Advancing Medical Image Segmentation with Mini-Net: A Lightweight Solution Tailored for Efficient Segmentation of Medical Images

TL;DR

Mini-Net is a lightweight, efficient medical image segmentation network with fewer than 38,000 parameters, designed for real-time applications and validated across multiple datasets, outperforming some existing methods.

Contribution

The paper introduces Mini-Net, a novel lightweight segmentation network tailored for medical images, addressing computational challenges and optimizing performance for medical imaging tasks.

Findings

Mini-Net achieves robust segmentation across diverse datasets.

Mini-Net operates in real-time with fewer than 38,000 parameters.

Mini-Net outperforms some state-of-the-art methods in accuracy.

Abstract

Accurate segmentation of anatomical structures and abnormalities in medical images is crucial for computer-aided diagnosis and analysis. While deep learning techniques excel at this task, their computational demands pose challenges. Additionally, some cutting-edge segmentation methods, though effective for general object segmentation, may not be optimised for medical images. To address these issues, we propose Mini-Net, a lightweight segmentation network specifically designed for medical images. With fewer than 38,000 parameters, Mini-Net efficiently captures both high- and low-frequency features, enabling real-time applications in various medical imaging scenarios. We evaluate Mini-Net on various datasets, including DRIVE, STARE, ISIC-2016, ISIC-2018, and MoNuSeg, demonstrating its robustness and good performance compared to state-of-the-art methods.