Simple is what you need for efficient and accurate medical image segmentation

TL;DR

This paper introduces SimpleUNet, an ultra-lightweight yet high-performing medical image segmentation model that combines simplicity, efficiency, and competitive accuracy across multiple datasets, challenging the notion that larger models are necessary for good performance.

Contribution

The paper proposes SimpleUNet, a novel lightweight segmentation architecture with three key innovations, achieving state-of-the-art results with minimal parameters and computational cost.

Findings

Record-breaking 16 KB parameter model outperforms benchmarks.

Achieves 85.76% DSC on breast lesion datasets, surpassing U-Net.

Maintains high accuracy on skin lesion and polyp segmentation datasets.

Abstract

While modern segmentation models often prioritize performance over practicality, we advocate a design philosophy prioritizing simplicity and efficiency, and attempted high performance segmentation model design. This paper presents SimpleUNet, a scalable ultra-lightweight medical image segmentation model with three key innovations: (1) A partial feature selection mechanism in skip connections for redundancy reduction while enhancing segmentation performance; (2) A fixed-width architecture that prevents exponential parameter growth across network stages; (3) An adaptive feature fusion module achieving enhanced representation with minimal computational overhead. With a record-breaking 16 KB parameter configuration, SimpleUNet outperforms LBUNet and other lightweight benchmarks across multiple public datasets. The 0.67 MB variant achieves superior efficiency (8.60 GFLOPs) and accuracy, attaining a mean DSC/IoU of 85.76%/75.60% on multi-center breast lesion datasets, surpassing both U-Net and TransUNet. Evaluations on skin lesion datasets (ISIC 2017/2018: mDice 84.86%/88.77%) and endoscopic polyp segmentation (KVASIR-SEG: 86.46%/76.48% mDice/mIoU) confirm consistent dominance over state-of-the-art models. This work demonstrates that extreme model compression need not compromise performance, providing new insights for efficient and accurate medical image segmentation. Codes can be found at https://github.com/Frankyu5666666/SimpleUNet.