CN119833138A - Hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density - Google Patents
Hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density Download PDFInfo
- Publication number
- CN119833138A CN119833138A CN202411896580.7A CN202411896580A CN119833138A CN 119833138 A CN119833138 A CN 119833138A CN 202411896580 A CN202411896580 A CN 202411896580A CN 119833138 A CN119833138 A CN 119833138A
- Authority
- CN
- China
- Prior art keywords
- density
- area
- critical value
- cell
- tumor
- 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.)
- Granted
Links
Landscapes
- Investigating Or Analysing Biological Materials (AREA)
Abstract
The invention discloses a hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density, wherein the method comprises the steps of obtaining full-slide imaging, carrying out image preprocessing, determining a critical value of a cell pixel level division result, carrying out cell pixel density calculation according to the critical value of the cell pixel level division result, determining CD8+TILs density of a tumor area, carrying out standardization and average calculation treatment on the CD8+TILs density of the tumor area, obtaining a critical value of ATLS-8, and carrying out prognosis evaluation on hepatocellular carcinoma of a patient according to the critical value of ATLS-8. By using the method, the cell density on the whole glass slide can be accurately quantified, so that the accuracy of cell density measurement is improved, and the accuracy of prognosis evaluation of the hepatocellular carcinoma of a patient is improved. The method and the system for prognosis evaluation of the hepatocellular carcinoma based on the cell pixel density can be widely applied to the technical field of prognosis evaluation of the hepatocellular carcinoma.
Description
Technical Field
The invention relates to the technical field of prognosis evaluation of hepatocellular carcinoma, in particular to a method and a system for prognosis evaluation of hepatocellular carcinoma based on cell pixel density.
Background
For the complexity of tumor heterogeneity, related art has focused on determining new prognostic parameters and indicators as comprehensive biomarkers for a wider range of HCC patients. Currently, a variety of biomarkers have been discovered and used for biological characterization and prognosis evaluation of tumors. Tumor Microenvironment (TME) is composed of tumor cells, stromal cells, immune cells and epithelial tissue and is an important component affecting tumor survival and progression. Tumor heterogeneity plays a critical role in limiting the effectiveness of widely accepted drugs and radiation therapies in cancer patients, closely related to TME. TILs are widely recognized as immune cells that are correlated with prognosis of patients with solid tumors. This includes subtypes of CD3, CD8, and CD45 lymphocytes, where cd8+ T cells have been shown to be key members in the destruction of tumor cells.The effectiveness and versatility of tumor cells in various cancer types have been verified using the proportion of T lymphocytes at the tumor core and infiltrating margin as the main evaluation index. Various TIL-centric scoring methods have been widely studied and used for prognostic evaluation of various tumor types. However, there is no wide consensus on liver cancer. In previous studies, some pathologists reported artificial scoring of immune responses of tissues and tumor areas using traditional pathology methods to assess prognosis. These methods have limitations and subjectivity, resulting in contradictory results from different studies. Due to the large number of cells in WSI, traditional pathology quantification methods have little accuracy in quantifying cell density on whole slides. Conventional methods often have bias in selecting a region of interest (ROI) and rely heavily on the subjective expertise of pathologists, which can affect the accuracy of cell density measurements, limiting them to qualitative or semi-quantitative assessments.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density, which can accurately quantify the cell density on a whole glass slide, further improve the accuracy of cell density measurement and improve the accuracy of prognosis evaluation of hepatocellular carcinoma of a patient.
The first technical scheme adopted by the invention is that the hepatocellular carcinoma prognosis evaluation method based on cell pixel density comprises the following steps:
acquiring full slide imaging, performing image preprocessing, and determining a critical value of a cell pixel fraction separation result;
calculating cell pixel density according to the critical value of the cell pixel level segmentation result, and determining the CD8+TILs density of the tumor area;
Carrying out standardization and average calculation treatment on the CD8+TILs density of the tumor area to obtain a critical value of ATLS-8;
Prognosis evaluation is performed on the hepatocellular carcinoma of the patient according to the critical value of ATLS-8.
