Multi-scale Microaneurysms Segmentation Using Embedding Triplet Loss


Deep learning techniques are recently being used in fundus image analysis and diabetic retinopathy detection. Microaneurysms are an important indicator of diabetic retinopathy progression. We introduce a two-stage deep learning approach for microaneurysms segmentation using multiple scales of the input with selective sampling and embedding triplet loss. The model first segments on two scales and then the segmentations are refined with a classification model. To enhance the discriminative power of the classification model, we incorporate triplet embedding loss with a selective sampling routine. The model is evaluated quantitatively to assess the segmentation performance and qualitatively to analyze the model predictions. This approach introduces a 30.29% relative improvement over the fully convolutional neural network.

International Conference on Medical Image Computing and Computer-Assisted Intervention
Shadi Albarqouni
Shadi Albarqouni
Professor of Computational Medical Imaging Research at University of Bonn | AI Young Investigator Group Leader at Helmholtz AI