CN115862857A - Tumor immune subtype prediction method, system and computer equipment - Google Patents
Tumor immune subtype prediction method, system and computer equipment Download PDFInfo
- Publication number
- CN115862857A CN115862857A CN202211488320.7A CN202211488320A CN115862857A CN 115862857 A CN115862857 A CN 115862857A CN 202211488320 A CN202211488320 A CN 202211488320A CN 115862857 A CN115862857 A CN 115862857A
- Authority
- CN
- China
- Prior art keywords
- tumor
- imaging
- data
- immune subtype
- magnetic resonance
- 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
Images
Landscapes
- Magnetic Resonance Imaging Apparatus (AREA)
Abstract
The tumor immune subtype prediction method, the tumor immune subtype prediction system and the computer equipment acquire a nuclear magnetic resonance image dataset of a tumor patient; extracting an imaging feature from the nuclear magnetic resonance image dataset; carrying out feature screening on the imaging features to form an imaging feature set; the immune subtype of the tumor is predicted according to the imaging characteristic set, the prediction method and the system provided by the application can predict the immune subtype of the tumor by adopting a non-invasive method without acquiring tumor tissues, reduce the damage to patients, can be operated repeatedly, realize the dynamic monitoring of the immune subtype of the tumor in the whole disease course, and have controllable technical cost and smaller economic burden on the patients compared with methods such as high-throughput sequencing and tissue staining.
Description
Technical Field
The application relates to the technical field of medical image processing, in particular to a tumor immune subtype prediction method, a tumor immune subtype prediction system and computer equipment.
Background
The immune subtype of the tumor refers to a subtype generated by classifying the tumor according to the infiltration level of immune cells and stromal cells in a tumor immune microenvironment. The immune subtype of the tumor plays an important guiding role in the selection of a tumor treatment scheme, and directly influences the prognosis of a patient. Therefore, the immune subtype of the tumor needs to be first defined when making a treatment decision. On the other hand, the immune microenvironment of the tumor can be dynamically changed in real time under the influence of the treatment and the tumor itself. As the disease condition of the patient worsens or improves, the immune subtype of the tumor also changes. When a follow-up treatment and disease monitoring strategy is formulated, a technology capable of monitoring the change of tumor immune subtypes in real time is also urgently needed.
The existing tumor immune subtype classification method mainly comprises a tissue section staining counting method, an immunofluorescence staining counting method and a transcriptome and single cell transcriptome deconvolution algorithm. The tissue section staining technique method is from a research guide of lymphocyte infiltration of solid tumors provided by the International immunooncology working group, and immune cell counting and immune typing are carried out based on a hematoxylin-eosin staining method. Immunofluorescence staining methods are also very similar, and the cells are subjected to marker specificity staining, cell attributes are judged, the number of the cells is counted, and then immune typing is carried out. And analyzing the expression quantity of immune cell specific gene markers to estimate the content of corresponding immune cells by transcriptome sequencing and single-cell transcriptome sequencing data, and performing tumor immune typing by unsupervised clustering.
The prior art mainly has the defects that tumor tissue sampling is needed, and all the technical means are invasive. The tissue section staining method and the immunofluorescence staining counting method need to firstly carry out paraffin embedding and fixing and freezing sectioning on tumor tissues, and then carry out staining and discrimination. High throughput sequencing methods, such as transcriptome sequencing and single cell transcriptome sequencing, require the provision of fresh tumor tissue blocks for sequencing experiments. These above techniques are suitable for patients undergoing surgery, as the tumor tissue removed during surgery can be used for subsequent experimental procedures. However, for patients who cannot be operated, only biopsy puncture can be performed to take tumor tissue samples, so that the risk of wound infection is increased, and the possibility of tumor metastasis is increased. More importantly, the immune subtype of the tumor changes continuously during the course of disease, but repeated collection of tumor tissue samples is clinically impossible for judging the immune subtype.
Disclosure of Invention
In view of the above, there is a need for a method, a system and a computer device for predicting an immune subtype of a tumor without acquiring tumor tissue.
In order to solve the above problems, the following technical solutions are adopted in the present application:
one of the objectives of the present application is to provide a method for predicting tumor immune subtype, comprising the following steps:
acquiring a nuclear magnetic resonance image dataset of a tumor patient;
extracting an imaging feature from the nuclear magnetic resonance image dataset;
performing feature screening on the imaging features to form an imaging feature set;
and predicting the immune subtype of the tumor according to the imaging characteristic set.
In some embodiments, the step of acquiring the magnetic resonance image dataset of the tumor patient specifically includes the following steps: collecting nuclear magnetic resonance images of tumor patients and paired transcriptome sequencing data, and establishing a data set, wherein the nuclear magnetic resonance images are from patient data sets jointly included in hospitals, TCIA and TCGA.
In some of these embodiments, wherein: the mri images include T1, T2, T1C and Flair modalities.
In some of these embodiments, the transcriptome sequencing data is processed by: firstly, removing a linker sequence from sequencing initial data, removing sequences with the number of N bases reaching 5bp and above, removing sequences with the quality value of less than or equal to 20 and more than 40%, and performing sliding window quality cutting by taking 4bp as the size of a sliding window to finally obtain initially processed transcriptome sequencing data; comparing the sequencing data of the primary processing transcriptome to a reference genome by using HISAT2 software, and recording genome annotation information according to a sequence comparison result; carrying out statistics on expression quantities of different genes by using HTSeq, and calculating an FPKM value to complete standardization; finally, a gene expression data matrix is generated.
In some embodiments, before the step of acquiring the mri data set of the tumor patient is completed, the following steps are further included: format conversion, resampling and registration of image data in the MRI image dataset.
In some embodiments, the step of extracting the imaging characteristics from the mri dataset specifically includes the following steps:
delineating a tumor region in the nuclear magnetic image under the T1C mode, and performing cross comparison with the tumor regions under the T1, T2 and Flair modes to determine that all the tumor regions are wrapped in the delineation;
and (3) re-sampling and registering the T1, T2, T1C and Flair modal nuclear magnetic resonance images, and respectively extracting the image omics characteristics from each case, wherein the image omics characteristics comprise first-order characteristics, shape characteristics and texture characteristics.
In some embodiments, the step of performing feature screening on the imaging features to form the imaging feature set specifically includes the following steps:
calculating whether the data set of the imaging characteristics conforms to normal distribution by adopting Shapiro-Wilk test;
calculating the homogeneity of the variance of the data set of the imaging characteristics by adopting an F test;
calculating statistical differences of data sets passing the two tests by using One-way ANOVA test;
the failed data set was calculated using Wilcoxon rank sum test, and a P value <0.05 was considered to satisfy significant differences, preserving the iconological features with significant statistical differences.
In some embodiments, the step of predicting an immune subtype of the tumor from the set of imaging features specifically comprises the steps of:
constructing a random forest model by using the reserved image characteristics as input data, wherein the output data of the model is a label of an immune subtype;
on a training set, giving a rough training model of a real immune subtype result of a tumor sample, and keeping the top ten image characteristics according to the contribution degree of the image characteristics to the model;
reconstructing a random forest model by taking the image characteristics of the first ten as input data;
retraining the model with the real immune subtype result of the tumor sample to realize the tumor immune subtype prediction, and testing the prediction effect of the tumor immune subtype on a verification set.
In another aspect, the present application provides a tumor immune subtype prediction system, comprising:
the data acquisition unit is used for acquiring a nuclear magnetic resonance image dataset of a tumor patient;
a feature extraction unit for extracting an imaging feature from the nuclear magnetic resonance image dataset;
the screening unit is used for carrying out feature screening on the imaging features to form an imaging feature set;
and the prediction unit is used for predicting the immune subtype of the tumor according to the imaging characteristic set.
A third object of the present application is to provide a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements any of the prediction methods when executing the computer program.
This application adopts above-mentioned technical scheme, and its beneficial effect as follows:
the tumor immune subtype prediction method, the tumor immune subtype prediction system and the computer equipment acquire a nuclear magnetic resonance image dataset of a tumor patient; extracting an imaging feature from the nuclear magnetic resonance image dataset; carrying out feature screening on the imaging features to form an imaging feature set; the immune subtype of the tumor can be predicted by adopting a non-invasive method without acquiring tumor tissues, so that the damage to a patient is reduced, the method and the system can be operated repeatedly, the dynamic monitoring of the immune subtype of the tumor in the whole disease course is realized, and compared with methods such as high-throughput sequencing and tissue staining, the technical cost is controllable, and the economic burden of the patient is small.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application or the prior art will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of the steps of the tumor immune subtype prediction method provided in example 1 of the present application.
Fig. 2 is a flowchart illustrating steps of performing feature screening on the imaging features to form an imaging feature set according to embodiment 1.
Fig. 3 is a flowchart illustrating the steps of predicting the immune subtype of a tumor according to the set of imaging characteristics provided in this example 1.
Fig. 4 is a schematic structural diagram of a tumor immune subtype prediction system provided in example 2 of the present application.
Fig. 5 is a schematic structural diagram of the computer device provided in this embodiment 3.
Detailed Description
Reference will now be made in detail to the embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are illustrative and intended to explain the present application and should not be construed as limiting the present application.
In the description of the present application, it is to be understood that the terms "upper", "lower", "horizontal", "inner", "outer", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are only for convenience in describing the present application and simplifying the description, and do not indicate or imply that the referred devices or elements must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present application.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present application, "a plurality" means two or more unless specifically limited otherwise.
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments.
Example 1
Referring to fig. 1, a flowchart of steps of a tumor immune subtype prediction method provided in embodiment 1 of the present application includes the following steps S110 to S140, and implementation of the steps is described in detail below.
Step S110: a nuclear magnetic resonance image dataset of a tumor patient is acquired.
The step of acquiring the magnetic resonance image dataset of the tumor patient specifically comprises the following steps: collecting nuclear magnetic resonance images and paired transcriptome sequencing data of tumor patients, and establishing a data set.
Further, the magnetic resonance images of The patients obtained in this example were from The patient data sets co-registered in hospitals and TCIA (The Cancer Imaging Archive) and TCGA (The Cancer Genome Atlas). The size of the data set is not limited in the embodiment, and the larger the data set is, the stronger the generalization capability is.
In this example, the data for each patient includes magnetic resonance images and transcriptome sequencing data. The nmr image should include four common modalities, i.e., T1, T2, T1C, flair modality. The transcriptome sequencing data is obtained by removing a linker sequence from sequencing initial data, removing sequences with N base number reaching 5bp or more, removing sequences with low-quality bases (quality value less than or equal to 20) exceeding 40%, performing sliding window quality cutting by using 4bp as a sliding window size, and finally obtaining primary processing data. And (3) comparing the transcriptome sequencing data to a reference genome by using HISAT2 software, and recording genome annotation information according to a sequence comparison result. Statistics of the expression levels of different genes were carried out using HTSeq, and FPKM (fragments per kilobase of exon per mile fragments mapped) values were calculated to complete normalization. Finally, a gene expression data matrix is generated.
In this example, patients were randomized into training and validation groups and differences in the clinicopathologic characteristics of the data sets were validated using the rank-sum test or chi-square test. Statistical analysis was performed using R software v3.4.0. P values <0.05 were considered to satisfy significant differences. Chi fang test is suitable for two independent sets of two-class comparison or two independent sets of two-class comparison, and for the comparison of multiple sets of multi-class data, rank sum test should be used. In this study, wilcoxon rank-sum test was used to assess the differences in age, tumor length and thickness between the training and test sets, while Chi-Square test was used for gender, tumor location, stage of patients in the training and test sets.
In this embodiment, before the step of acquiring the mri data set of the tumor patient is completed and the next step is performed, the method further includes the following steps: format conversion, resampling and registration of image data in the MRI image dataset.
It is understood that DICOM (Digital Imaging and Communications in Medicine) refers to Digital image transmission protocol in Medicine, which is a set of common standard protocols for processing, storing, printing, and transmitting medical images. The data obtained on the medical instrument is in a DICOM format, and the DICOM format is firstly converted into a NIFTI (Neuroimaging information Technology Initiative) format; then, resampling the image to improve the resolution of the image; and then registering the images, wherein points corresponding to the same spatial position in a plurality of time points are in one-to-one correspondence, a rigid registration mode is used during registration, mutual information is used as image similarity measurement, and the spatial resolution of the images after registration and resampling are both 1mm.
Step S120: and extracting the imaging characteristics from the nuclear magnetic resonance image dataset.
In this embodiment, the step of extracting the imaging characteristics from the magnetic resonance image dataset specifically includes the following steps:
step S121: and (3) delineating a tumor region in the nuclear magnetic image under the T1C mode, and performing cross comparison with the tumor regions under the T1, T2 and Flair modes to determine that all the tumor regions are wrapped in the delineation. And eliminating the image outside the three-dimensional delineation area and only keeping the image of the tumor area.
Step S122: and (3) re-sampling and registering the T1, T2, T1C and Flair modal nuclear magnetic resonance images, and respectively extracting the image omics characteristics from each case, wherein the image omics characteristics comprise first-order characteristics, shape characteristics and texture characteristics.
In this embodiment, the resampled and registered T1, T2, T1C, flair modal nmr images are used to extract 851 image omics features, including a first-order feature, a shape feature and a texture feature, from each case respectively, for a total of 3404 image omics features.
It can be understood that the quantity of the image group features of each case is not fixed in practice, but is adjusted according to the actual situation.
Step S130: and carrying out feature screening on the imaging features to form an imaging feature set.
Referring to fig. 2, a flowchart of steps for performing feature screening on the imaging features to form an imaging feature set provided in this embodiment specifically includes the following steps S131 to S134, and the implementation of each step is described in detail below.
Step S131: the data set of the imaging features was calculated to fit a normal distribution using the Shapiro-Wilk test.
Step S132: the homogeneity of variance of the dataset of the imaging features is calculated using an F-test.
Step S133: the data sets that passed both tests were statistically different using the One-way ANOVA test.
Step S134: the failed data set was calculated using Wilcoxon rank sum test, and a P value <0.05 was considered to satisfy significant differences, preserving the iconological features with significant statistical differences.
It is understood that in order to screen for a proteomic signature with predictive potency, this example uses R software v3.4.0 for statistical analysis to calculate whether these proteomic signatures are statistically different between immune subtypes.
Step S140: and predicting the immune subtype of the tumor according to the imaging feature set.
Referring to fig. 3, a flowchart of steps for predicting an immune subtype of a tumor according to the set of imaging characteristics provided in the embodiment of the present application includes the following steps S141 to S144, and the implementation of each step is described in detail below.
S141: and (4) taking the reserved image characteristics as input data, constructing a random forest model, wherein the output data of the model is a label of the immune subtype.
S142: on the training set, a coarse training model of the real immune subtype result of a given tumor sample is obtained, and the image characteristics of the first ten are reserved according to the contribution degree of the image characteristics to the model.
S143: and taking the image characteristics of the former ten images as input data to reconstruct the random forest model.
S144: retraining the model with the real immune subtype result of the tumor sample to realize the tumor immune subtype prediction, and testing the prediction effect of the tumor immune subtype on a verification set.
The above examples examine the effectiveness of the model in terms of accuracy, recall, sensitivity, specificity, F1 score, and area under the subject's operating characteristic curve.
The tumor immune subtype prediction method provided by the application adopts a non-invasive method, can predict the immune subtype of the tumor without obtaining tumor tissues, reduces the damage to patients, can be operated repeatedly, realizes dynamic monitoring of the tumor immune subtype in the whole disease course, and has controllable technical cost and smaller economic burden on the patients compared with methods such as high-throughput sequencing and tissue staining.
Example 2
Referring to fig. 4, a tumor immune subtype prediction system provided in this embodiment 2 includes: a data acquisition unit 110, configured to acquire a nuclear magnetic resonance image dataset of a tumor patient; a feature extraction unit 120 configured to extract an imaging feature from the mri dataset; a screening unit 130, configured to perform feature screening on the imaging features to form an imaging feature set; the prediction unit 140 is configured to predict an immune subtype of the tumor according to the set of imaging characteristics.
The detailed working manner of the prediction system provided in this embodiment 2 may be referred to as embodiment 1, and is not described herein again.
The tumor immune subtype prediction system provided by the application adopts a non-invasive method, can predict the immune subtype of the tumor without obtaining tumor tissues, reduces the damage to patients, can be operated repeatedly, realizes dynamic monitoring of the tumor immune subtype in the whole disease course, and has controllable technical cost and smaller economic burden on the patients compared with methods such as high-throughput sequencing and tissue staining.
Example 3
Please refer to fig. 3, which is a schematic structural diagram of a computer device according to an embodiment of the present application. The computer device 50 comprises a processor 51, a memory 52 coupled to the processor 51.
The memory 52 stores program instructions for implementing the tumor immune subtype method of the memristor accuracy reconstruction calculation described above.
The processor 51 is configured to execute the program instructions stored in the memory 52 to implement the tumor immune subtype prediction method.