Further, the step of acquiring whole glass imaging and performing image preprocessing to determine a critical value of a cell pixel fraction, specifically includes:
performing image preprocessing on the whole glass slide imaging to obtain a tumor region mask image;
Positive cell identification and segmentation are carried out on the whole glass imaging and the tumor region mask image, and the critical value of the cell pixel fraction result is determined.
Further, the step of performing image preprocessing on the whole glass slide imaging to obtain a mask image of the tumor region specifically comprises the following steps:
acquiring full slide imaging and performing dyeing treatment to obtain an H & E dyed slice image;
extracting non-overlapping plaques and removing the plaques smaller than a preset threshold value under the magnification of 10 times of the H & E staining slice images based on a quality control instrument to obtain images to be segmented of the tumor area;
and (3) segmenting the image to be segmented of the tumor region through nnU-Net deep neural network to obtain a mask image of the tumor region.
Further, the step of positive cell identification and segmentation of the whole slide imaging and the tumor region mask image and determining the critical value of the cell pixel level segmentation result specifically comprises the following steps:
Registering the full slide imaging with the same tissue staining and different staining according to the H & E staining slice images to obtain aligned full slide imaging;
The aligned full slide imaging is imported to a Qupath platform for observation, and a tiled image is exported;
And registering and pixel segmentation processing is carried out on the tiled image and the tumor area mask image through a random forest algorithm, and a critical value of a cell pixel fraction result is determined.
Further, the step of calculating the cell pixel density according to the critical value of the cell pixel level segmentation result to determine the cd8+tils density of the tumor region specifically includes:
determining the area of a tumor area and the area of CD8+ pixels according to the critical value of the cell pixel level segmentation result;
Determining the area proportion of CD8+ cells in tumor tissues according to the area of the tumor area and the area of the CD8+ pixels;
the area ratio of CD8+ cells in tumor tissue is normalized in units, and the pixel density is converted into the cell number in unit area, so as to obtain the CD8+ TILs density of the tumor area.
Further, the step of normalizing and averaging the cd8+tils density of the tumor area to obtain a critical value of ats-8 specifically includes:
Carrying out standardization treatment on CD8+TILs density of the tumor area through a normalization factor to obtain the ratio of the normalized patch area to the unit area;
obtaining non-normalized cell density and carrying out product calculation on the non-normalized cell density and the ratio of the normalized patch area to the unit area to obtain the cell density of the unit area of the patch;
and carrying out average calculation treatment on the cell density of the unit area of the patch to obtain the critical value of ATLS-8.
The second technical scheme adopted by the invention is that the hepatocellular carcinoma prognosis evaluation system based on cell pixel density comprises:
The first module is used for acquiring full slide imaging and carrying out image preprocessing, and determining a critical value of a cell pixel fraction separation result;
the second module is used for calculating the cell pixel density according to the critical value of the cell pixel fraction result and determining the CD8+TILs density of the tumor area;
the third module is used for carrying out standardization and average calculation treatment on the CD8+TILs density of the tumor area to obtain a critical value of ATLS-8;
and a fourth module for prognosis evaluation of hepatocellular carcinoma in patients according to the threshold value of ATLS-8.
The method and the system have the beneficial effects that the critical value of the cell pixel level segmentation result is determined, so that the CD8+TILs density of the tumor area is determined, the CD8+TILs density of the tumor area is subjected to standardization and average calculation treatment to obtain the critical value of ATLS-8, and finally, the prognosis evaluation of the hepatocellular carcinoma of a patient is carried out according to the critical value of ATLS-8, so that the cell density on the whole glass slide can be accurately quantified, the accuracy of cell density measurement is further improved, and the accuracy of the prognosis evaluation of the hepatocellular carcinoma of the patient is improved.
