Toward real world medical image analysis

Motivation

Medical image analysis empowered by AI has significantly reshaped medicine. However, real-world medical image analysis poses significant challenges for medical AI, including domain shifts due to diverse imaging protocols and medical centers, variability in pathological regions of interest (ROIs) such as lesions and scars, and limited generalization capabilities of existing deep learning models. This challenge aims to encourage the development of novel and effective methods that facilitate the translation of medical AI into real-world clinical practice.

Tracks

This challenge consists of four challenging tracks and encompasses imaging data from over 1300 patients collected across three continents, capturing diverse acquisition protocols, clinical centers, and patient populations. Participants can explore each track below to access the corresponding datasets and detailed task descriptions.

CARE-Cardiac: Unified Cardiac Structure and Pathology Segmentation

CARE-Myops: Myocardial Pathology Segmentation

CARE-WHS: Whole Heart Segmentation

CARE-Liver: Liver Segmentation and Fibrosis Staging

Submission

After registration, we will assign participants an account to login into our evaluation platform. Participants can directly upload your predictions on the validation and test data via the website.

Paper Submission

Please refer to our paper submission guidance.

Timeline

The schedule for this track is as follows. All deadlines(DDLs) are on 23:59 in Pacific Standard Time.

Training Data Release	May 8, 2025
Validation Phase start	June 20, 2025
Test data release	July 7, 2025
Abstract Submission	July 15, 2025
Paper Submission	August 15, 2025
Release date of the results	September 5, 2025
Camera Ready	September 10, 2025
Workshop (Half-Day)	September 23-27, 2025

Contact

If you have any questions, please feel free to contact: care25challenge@163.com or care25challenge@outlook.com.