Loss Ensembles for Extremely Imbalanced Segmentation
This paper addresses the problem of segmenting intracranial aneurysms from brain MR scans, which is significant for medical diagnosis and treatment planning, demonstrating an incremental improvement in a specific medical imaging challenge.
This paper presents a method for automatic intracranial aneurysm segmentation from brain MR scans, achieving first place in the ADAM challenge segmentation task by using ensembles of multiple models trained with different loss functions.
This short paper briefly presents our methodology details of automatic intracranial aneurysms segmentation from brain MR scans. We use ensembles of multiple models trained from different loss functions. Our method ranked first place in the ADAM challenge segmentation task. The code and trained models are publicly available at https://github.com/JunMa11/ADAM2020.