Drawings
FIG. 1 is a flow chart of steps of a method for prognosis evaluation of hepatocellular carcinoma based on pixel density of cells of the present invention;
FIG. 2 is a block diagram of a cellular pixel density-based prognosis evaluation system for hepatocellular carcinoma of the present invention;
FIG. 3 is a schematic diagram of the results of Bland Altman consistency correlation analysis of 40 plaques according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a discovery queue Kaplan-Meier survival curve analysis provided by an embodiment of the invention;
FIG. 5 is a schematic diagram of a verification queue 1Kaplan-Meier survival curve analysis provided in an embodiment of the present invention;
FIG. 6 is a schematic diagram of a verification queue 2Kaplan-Meier survival curve analysis provided by an embodiment of the present invention.
Detailed Description
The invention will now be described in further detail with reference to the drawings and to specific examples. The step numbers in the following embodiments are set for convenience of illustration only, and the order between the steps is not limited in any way, and the execution order of the steps in the embodiments may be adaptively adjusted according to the understanding of those skilled in the art.
The invention introduces a new full-automatic process, and utilizes an artificial intelligence algorithm to quantify CD8 tumor infiltrating lymphocytes, namely ATLS. In addition, it also evaluates the prognosis of HCC patients using a cd8+ lymphocyte-based scoring system (called ats-8).
Referring to fig. 1, the present invention provides a hepatocellular carcinoma prognosis evaluation method based on cell pixel density, the method comprising the steps of:
s100, acquiring full slide imaging, performing image preprocessing, and determining a critical value of a cell pixel fraction result;
S110, performing image preprocessing on the whole glass slide imaging to obtain a mask image of a tumor area;
The method comprises the steps of obtaining full-slide imaging, performing dyeing treatment to obtain an H & E dyed slice image, extracting non-overlapping plaques and removing plaques smaller than a preset threshold value under the 10-time magnification of the H & E dyed slice image based on a quality control instrument to obtain an image to be segmented of a tumor area, and performing segmentation treatment to the image to be segmented of the tumor area through a nnU-Net deep neural network to obtain a mask image of the tumor area.
In this embodiment, to remove unwanted white background, artifacts, and other types of noise in the WSI, a quality control instrument is employed by a expert pathologist to identify the ROI available in the WSI. To obtain more accurate tumor segmentation results and higher efficiency, non-overlapping plaques (1000 x 1000 pixels) were extracted at 10-fold magnification. Plaques with a tissue proportion of less than 0.05 will be discarded. Next, the tumor region was segmented using pre-trained nnU-Net. Finally, pathologists should invite to assist in refining tumor masks obtained by aggregating Patch-level masks.
S120, positive cell identification and segmentation are carried out on the whole glass imaging and the tumor region mask image, and a critical value of a cell pixel fraction result is determined.
The method comprises the steps of registering full-slide imaging with different tissue staining according to H & E staining slice images to obtain aligned full-slide imaging, guiding the aligned full-slide imaging to a Qupath platform for observation to obtain a tiled image, registering and pixel segmentation of the tiled image and a mask image of a tumor area through a random forest algorithm, and determining a critical value of a cell pixel fraction result.
In this embodiment, to align adjacent tissue WSI stained with H & E and IHC, a method of combining rigid registration and two non-rigid registration is employed. This approach ensures that the two slices are as similar in spatial position and morphology as possible. In this step, all WSI images of the same tissue but differently stained are registered from H & E stained images as reference for multiple IHC staining registration. The aligned images are then imported to a Qupath platform to observe the registration effect. A 1024x1024 pixel tile is generated and derived. Subsequently, the mask is registered with the calibrated IHC slices using a Python procedure. The tumor and non-tumor regions were segmented into 1024x1024 pixels and blocks with more than 50% of the tissue in the mask were available for further analysis.
And further adopting a random forest algorithm model to divide cells of the tiles. Initially, a pathologist makes a small number of accurate annotations to observe the segmentation results. Based on these observations, the model is then trained on a larger scale. The probability generated for each pixel represents the likelihood that pixel will become the target cell pixel. The final threshold for the cell pixel level segmentation result was determined to be p=0.25.
