Deep Learning-based Bio-Medical Image Segmentation using UNet Architecture and Transfer Learning

TL;DR

This paper implements a UNet-based deep learning approach for biomedical image segmentation, demonstrating that transfer learning improves performance over a from-scratch UNet model.

Contribution

It introduces a transfer learning application to UNet for biomedical segmentation, showing enhanced results compared to a basic implementation.

Findings

Transfer learning outperforms from-scratch UNet in segmentation accuracy.

Pre-trained models require fine-tuning for specific biomedical datasets.

The approach achieves high performance on multiple biomedical image datasets.

Abstract

Image segmentation is a branch of computer vision that is widely used in real world applications including biomedical image processing. With recent advancement of deep learning, image segmentation has achieved at a very high level performance. Recently, UNet architecture is found as the core of novel deep learning segmentation methods. In this paper we implement UNet architecture from scratch with using basic blocks in Pytorch and evaluate its performance on multiple biomedical image datasets. We also use transfer learning to apply novel modified UNet segmentation packages on the biomedical image datasets. We fine tune the pre-trained transferred model with each specific dataset. We compare its performance with our fundamental UNet implementation. We show that transferred learning model has better performance in image segmentation than UNet model that is implemented from scratch.