FunduSAM: A Specialized Deep Learning Model for Enhanced Optic Disc and Cup Segmentation in Fundus Images

TL;DR

FunduSAM is a specialized deep learning model that enhances optic disc and cup segmentation in fundus images by incorporating adapters, attention modules, and polar transformation, outperforming existing methods on the REFUGE dataset.

Contribution

The paper introduces FunduSAM, a novel adaptation of SAM with adapters and attention modules tailored for OD and OC segmentation in fundus images, addressing challenges of low contrast and blurred boundaries.

Findings

FunduSAM outperforms five mainstream approaches on the REFUGE dataset.

Incorporating adapters and CBAM improves segmentation accuracy.

Polar transformation enhances model training and evaluation.

Abstract

The Segment Anything Model (SAM) has gained popularity as a versatile image segmentation method, thanks to its strong generalization capabilities across various domains. However, when applied to optic disc (OD) and optic cup (OC) segmentation tasks, SAM encounters challenges due to the complex structures, low contrast, and blurred boundaries typical of fundus images, leading to suboptimal performance. To overcome these challenges, we introduce a novel model, FunduSAM, which incorporates several Adapters into SAM to create a deep network specifically designed for OD and OC segmentation. The FunduSAM utilizes Adapter into each transformer block after encoder for parameter fine-tuning (PEFT). It enhances SAM's feature extraction capabilities by designing a Convolutional Block Attention Module (CBAM), addressing issues related to blurred boundaries and low contrast. Given the unique requirements of OD and OC segmentation, polar transformation is used to convert the original fundus OD images into a format better suited for training and evaluating FunduSAM. A joint loss is used to achieve structure preservation between the OD and OC, while accurate segmentation. Extensive experiments on the REFUGE dataset, comprising 1,200 fundus images, demonstrate the superior performance of FunduSAM compared to five mainstream approaches.