{"title":"MoE-NuSeg: Enhancing nuclei segmentation in histology images with a two-stage Mixture of Experts network","authors":"","doi":"10.1016/j.aej.2024.10.011","DOIUrl":null,"url":null,"abstract":"<div><div>Accurate nuclei segmentation is essential for extracting quantitative information from histology images to support disease diagnosis and treatment decisions. However, precise segmentation is challenging due to the presence of clustered nuclei, varied morphologies, and the need to capture global spatial correlations. While state-of-the-art Transformer-based models employ tri-decoder architectures to decouple the segmentation task into nuclei, edges, and cluster edges segmentation, their complexity and long inference times hinder clinical integration. To address this, we introduce MoE-NuSeg, a novel Mixture of Experts (MoE) network that consolidates the tri-decoder into a single decoder. MoE-NuSeg employs three specialized experts for nuclei segmentation, edge delineation, and cluster edge detection, thereby mirroring the functionality of tri-decoders while surpassing their performance and reducing parameters by sharing attention heads. We propose a two-stage training strategy: the first stage independently trains the three experts, and the second stage fine-tunes their interactions to dynamically allocate the contributions of each expert using a learnable attention-based gating network. Evaluations across three datasets demonstrate that MoE-NuSeg outperforms the state-of-the-art methods, achieving an average increase of 0.99% in Dice coefficient, 1.14% in IoU and 0.92% in F1 Score, while reducing parameters by 30.1% and FLOPs by 40.2%. The code is available at <span><span>https://github.com/deep-geo/MoE-NuSeg</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":null,"pages":null},"PeriodicalIF":6.2000,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"alexandria engineering journal","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1110016824011669","RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
Accurate nuclei segmentation is essential for extracting quantitative information from histology images to support disease diagnosis and treatment decisions. However, precise segmentation is challenging due to the presence of clustered nuclei, varied morphologies, and the need to capture global spatial correlations. While state-of-the-art Transformer-based models employ tri-decoder architectures to decouple the segmentation task into nuclei, edges, and cluster edges segmentation, their complexity and long inference times hinder clinical integration. To address this, we introduce MoE-NuSeg, a novel Mixture of Experts (MoE) network that consolidates the tri-decoder into a single decoder. MoE-NuSeg employs three specialized experts for nuclei segmentation, edge delineation, and cluster edge detection, thereby mirroring the functionality of tri-decoders while surpassing their performance and reducing parameters by sharing attention heads. We propose a two-stage training strategy: the first stage independently trains the three experts, and the second stage fine-tunes their interactions to dynamically allocate the contributions of each expert using a learnable attention-based gating network. Evaluations across three datasets demonstrate that MoE-NuSeg outperforms the state-of-the-art methods, achieving an average increase of 0.99% in Dice coefficient, 1.14% in IoU and 0.92% in F1 Score, while reducing parameters by 30.1% and FLOPs by 40.2%. The code is available at https://github.com/deep-geo/MoE-NuSeg.
期刊介绍:
Alexandria Engineering Journal is an international journal devoted to publishing high quality papers in the field of engineering and applied science. Alexandria Engineering Journal is cited in the Engineering Information Services (EIS) and the Chemical Abstracts (CA). The papers published in Alexandria Engineering Journal are grouped into five sections, according to the following classification:
• Mechanical, Production, Marine and Textile Engineering
• Electrical Engineering, Computer Science and Nuclear Engineering
• Civil and Architecture Engineering
• Chemical Engineering and Applied Sciences
• Environmental Engineering