S200, calculating cell pixel density according to a critical value of a cell pixel level segmentation result, and determining CD8+TILs density of a tumor area;
S210, determining the area of a tumor area and the area of a CD8+ pixel according to the critical value of the cell pixel level segmentation result;
S220, determining the area proportion of CD8+ cells in tumor tissues according to the area of the tumor area and the area of CD8+ pixels;
In this example, after cell division is completed, the flat sheet is divided into tumor regions according to a predetermined strategy. The pixel density for each segment is calculated by determining the area occupied by the pixels labeled cd8+ in the entire segment. This measurement provides a WSI-grade cd8+ cell pixel density, representing the area fraction of cd8+ cells in tumor tissue, expressed as:
Wherein Pixel CD8label counting patch refers to dividing pixels at the plaque level using a random forest algorithm. Thus Pixel total label countiong patch encompasses all pixels within the plaque, including cd8+ cells (1) and non-cd8+ cells (0). By this method, cd8+ cell density in the tumor region can be calculated at WSI level. The calculation formula is summarized as follows:
where n represents the number of plaques in the available tumor area.
And S230, carrying out unit standardization on the area proportion of CD8+ cells in the tumor tissue, and converting the pixel density into the cell number of unit area to obtain the CD8+ TILs density of the tumor area.
In this example, the pixel density was converted to the number of cells per unit area for standardization with the units used in most studies. This involves measuring the number of cells in each sample from the cell segmentation marker image. The cell number/mm 2 was determined by calculating the area fraction and pixel size of each cell. This gives the CD8+ TILs density of the tumor area. The calculation formula is as follows:
In the above formula PSTCH CELL DENSITY unnormalised represents the cd8+ TILs density of the tumor region.
S300, carrying out standardization and average calculation treatment on the CD8+TILs density of the tumor area to obtain a critical value of ATLS-8;
s310, carrying out standardization treatment on CD8+TILs density of a tumor area through a normalization factor to obtain the ratio of a normalized patch area to a unit area;
s320, obtaining the non-normalized cell density, and performing product calculation on the non-normalized cell density and the ratio of the normalized patch area to the unit area to obtain the cell density of the unit area of the patch;
S330, carrying out average calculation processing on the cell density of the unit area of the patch to obtain the critical value of ATLS-8.
In this example, after calculating the cell density of each plaque, we normalized the pixel density of the three centers using normalization factors to reflect the cell size. The calculation formula is as follows
Where F mag represents the actual area of each patch and F nor represents the ratio of normalized patch area to unit area (unit: mm 2). In this experiment, the Patch height height and the Patch weight weight were 1024 pixels. Then, F nor was multiplied by the non-normalized cell density to obtain the cell density per patch unit area (mm 2). The calculation formula is as follows
Cell densitypatch-normalised＝Cell densitypatch-unnormalised*Fnor
Finally, the cd8+ cell densities of all plaques were averaged to reflect the cd8+ cell density in the tumor area on WSI. This is known as ATLS-8. The calculation formula is as follows
Where n represents the number of plaques in the available tumor area.
S400, performing prognosis evaluation on the hepatocellular carcinoma of the patient according to the critical value of ATLS-8.
In summary, embodiments of the present invention obtain a mask of a tumor region labeled by a expert pathologist by scanning the H & E stained sections using a full slide scanner, applying quality control measures, and segmenting the image using a deep learning method. Registering adjacent CD8 IHC stained sections with the H & E image, dividing the registered images into blocks, identifying and dividing annotated cells using a machine learning algorithm, calculating and converting pixel densities of positive cells to cell densities, determining acls-8 threshold values based on maximum selection rank statistics, dividing patients into low index groups and high index groups, and drawing Kaplan-Meier (KM) curves to visually display survival differences between the groups, as shown in fig. 4. The prognostic value of ATLS-8 was evaluated in two independent external validation groups.
