Enhance dataset card with task categories, paper/code links, and improved sample usage

README.md

@@ -1,7 +1,21 @@
 ---
 license: cc-by-nc-sa-4.0
+task_categories:
+- image-segmentation
+tags:
+- medical
+- biomedical
+- 3d
+- cvpr2025
 ---
 
+This repository contains the BiomedSegFM dataset, a crucial resource for the **CVPR 2025 Competition: Foundation Models for 3D Biomedical Image Segmentation**.
+
+The dataset is utilized by the model presented in the paper [Medal S: Spatio-Textual Prompt Model for Medical Segmentation](https://huggingface.co/papers/2511.13001). This paper introduces a medical segmentation foundation model that supports native-resolution spatial and textual prompts, achieving channel-wise alignment between volumetric prompts and text embeddings. The dataset preserves full 3D context, efficiently processes multiple native-resolution masks in parallel, and supports up to 243 classes across CT, MRI, PET, ultrasound, and microscopy modalities.
+
+**Paper:** [https://huggingface.co/papers/2511.13001](https://huggingface.co/papers/2511.13001)
+**Code:** [https://github.com/yinghemedical/Medal-S](https://github.com/yinghemedical/Medal-S)
+
 # CVPR 2025 Competition: Foundation Models for 3D Biomedical Image Segmentation
 
 **Highly recommend watching the [webinar recording](https://www.youtube.com/playlist?list=PLWPTMGguY4Kh48ov6WTkAQDfKRrgXZqlh) to learn about the task settings and baseline methods.**
@@ -21,19 +35,20 @@ Folder structure
 - CVPR25_TextSegFMData_with_class.json: text prompt for test-guided segmentation task
 
 
-
-## Interactive 3D segmentation ([Homepage](https://www.codabench.org/competitions/5263/))
+## Sample Usage (Interactive 3D Segmentation)
 
 The training `npz` files contain three keys: `imgs`, `gts`, and `spacing`.
 The validation (and testing) `npz` files don't have `gts` keys. We provide an optional box key in the `npz` file, which is defined by the middle slice 2D bounding box and the top and bottom slice (closed interval).
 Here is a demo to load the data:
 
 ```python
-npz = np.load(‘path to npz file’, allow_pickle=True)
+import numpy as np
+
+npz = np.load('path to npz file', allow_pickle=True)
 print(npz.keys())
-imgs = npz[‘imgs’]
-gts = npz[‘gts’] # will not be in the npz for testing cases
-boxes = npz[‘boxes’] # a list of bounding box prompts
+imgs = npz['imgs']
+gts = npz['gts'] # will not be in the npz for testing cases
+boxes = npz['boxes'] # a list of bounding box prompts
 print(boxes[0].keys()) # dict_keys(['z_min', 'z_max', 'z_mid', 'z_mid_x_min', 'z_mid_y_min', 'z_mid_x_max', 'z_mid_y_max'])
 ```
 
@@ -46,7 +61,7 @@ Remarks:
 3. The provided box prompts is designed for better efficiency for annotators, which may not cover the whole object. [Here](https://github.com/JunMa11/CVPR-MedSegFMCompetition/blob/main/get_boxes.py ) is the script to generate box prompts from ground truth.
 
 
-## Text-guided segmentation ([Homepage](https://www.codabench.org/competitions/5651/))
+## Sample Usage (Text-Guided Segmentation)
 
 For the training set, we provide a json file with dataset-wise prompts `CVPR25_TextSegFMData_with_class.json`.
 
@@ -57,16 +72,16 @@ In the text prompts,
 For the validation (and hidden testing) set, we provided a text key for each validation npz file
 
 ```python
-npz = np.load(‘path to npz file’, allow_pickle=True)
+import numpy as np
+
+npz = np.load('path to npz file', allow_pickle=True)
 print(npz.keys())
-imgs = npz[‘imgs’]
-print(npz[‘text_prompts’])
+imgs = npz['imgs']
+print(npz['text_prompts'])
 ```
 
 Remarks:
 
 1. To ensure rotation consistency, all testing cases will be preprocessed to standard rotation by https://nipy.org/nibabel/reference/nibabel.funcs.html#nibabel.funcs.as_closest_canonical
 2. Some datasets don't have text prompts, please simply exclude them during model training.
-3. For instance labels, the evaluate metric is [F1 score](https://github.com/JunMa11/NeurIPS-CellSeg/blob/main/baseline/compute_metric.py) where the order of instance id doesn't matter. 
-
-
+3. For instance labels, the evaluate metric is [F1 score](https://github.com/JunMa11/NeurIPS-CellSeg/blob/main/baseline/compute_metric.py) where the order of instance id doesn't matter.