Improving Lesion Segmentation for Diabetic Retinopathy using Adversarial Learning
This work addresses the challenge of automatic lesion segmentation for diabetic retinopathy, a leading cause of blindness, but it appears incremental as it builds on existing methods like HEDNet and cGANs.
The paper tackled the problem of segmenting diabetic retinopathy lesions in fundus images by proposing an end-to-end system that incorporates HEDNet into a Conditional Generative Adversarial Network (cGAN) with a combined adversarial and segmentation loss. The result showed that this approach improves lesion segmentation performance over the baseline, though no concrete numbers are provided.
Diabetic Retinopathy (DR) is a leading cause of blindness in working age adults. DR lesions can be challenging to identify in fundus images, and automatic DR detection systems can offer strong clinical value. Of the publicly available labeled datasets for DR, the Indian Diabetic Retinopathy Image Dataset (IDRiD) presents retinal fundus images with pixel-level annotations of four distinct lesions: microaneurysms, hemorrhages, soft exudates and hard exudates. We utilize the HEDNet edge detector to solve a semantic segmentation task on this dataset, and then propose an end-to-end system for pixel-level segmentation of DR lesions by incorporating HEDNet into a Conditional Generative Adversarial Network (cGAN). We design a loss function that adds adversarial loss to segmentation loss. Our experiments show that the addition of the adversarial loss improves the lesion segmentation performance over the baseline.