13 October 2019 · MICCAI 2019, Shenzhen, China
Organizing Committee Member at MICCAI DART 2019
1st MICCAI workshop on Domain Adaptation and Representation Transfer ↗

On 13 October 2019 we held the first MICCAI Workshop on Domain Adaptation and Representation Transfer (DART), Towards Learning Transferable, Interpretable, and Robust Representations, at MICCAI 2019 in Shenzhen, China. DART is a forum to compare and discuss methods that keep machine learning robust when data come from different domains, such as scanners, protocols or sites, so that models can be used in the clinic. Topics include domain adaptation, transfer learning, few-shot learning and interpretable representations.
Klaus Maier-Hein (DKFZ and Heidelberg University Hospital) gave the keynote, Towards generalizability and clinical applicability of deep learning models in medical imaging. The workshop accepted 12 papers, and the best paper award, a GPU sponsored by NVIDIA, went to Ilja Manakov (LMU Munich).
Organisers: Fausto Milletari, Maximilian Baust, Abood Quraini, Nicola Rieke, Daguang Xu and Ziyue Xu (NVIDIA), M. Jorge Cardoso (King’s College London), Konstantinos Kamnitsas (Imperial College London) and Shadi Albarqouni (TU Munich). The papers appeared in the joint DART 2019 and MIL3ID 2019 proceedings, Springer LNCS 11795, which I co-edited. Source: DART 2019 website.