Finally, the cell segmentation results of the embodiments of the present invention were analyzed by simulation experiments, and as shown in fig. 3, 40 plaques were randomly selected in total for cell segmentation. The segmentation results were compared and analyzed with manual counts by pathologists using conventional pathology counting methods, which were not visible to the pathologist as the results generated by the algorithm. The comparison results were analyzed by Bland-Altman method. The agreement between manual and automatic counting using our algorithm is good, with ICC 0.98 (95% CI,0.97-0.99; P < 0.001).
And further analyzing the prognostic value of the ATLS-8 analysis, wherein the critical value of the ATLS-8 is determined by adopting a maximum selection rank statistical method, and the critical value of the tumor area is 67.25 cells/mm 2. In the discovery cohort, 90 patients (72.58%) had an ats-8 value above the threshold, and were classified as low risk, and 34 patients (27.42%) had an ats-8 value below the threshold, and were classified as high risk. The 5 year Overall Survival (OS) was 65.03% for the low risk group and 34.05% for the high risk group. As shown in fig. 5, in verification queue 1, 74 patients (90.24%) had an ats-8 value above the threshold, and 8 patients (9.76%) had an ats-8 value below the threshold, and were classified as high risk. The 5 year OS rate was 68.73% for the low risk group and 28.57% for the high risk group. As shown in fig. 6, in verification queue 2, 27 patients (50.0%) had an ats-8 value above the threshold, and were classified as low risk, and 27 patients (50.0%) had an ats-8 value below the threshold, and were classified as high risk. The 5 year OS rate was 81.48% for the low risk group and 59.26% for the high risk group.
KM curves show that patients with lower ats-8 scores had worse OS than patients with higher scores (found cohorts, unadjusted HR,2.23 (95% ci, 1.27-3.91); p=0.0042; validated cohorts 1, HR 3.38 (95% ci, 1.27-9.02) p=0.0096; validated cohorts 2, HR 2.74 (95% ci, 1.05-7.15); p=0.031).
Table 1 univariate Cox regression analysis results data sheet for three queues
Table 2 multivariate Cox regression analysis results data sheet for three queues
Table 1 lists the univariate Cox regression analysis results for the three queues and table 2 lists the multivariate Cox regression analysis results for the three queues. Factors that reached statistical significance (P < 0.05) in the univariate analysis of the discovery cohort, including age, HBV, tumor size, grade of differentiation, MVI, BCLC staging, and ats-8, were included in the multivariate analysis to evaluate robustness of ats-8 in the validation cohort. Univariate and multivariate model analysis precluded data for which MVI status is unknown. Multivariate analysis showed that after adjusting other clinical pathology factors, ats-8 score was independent of OS (found cohorts, low HR and high HR,2.015 (95% ci: 1.115-3.641), p=0.0203, validation cohorts 1, HR,4.871 (95% ci: 1.753-13.534), p=0.0024, validation cohorts 2,3.925 ((95% ci:1.212-12.714; p= 0.0226)) finally developed and validated a prognostic prediction model, in which age, HBV, ats-8, BCLC stage, tumor size, differentiation stage and MVI were considered independent factors affecting OS, so we established a complete prognostic prediction model (age+hbv+ats-8+bclc stage+tumor size+differentiation stage+mvi) based on the above factors (so we refer to this model as complete model.) we further compared the performance of complete model and other four models, including age, ats-8, BCLC stage and 8, and a combination of stage, HBV stage and atlc model, and a clinical differentiation stage and a pathological stage model (BCLC-8, a combination of stage, a clinical differentiation stage and a model, a clinical stage and a pathological stage model) as shown in table 3.
TABLE 3 Performance index of ATLS-8 model and Multi-reference model
In all three queues, the discrimination ability (estimated in C-index) and the calibration ability (estimated in AIC) of the complete model outperformed the clinical pathology model. In particular the number of the elements,
In the discovery cohort, the C index of the full model was 0.770, while the clinical pathology model was 0.757, aic was 381.6, and the clinical pathology model was 384.8. In validation queue 1, the C index is 0.769vs 0.727,AIC at 285.8vs 291.3. In validation set 2, the C index was 0.712vs 0.642,AIC to 152.7vs 156.6. The BCLC classification and ats-8 model also showed better discrimination and calibration capabilities than the BCLC classification and ats-8 models.
