Going Beyond U-Net: Assessing Vision Transformers for Semantic Segmentation in Microscopy Image Analysis
This work addresses segmentation challenges in microscopy image analysis for biomedical researchers, offering incremental improvements through model modifications.
The study assessed transformer-based models like UNETR, Segment Anything Model, and Swin-UPerNet for semantic segmentation in microscopy images, finding that architectural modifications to Swin Transformer improved performance over U-Net and unmodified versions.
Segmentation is a crucial step in microscopy image analysis. Numerous approaches have been developed over the past years, ranging from classical segmentation algorithms to advanced deep learning models. While U-Net remains one of the most popular and well-established models for biomedical segmentation tasks, recently developed transformer-based models promise to enhance the segmentation process of microscopy images. In this work, we assess the efficacy of transformers, including UNETR, the Segment Anything Model, and Swin-UPerNet, and compare them with the well-established U-Net model across various image modalities such as electron microscopy, brightfield, histopathology, and phase-contrast. Our evaluation identifies several limitations in the original Swin Transformer model, which we address through architectural modifications to optimise its performance. The results demonstrate that these modifications improve segmentation performance compared to the classical U-Net model and the unmodified Swin-UPerNet. This comparative analysis highlights the promise of transformer models for advancing biomedical image segmentation. It demonstrates that their efficiency and applicability can be improved with careful modifications, facilitating their future use in microscopy image analysis tools.