Method paper
LNCS format, no more than 4 pages, including results.
Machine Learning/CORE Unranked
TREAT-MMTB 2026
Transformative Research and Efficient AI Technologies for Multimodal Management of Tuberculosis 2026
TREAT-MMTB 2026 is a MICCAI challenge focused on transformative AI technologies for multimodal management of tuberculosis, with two tasks: cavity detection and segmentation (Task 1), and TB diagnosis (Task 2). Participants submit Dockerized models for evaluation on internal and external datasets, with opportunities for journal co-authorship and oral presentation at MICCAI 2026.
Paper fit
Contribution paths
A strong submission should clearly identify its contribution and evaluate it appropriately.
Research areas in scope
01
Challenge Tasks
Cavity Detection and SegmentationTB Diagnosis
Policies worth checking twice
- Submissions must include both a Dockerfile and a built Docker image with all dependencies and model weights included.
- Docker images must run without network access.
- Only the most recent submission before the deadline will be counted.
- Internal validation submissions are due by July 24, 2026, 23:59 KST.
- Submissions must use the provided evaluation scripts (evaluate_task1.py / evaluate_task2.py).
- Ground-truth labels for internal phase are provided as segmentation masks + CSV for Task 1 and CSV for Task 2.
Official sources
Compiled from the official call for papers. The organizers’ pages remain authoritative.
Last verified September 9, 2026