# Glioma-MDC 2025 (ISBI) — Data

来源：https://www.kaggle.com/competitions/glioma-mcd-2025/data

采集时间：2026-10-19T08:36:30Z

The target cohort for the Glioma-MDC 2025 challenge consists of patients diagnosed with glioma, specifically those who have undergone surgical resection or biopsy. The data used in the final biomedical application would be acquired from human glioma tissue samples, typically stained with Hematoxylin and Eosin (H&E) for histopathological examination.

The challenge cohort for the Glioma-MDC 2025 challenge consists of histopathological image patches obtained from glioma tissue samples. These samples were sourced from patients treated at Beijing Tiantan Hospital,  Sanbo Brain Hospital and supplemented with data derived from The Cancer Genome Atlas (TCGA) database.

**All of the data provided for the Glioma-MDC 2025 challenge is licensed under CC BY-NC-SA (Attribution-NonCommercial-ShareAlike)**.

Participants should read the data description before starting:

## Files

*   **Glioma_MDC_2025_training.zip** - compressed zip file contains following two types of training data, 1,439 annotated images in total
*   *training[4-digit-ID].jpg* - the training HE slide image with`512` by `512` pixels
*   *training[4-digit-ID].json* - the training annotation with HE slide ID correlated to image
*   **Glioma_MDC_2025_testing.zip** - compressed zip file contains following two types of testing data, 200 annotated images in total
*   *testing[4-digit-ID].jpg* - the testing HE slide image with`512` by `512` pixels
*   *testing[4-digit-ID].json* - the testing annotation with HE slide ID correlated to image
*   **Submission_sample.csv** - a sample submission file in the correct format

## Columns of `Submission_sample.csv`

*   `Image ID` - Image ID for the HE slide
*   `Label ID` - Location ID for the HE slide
*   `Prediction` - Prediction of labels in specific location of an HE slide. If the cell is identified as mitosis, a 1 is filled in the corresponding position in the result table. If it is "Non-mitosis", a 0 is filled in the corresponding position.

## Important notes

* Each image may have multiple ROIs (Region of Interests), and the number are not always the same. We have annotated all of these ROIs so you don't have to, the task of this challenge is to identify the label of every ROIs: `Mitosis` or `Non-mitosis`
* The ROI is circled by polygon shapes. For example in the `training0001.json` file you will see:
```
"shapes": [
        {
            "label": "Non-mitosis",
            "points": [
                [
                    262.3829787234043,
                    187.56028368794324
                ],
                [
                    269.4751773049645,
                    187.56028368794324
                ],
                ...
            ],
            "group_id": null,
            "description": "",
            "shape_type": "polygon",
            "flags": {},
            "mask": null
        }
    ]
```
This location is labeled as `Non-mitosis` circled by the points in x-axis and y-axis coordinates. To view the annotation on top of the image, you can download open-source software [*labelme*](https://github.com/wkentaro/labelme). The following is an example view of `training0001.jpg` with annotation `training0001.json`. ![](assets/e8b49260d6ff72012d70dcee.jpg)
* The label in the annotation files of the **testing** set is `Blank[position-ID]`, your task is to predict the label. For example in the `testing0001.json` file you will see:
```
 "shapes": [
        {
            "label": "Blank1",
            "points": [
                [
                    303.23756906077347,
                    229.75690607734808
                ],
              ...
            ],
            "group_id": null,
            "description": "",
            "shape_type": "polygon",
            "flags": {},
            "mask": null
        },
        {
            "label": "Blank2",
            "points": [
                [
                    164.0110497237569,
                    243.56906077348066
                ],
         ...
            ],
            "group_id": null,
            "description": "",
            "shape_type": "polygon",
            "flags": {},
            "mask": null
        },
        ...
    ],
```
You will see there will be `Blank1-6` in the JSON file, which means there are **6** ROIs to be predict in total. If the cell is identified as mitosis, a 1 is filled in the corresponding position in the result table. If it is "Non-mitosis", a 0 is filled in the corresponding position. **Make sure filling the prediction accordingly in the `Submission_template.csv` file**.