MM-WHS: Multi-Modality Whole Heart Segmentation

* WHS of other datasets or Challenge data: NEW!!! (see blow)

Accurate computing, modeling and analysis of the whole heart substructures is important in the development of clinical applications. Segmentation and registration of whole heart images is however challenging and still relies heavily on manual work, which is time-consuming and prone to errors.

The Challenge provides 120 multi-modality cardiac images acquired in real clinical environment. It aims at creating an open and fair competition for various research groups to test and validate their methods, particularly for the multi-modality whole heart segmentation[1]. It is not only to benchmark various whole heart segmentation algorithms, but also to cover the topic of general cardiac image segmentation and registration and modeling.

Data and Evaluation:

The purpose of disseminating the Data is to perform a multi-institutional analysis of a database of anonymized clinical MRI and CT scans for whole heart segmentation. This analysis is taking place in the context of the Multi-Modality Whole Heart Segmentation Challenge 2017. The Recipient(s) commit to not disseminate the Data to any third party. For details of the data, please refer to here.

For participants who want to download and use the data, they need to agree with the conditions above and the terms in the registration form (please sign the form and send to the organizers.)
Agree and download from FDU server or Agree and download from mega link.
* Users can evaluate their results by themselves now, link
* Tools for MM-atlas WHS [1]: Atlas-based WHS (matlab: zxhwhs.m) ; Multi-atlas WHS (zxhLabelFuse for label fusion) ; ZXHPROJ ;

You are welcomed to use the data or results for your publications. Please cite the two references below when using them:
[1] Xiahai Zhuang and Juan Shen: Multi-scale patch and multi-modality atlases for whole heart segmentation of MRI, Medical Image Analysis 31: 77-87, 2016 (download)
[2] Xiahai Zhuang: Challenges and Methodologies of Fully Automatic Whole Heart Segmentation: A Review. Journal of Healthcare Engineering 4 (3): 371-407, 2013 (download)

* WHS of other datasets or Challenge data: NEW!!!
(1) Twenty MR images with manual segmentation [1], a sub set of MM-WHS challenge test set.
(2) Left Atrium Segmentation Challenge 13: WHS of 30 MR and 30 CT using [1] (challenge link).
(3) Coronary Centerline Extraction Challenge 08: WHS of 32 CTA using [1], (original challenge-invalid now, challenge paper).
(4) Coronary Artery Stenosis Detection Challenge 13: WHS of 48 CTA using [1], (original challenge-invalid now, challenge paper).

* Pre-processed training data registered to a common space (2x2x2 mm), e.g. for multi-atlas segmentation [1] or statistical shape model studies [2]: affregcommon2mm_roi_mr_train or affregcommon2mm_roi_ct_train.
* All cardiac cross-modality domain adaptation works from zmic link
(1) Wu,FP's medical cross modality domain adaptation work & Data link1 link2-2DWHS
(2) Dou, Q et al.'s medical cross modality domain adaptation work link