Universal Lesion Segmentation Challenge 2023: A Comparative Research of Different Algorithms

TL;DR

This paper compares various algorithms for universal lesion segmentation, ultimately selecting SwinUnet as the most effective model, and discusses future research directions in this domain.

Contribution

The study systematically evaluates multiple architectures for universal lesion segmentation and identifies SwinUnet as the best performing model for diverse tissue types.

Findings

SwinUnet outperforms other tested architectures.

The paper provides insights into the strengths and weaknesses of different models.

Future directions for improving universal lesion segmentation are proposed.

Abstract

In recent years, machine learning algorithms have achieved much success in segmenting lesions across various tissues. There is, however, not one satisfying model that works well on all tissue types universally. In response to this need, we attempt to train a model that 1) works well on all tissue types, and 2) is capable of still performing fast inferences. To this end, we design our architectures, test multiple existing architectures, compare their results, and settle upon SwinUnet. We document our rationales, successes, and failures. Finally, we propose some further directions that we think are worth exploring. codes: https://github.com/KWFredShi/ULS2023NGKD.git