The BCLC classification and acls-8 models also showed better discrimination and calibration capabilities than the BCLC classification model in all three groups. In the discovery queue, the C index is 0.715vs 0.655,AIC to 389.4vs 392.8. In validation queue 1, the C index is 0.619vs 0.674,AIC is 292.6vs 296. In validation queue 2, the C index is 0.643vs 0.516,AIC to 146.2vs 160.
Incorporating the ats-8 score into the BCLC staging model can increase the predictive OS (likelihood ratio p=0.0013). Furthermore, inclusion of the ats-8 score into the clinical pathology model also improved the predictive ability of OS (likelihood ratio p=0.0011).
Referring to fig. 2, a hepatocellular carcinoma prognosis evaluation system based on cell pixel density, comprising:
The first module is used for acquiring full slide imaging and carrying out image preprocessing, and determining a critical value of a cell pixel fraction separation result;
the second module is used for calculating the cell pixel density according to the critical value of the cell pixel fraction result and determining the CD8+TILs density of the tumor area;
the third module is used for carrying out standardization and average calculation treatment on the CD8+TILs density of the tumor area to obtain a critical value of ATLS-8;
and a fourth module for prognosis evaluation of hepatocellular carcinoma in patients according to the threshold value of ATLS-8.
The content in the method embodiment is applicable to the system embodiment, the functions specifically realized by the system embodiment are the same as those of the method embodiment, and the achieved beneficial effects are the same as those of the method embodiment.
While the preferred embodiment of the present application has been described in detail, the application is not limited to the embodiment, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the application, and these equivalent modifications and substitutions are intended to be included in the scope of the present application as defined in the appended claims.
Claims (7)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411896580.7A CN119833138B (en) | 2024-12-23 | 2024-12-23 | Hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202411896580.7A CN119833138B (en) | 2024-12-23 | 2024-12-23 | Hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN119833138A true CN119833138A (en) | 2025-04-15 |
| CN119833138B CN119833138B (en) | 2025-11-07 |
Family
ID=95297058
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202411896580.7A Active CN119833138B (en) | 2024-12-23 | 2024-12-23 | Hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN119833138B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4856528A (en) * | 1987-06-26 | 1989-08-15 | John Hopkins University | Tumor volume determination |
| US20170270346A1 (en) * | 2014-09-03 | 2017-09-21 | Ventana Medical Systems, Inc. | Systems and methods for generating fields of view |
| US20200234442A1 (en) * | 2017-07-24 | 2020-07-23 | Ventana Medical Systems, Inc. | Methods and systems for evaluation of immune cell infiltrate in tumor samples |
| CN112011616A (en) * | 2020-09-02 | 2020-12-01 | 复旦大学附属中山医院 | Immune gene prognosis model for predicting hepatocellular carcinoma tumor immune infiltration and postoperative survival time |
| CN113930506A (en) * | 2021-09-23 | 2022-01-14 | 江苏大学附属医院 | Glutamine metabolism gene label scoring system for predicting hepatocellular carcinoma prognosis and treatment resistance |
| CN116258676A (en) * | 2022-12-30 | 2023-06-13 | 广州市第一人民医院（广州消化疾病中心、广州医科大学附属市一人民医院、华南理工大学附属第二医院） | Cell density quantification method and system for colorectal cancer IHC pathological image |
| CN116265947A (en) * | 2021-12-17 | 2023-06-20 | 许大康 | Immune biomarker combination based on CD8/FOXP3 and PD-1 and PD-L1 and application thereof |
-
2024
- 2024-12-23 CN CN202411896580.7A patent/CN119833138B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4856528A (en) * | 1987-06-26 | 1989-08-15 | John Hopkins University | Tumor volume determination |