The processor 51 may also be referred to as a CPU (Central Processing Unit). The processor 51 may be an integrated circuit chip having signal processing capabilities. The processor 51 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
It is to be understood that various features of the above-described embodiments may be combined in any combination, and for the sake of brevity, all possible combinations of features in the above-described embodiments may not be described in detail, but rather, all combinations of features may be considered to fall within the scope of the present disclosure unless there is a conflict between such combinations.
The foregoing is considered as illustrative only of the preferred embodiments of the invention, and is presented only for the purpose of illustrating the principles of the invention and not in any way to limit its scope. Any modifications, equivalents and improvements made within the spirit and principles of the present application and other embodiments of the present application without the exercise of inventive faculty will occur to those skilled in the art and are intended to be included within the scope of the present application.
Claims (10)
1. A method for predicting tumor immune subtype, which is characterized by comprising the following steps:
acquiring a nuclear magnetic resonance image dataset of a tumor patient;
extracting an imaging feature from the nuclear magnetic resonance image dataset;
carrying out feature screening on the imaging features to form an imaging feature set;
and predicting the immune subtype of the tumor according to the imaging characteristic set.
2. The method of claim 1, wherein the step of obtaining the magnetic resonance image dataset of the tumor patient comprises the steps of: collecting nuclear magnetic resonance images of tumor patients and paired transcriptome sequencing data, and establishing a data set, wherein the nuclear magnetic resonance images are from patient data sets jointly included in hospitals, TCIA and TCGA.
3. The method of claim 2, wherein the mri images comprise T1, T2, T1C and Flair modalities.
4. The method of tumor immune subtype prediction according to claim 2 characterized in that said transcriptome sequencing data is processed by: firstly, removing a linker sequence from sequencing initial data, removing sequences with the number of N bases reaching 5bp and above, removing sequences with the quality value of less than or equal to 20 and more than 40%, and performing sliding window quality cutting by taking 4bp as the size of a sliding window to finally obtain initially processed transcriptome sequencing data; comparing the sequencing data of the primary processing transcriptome to a reference genome by using HISAT2 software, and recording genome annotation information according to a sequence comparison result; carrying out statistics on expression quantities of different genes by using HTSeq, and calculating an FPKM value to complete standardization; finally, a gene expression data matrix is generated.
5. The method of claim 3, wherein the step of obtaining the MRI image data set of the tumor patient is further followed by the step of: format conversion, resampling and registration of image data in the MRI image dataset.
6. The method of claim 5, wherein the step of extracting the imaging characteristics from the NMR image dataset comprises the steps of:
delineating a tumor region in the nuclear magnetic image under the T1C mode, cross-comparing the tumor region with the tumor regions under the T1, T2 and Flair modes, and determining that all the tumor regions are wrapped in the delineation;
and (3) re-sampling and registering the T1, T2, T1C and Flair modal nuclear magnetic resonance images, and respectively extracting the image omics characteristics from each case, wherein the image omics characteristics comprise first-order characteristics, shape characteristics and texture characteristics.
7. The method of claim 6, wherein the step of screening the imaging characteristics to form an imaging characteristic set comprises the following steps:
calculating whether the data set of the imaging characteristics conforms to normal distribution by adopting Shapiro-Wilk test;
calculating the homogeneity of the variance of the data set of the imaging characteristics by adopting an F test;
calculating statistical differences of the data sets passing through the two tests by using One-way ANOVA test;
the failed data set was calculated using Wilcoxon rank sum test, and a P value <0.05 was considered to satisfy significant differences, preserving the iconological features with significant statistical differences.
8. The method of claim 7, wherein the step of predicting the immune subtype of the tumor according to the set of imaging characteristics comprises the steps of:
constructing a random forest model by using the reserved image characteristics as input data, wherein the output data of the model is a label of an immune subtype;
on a training set, giving a rough training model of a real immune subtype result of a tumor sample, and reserving image characteristics of the first ten according to the contribution degree of the image characteristics to the model;
taking the image characteristics of the former ten as input data, and reconstructing a random forest model;
retraining the model with the real immune subtype result of the tumor sample to realize the tumor immune subtype prediction, and testing the prediction effect of the tumor immune subtype on a verification set.
9. A tumor immune subtype prediction system, comprising:
the data acquisition unit is used for acquiring a nuclear magnetic resonance image dataset of a tumor patient;
a feature extraction unit for extracting an imaging feature from the nuclear magnetic resonance image dataset;
the screening unit is used for carrying out feature screening on the imaging features to form an imaging feature set;
and the prediction unit is used for predicting the immune subtype of the tumor according to the imaging characteristic set.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the prediction method according to any one of claims 1 to 8 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211488320.7A CN115862857A (en) | 2022-11-25 | 2022-11-25 | Tumor immune subtype prediction method, system and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211488320.7A CN115862857A (en) | 2022-11-25 | 2022-11-25 | Tumor immune subtype prediction method, system and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115862857A true CN115862857A (en) | 2023-03-28 |
Family
ID=85666314
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211488320.7A Pending CN115862857A (en) | 2022-11-25 | 2022-11-25 | Tumor immune subtype prediction method, system and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115862857A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116385441A (en) * | 2023-06-05 | 2023-07-04 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of oligodendroglioma based on MRI |
| CN116403076A (en) * | 2023-06-06 | 2023-07-07 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of GBM patient based on DTI sequence |
-
2022
- 2022-11-25 CN CN202211488320.7A patent/CN115862857A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116385441A (en) * | 2023-06-05 | 2023-07-04 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of oligodendroglioma based on MRI |
| CN116385441B (en) * | 2023-06-05 | 2023-09-05 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of oligodendroglioma based on MRI |
| CN116403076A (en) * | 2023-06-06 | 2023-07-07 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of GBM patient based on DTI sequence |
| CN116403076B (en) * | 2023-06-06 | 2023-08-22 | 中国科学院深圳先进技术研究院 | Method and system for risk stratification of GBM patient based on DTI sequence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115862857A (en) | Tumor immune subtype prediction method, system and computer equipment | |
| CN111553892B (en) | Lung nodule segmentation calculation method, device and system based on deep learning | |
| Nakashima et al. | Does breast cancer growth rate really depend on tumor subtype? Measurement of tumor doubling time using serial ultrasonography between diagnosis and surgery | |
| JP2016505289A (en) | Integrated phenotypic analysis using image texture features | |
| CN111210441A (en) | Tumor prediction method and device, cloud platform and computer-readable storage medium | |
| Pareek et al. | [Retracted] Predicting the Spread of Vessels in Initial Stage Cervical Cancer through Radiomics Strategy Based on Deep Learning Approach | |
| CN113257360B (en) | Cancer screening model, and construction method and construction device of cancer screening model | |
| Ye et al. | Prediction of placenta accreta spectrum by combining deep learning and radiomics using T2WI: a multicenter study | |
| CN117253625A (en) | Construction device of lung cancer screening model, lung cancer screening device, equipment and medium | |
| Eun et al. | Applications of artificial intelligence in urological setting: a hopeful path to improved care | |
| US11817204B2 (en) | Specialized computer-aided diagnosis and disease characterization with a multi-focal ensemble of convolutional neural networks | |
| Yang et al. | Deep learning model for the detection of prostate cancer and classification of clinically significant disease using multiparametric MRI in comparison to PI-RADs score | |
| Shu et al. | Deep Learning model-based approach for preoperative prediction of Ki67 labeling index status in a noninvasive way using magnetic resonance images: A single-center study | |
| CN113611404A (en) | Plasma sample cancer early screening method based on ensemble learning | |
| CN116563224B (en) | Image histology placenta implantation prediction method and device based on depth semantic features | |
| WO2024108536A1 (en) | Tumor immune subtype prediction method and system, and computer device | |
| CN114092427A (en) | Crohn disease and tuberculosis of intestines classification method based on many sequence MRI images | |
| CN112329876A (en) | Colorectal cancer prognosis prediction method and device based on image omics | |
| Bardwell et al. | Cognitive Impairment Prediction by Normal Cognitive Brain MRI Scans Using Deep Learning | |
| Bhagat et al. | Computational Intelligence approach to improve the Classification accuracy of Brain Tumor Detection | |
| CN117853520A (en) | Tumor new auxiliary chemotherapy response prediction method and system | |
| US20230206432A1 (en) | Method and apparatus for tumor purity based on pathaological slide image | |
| CN117636064B (en) | Intelligent neuroblastoma classification system based on pathological sections of children | |
| US12094107B2 (en) | CAD device and method for analyzing medical images | |
| CN117876288A (en) | Brain glioma microenvironment immune state assessment model based on imaging and application |
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 |