and number of correctly predicted tags in experiment 2
| Type of tumor response | PD | SD | PR |
| Number of training sets | 6 | 13 | 11 |
| Predictive label | 1 | 1 | 0 |
| Correct number of | 6 | 12 | 9 |
3. Simulation experiment analysis
From the above graphs and the results of tables 3 and 4, the following conclusions can be drawn: the algorithm provided by the disclosure has higher prediction accuracy, has great advantages for PD prediction, has strong generalization capability, and can obtain more accurate prediction labels.
Example 2
In one embodiment of the present disclosure, there is provided a system for evaluating the efficacy of immunotherapy for advanced non-small cell lung cancer, comprising:
the data acquisition module is used for acquiring original CT image data and clinical information archive data of the advanced non-small cell lung cancer patient before immunotherapy and preprocessing the original CT image data;
the feature screening module is used for extracting wavelet features and Gaussian-Laplace features from original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model;
classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, and outputting matching results of the prediction labels and the label labels by using a clinical model;
and the fusion prediction module is used for fusing the random forest with the prediction result of the clinical model and outputting the result of the immunotherapy curative effect prediction in the fusion model.
Example 3
In one embodiment of the present disclosure, a computer readable storage medium is provided, in which a plurality of instructions are stored, the instructions being adapted to be loaded by a processor of a terminal device and to perform the one advanced non-small cell lung cancer immunotherapy efficacy assessment method.
Example 4
In one embodiment of the disclosure, a terminal device is provided, including a processor and a computer readable storage medium, where the processor is configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the method of evaluating the efficacy of immunotherapy of advanced non-small cell lung cancer.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the specific embodiments of the present disclosure have been described above with reference to the drawings, it should be understood that the present disclosure is not limited to the embodiments, and that various modifications and changes can be made by one skilled in the art without inventive effort on the basis of the technical solutions of the present disclosure while remaining within the scope of the present disclosure.
Claims (10)
1. A method for evaluating the efficacy of immunotherapy for advanced non-small cell lung cancer, comprising:
acquiring original CT image data and clinical information archive data of a patient with advanced non-small cell lung cancer before immunotherapy, and preprocessing the original CT image data;
extracting wavelet features and Gaussian-Laplace features from original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model;
classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, outputting a matching result of a prediction label and the label by using a clinical model, fusing a random forest with the prediction result of the clinical model, and outputting a result of predicting the immune therapy curative effect in the fusion model.
2. The method for evaluating the therapeutic effect of immunotherapy of advanced non-small cell lung cancer according to claim 1, wherein the preprocessing step comprises the steps of adjusting the window and the gray level normalization, adjusting the window level and the window width of the CT value of the image, and adjusting the gray level of the image after the window adjustment to obtain the CT image.
3. The method for evaluating the curative effect of the immunotherapy of the advanced non-small cell lung cancer according to claim 1, wherein the process of extracting the wavelet features and the Gaussian-Laplacian features in the original features of the CT image is as follows: three types of image histology features are extracted on a CT image, firstly, original features are directly extracted on the CT image, and wavelet features and Gaussian-Laplace features are extracted on the CT image subjected to wavelet filtering and Gaussian-Laplace filtering.
4. The method for evaluating the efficacy of immunotherapy for advanced non-small cell lung cancer according to claim 3, wherein each class of features comprises shape features, texture features, and histogram features.
5. The method for evaluating the curative effect of the immunotherapy of the advanced non-small cell lung cancer according to claim 1, wherein the method for screening the optimal characteristics by calculating the characteristic weighted ranking of the extracted characteristics is to use a LASSO method to calculate the characteristic weighted ranking, and perform variable screening and complexity adjustment while fitting a generalized linear model.
6. The method of claim 1, wherein the clinical profile data comprises gender, smoking history, TNM stage, tumor histology and age, and wherein the classification variables comprise gender, smoking history, TNM stage and tumor histology type; the continuous variable is age.
7. The method for evaluating the efficacy of immunotherapy of advanced non-small cell lung cancer according to claim 1, wherein the training process of the clinical model is as follows: establishing a clinical model, and performing label judgment on the training set by using the clinical model to obtain a preliminary clinical model; taking the obtained preliminary clinical model as a parameter of a test set to carry out label judgment; and matching the obtained prediction label with the label to obtain the accuracy of the treatment effect prediction of the immunotherapy.
8. An advanced non-small cell lung cancer immunotherapy efficacy evaluation system, comprising:
the data acquisition module is used for acquiring original CT image data and clinical information archive data of the advanced non-small cell lung cancer patient before immunotherapy and preprocessing the original CT image data;
the feature screening module is used for extracting wavelet features and Gaussian-Laplace features from original features of the CT image, and carrying out calculation feature weighted sorting on the extracted features to screen optimal features; labeling the optimal characteristics according to the reaction types, and outputting a matching result of the predicted label and the labeled label by using a random forest model;
classifying variables and continuous variables in clinical information archive data, screening clinical features in the classified variables and the continuous variables, judging labels, and outputting matching results of the prediction labels and the label labels by using a clinical model;
and the fusion prediction module is used for fusing the random forest with the prediction result of the clinical model and outputting the result of the immunotherapy curative effect prediction in the fusion model.
9. A computer readable storage medium, characterized in that a plurality of instructions are stored, said instructions being adapted to be loaded by a processor of a terminal device and to perform a method for evaluating the efficacy of an immunotherapy of advanced non-small cell lung cancer according to any one of claims 1-7.
10. A terminal device comprising a processor and a computer readable storage medium, the processor configured to implement instructions; a computer readable storage medium for storing a plurality of instructions adapted to be loaded by a processor and to perform a method of assessing the efficacy of an immunotherapy of advanced non-small cell lung cancer according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311486758.6A CN117524499A (en) | 2023-11-08 | 2023-11-08 | Method and system for evaluating curative effect of immunotherapy of advanced non-small cell lung cancer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311486758.6A CN117524499A (en) | 2023-11-08 | 2023-11-08 | Method and system for evaluating curative effect of immunotherapy of advanced non-small cell lung cancer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117524499A true CN117524499A (en) | 2024-02-06 |
Family
ID=89752430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311486758.6A Pending CN117524499A (en) | 2023-11-08 | 2023-11-08 | Method and system for evaluating curative effect of immunotherapy of advanced non-small cell lung cancer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117524499A (en) |
-
2023
- 2023-11-08 CN CN202311486758.6A patent/CN117524499A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Arunkumar et al. | Fully automatic model‐based segmentation and classification approach for MRI brain tumor using artificial neural networks | |
| Jaworek-Korjakowska et al. | Melanoma thickness prediction based on convolutional neural network with VGG-19 model transfer learning | |
| dos Santos Ferreira et al. | Convolutional neural network and texture descriptor-based automatic detection and diagnosis of glaucoma | |
| WO2022007337A1 (en) | Tumor cell content evaluation method and system, and computer device and storage medium | |
| CN112270666A (en) | Non-small cell lung cancer pathological section identification method based on deep convolutional neural network | |
| US20230420072A1 (en) | Method and system for predicting response to immune checkpoint inhibitor | |
| CN101551854B (en) | A processing system of unbalanced medical image and processing method thereof | |
| EP4105943A1 (en) | Method and system for predicting reaction to immuno-oncology agent | |
| EP4075325A1 (en) | Method and system for the classification of histopathological images based on multiple instance learning | |
| WO2023124562A1 (en) | Cell segmentation and typing method, apparatus and device based on machine learning, and medium | |
| CN112488992B (en) | Method, medium and electronic equipment for judging mutation state of epidermal growth factor receptor | |
| Junayed et al. | ScarNet: development and validation of a novel deep CNN model for acne scar classification with a new dataset | |
| Sreejini et al. | Retrieval of pathological retina images using Bag of Visual Words and pLSA model | |
| CN112861881A (en) | Honeycomb lung recognition method based on improved MobileNet model | |
| CN108376567B (en) | Label propagation algorithm-based clinical drug-drug adverse reaction detection method | |
| Mohapatra et al. | Automated invasive cervical cancer disease detection at early stage through deep learning | |
| Babayomi et al. | Convolutional xgboost (c-xgboost) model for brain tumor detection | |
| Kukreti et al. | Detection and Classification of Brain Tumour Using EfficientNet and Transfer Learning Techniques | |
| CN117524499A (en) | Method and system for evaluating curative effect of immunotherapy of advanced non-small cell lung cancer | |
| Chhabra et al. | Comparison of different edge detection techniques to improve quality of medical images | |
| Vijayadeep et al. | A hybrid feature extraction based optimized random forest learning model for brain stroke prediction | |
| Kadhim et al. | Ensemble Model for Prostate Cancer Detection Using MRI Images | |
| Akbarnejad et al. | Deep Fisher vector coding for whole slide image classification | |
| Chowdhury et al. | Segmentation of retina images to detect abnormalities arising from diabetic retinopathy | |
| Safia Naveed | Prediction of breast cancer through Random Forest |
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 |