| US20170270346A1 (en) * | 2014-09-03 | 2017-09-21 | Ventana Medical Systems, Inc. | Systems and methods for generating fields of view |
| US20200234442A1 (en) * | 2017-07-24 | 2020-07-23 | Ventana Medical Systems, Inc. | Methods and systems for evaluation of immune cell infiltrate in tumor samples |
| CN112011616A (en) * | 2020-09-02 | 2020-12-01 | 复旦大学附属中山医院 | Immune gene prognosis model for predicting hepatocellular carcinoma tumor immune infiltration and postoperative survival time |
| CN113930506A (en) * | 2021-09-23 | 2022-01-14 | 江苏大学附属医院 | Glutamine metabolism gene label scoring system for predicting hepatocellular carcinoma prognosis and treatment resistance |
| CN116265947A (en) * | 2021-12-17 | 2023-06-20 | 许大康 | Immune biomarker combination based on CD8/FOXP3 and PD-1 and PD-L1 and application thereof |
| CN116258676A (en) * | 2022-12-30 | 2023-06-13 | 广州市第一人民医院（广州消化疾病中心、广州医科大学附属市一人民医院、华南理工大学附属第二医院） | Cell density quantification method and system for colorectal cancer IHC pathological image |
Non-Patent Citations (1)
| Title |
|---|
| 何天基;葛波;: "肿瘤浸润淋巴细胞在泌尿系肿瘤中的研究进展", 中国免疫学杂志, no. 14, 27 July 2020 (2020-07-27), pages 1779 - 1783 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN119833138B (en) | 2025-11-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11842556B2 (en) | Image analysis method, apparatus, program, and learned deep learning algorithm | |
| CN113591919B (en) | AI-based analysis method and system for postoperative recurrence and prognosis of early hepatocellular carcinoma | |
| EP3977397B1 (en) | Quantitative phenotyping of fibrosis from images of biological tissue samples | |
| Mungle et al. | Automated characterization and counting of Ki-67 protein for breast cancer prognosis: A quantitative immunohistochemistry approach | |
| Xu et al. | Using transfer learning on whole slide images to predict tumor mutational burden in bladder cancer patients | |
| CN107567631B (en) | Tissue sample analysis techniques | |
| Smit et al. | Deep learning based tumor–stroma ratio scoring in colon cancer correlates with microscopic assessment | |
| CN112183557A (en) | MSI prediction model construction method based on gastric cancer histopathology image texture features | |
| CN115937130A (en) | Image processing method for predicting ovarian cancer Ki-67 expression based on dual-energy CT | |
| Chuang et al. | Deep Learning–Based Nuclear Morphometry Reveals an Independent Prognostic Factor in Mantle Cell Lymphoma | |
| CN119833138B (en) | Hepatocellular carcinoma prognosis evaluation method and system based on cell pixel density | |
| Fernandez et al. | Analytical validation of the PreciseDx digital prognostic breast cancer test in early-stage breast cancer | |
| KR20230120092A (en) | Method for predicting cancer prognosis and a model thereof | |
| CN113723441B (en) | Intelligent analysis system and method for lip gland pathology | |
| CN119205629A (en) | AI-based CBA fluorescence immunoassay method, system and storage medium | |
| CN115019882B (en) | Construction method and evaluation system of tumor internal heterogeneity evaluation model and its application | |
| US9122904B2 (en) | Method for optimization of quantitative video-microscopy and associated system | |
| CN118247548A (en) | Training method for colorectal cancer microsatellite status detection model | |
| JP2025533848A (en) | Prediction of biomarker expression levels in digital pathology images | |
| CN116188407A (en) | Prostate digital pathology image Gleason grading system and method | |
| Zhang et al. | Microscope based her2 scoring system | |
| CN116188406A (en) | HER2 immunochemistry breast cancer interpretation method and system | |
| CN117576127B (en) | Liver cancer area automatic sketching method based on pathological image | |
| CN111553369A (en) | Gleason automatic grading method and device for prostate cancer biopsy pathological images | |
| CN116309497B (en) | Auxiliary analysis method for cancer cell counting and prognosis prediction based on image recognition |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |