|Authors||K. Pogorelov, O. Ostroukhova, A. Petlund, P. Halvorsen, H. Espeland, T. Kupka, T. de Lange, C. Griwodz and M. Riegler|
|Title||Automatic Detection of Angiectasia: Evaluation of Deep Learning and Handcrafted Approaches|
|Project(s)||Efficient EONS: Execution of Large Workloads on Elastic Heterogeneous Resources|
|Publication Type||Talk, keynote|
|Year of Publication||2018|
|Location of Talk||IEEE Conference on Biomedical and Health Informatics (BHI) 2018|
Angiectasia, formerly called angiodysplasia, is one of the most frequent vascular lesions and often the cause of gastrointestinal bleedings. Medical specialists assessing videos of examinations reach a detection performance of 16% for the detection of bleeding to 69% for the detection of angiectasia. In this paper, we present several machine-learning-based approaches for angiectasia detection in wireless video capsule endoscopy images. The most promising results for pixel-wise localization and frame-wise detection are obtained by the proposed deep learning approach using generative adversarial networks (GANs) with a sensitivity of 88% and specificity of 99.9% for pixel-wise localization and a sensitivity of 98% and a specificity of 100% for frame-wise detection, which fits the requirements for automatic angiectasia detection in real clinical settings.