AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images

Recent MLP-Mixer has a good ability to handle long-range dependencies, however, to have a good performance, one requires huge data and expensive infrastructures for the pre-training process. In this study, we proposed a novel model for nuclei image segmentation namely Axial Convolutional-MLP Mixer,...

Full description

Saved in:
Bibliographic Details
Main Authors: Nguyen Thanh Thu, Dinh Binh Duong, Tran Thi Thao, Pham Van Truong
Format: Article
Language:English
Published: The University of Danang 2024-12-01
Series:Tạp chí Khoa học và Công nghệ
Subjects:
Online Access:https://jst-ud.vn/jst-ud/article/view/9327
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850192475534131200
author Nguyen Thanh Thu
Dinh Binh Duong
Tran Thi Thao
Pham Van Truong
author_facet Nguyen Thanh Thu
Dinh Binh Duong
Tran Thi Thao
Pham Van Truong
author_sort Nguyen Thanh Thu
collection DOAJ
description Recent MLP-Mixer has a good ability to handle long-range dependencies, however, to have a good performance, one requires huge data and expensive infrastructures for the pre-training process. In this study, we proposed a novel model for nuclei image segmentation namely Axial Convolutional-MLP Mixer, by replacing the token mixer of MLP-Mixer with a new operator, Axial Convolutional Token Mix. Specifically, in the Axial Convolutional Token Mix, we inherited the idea of axial depthwise convolution to create a flexible receptive field. We also proposed a Long-range Attention module that uses dilated convolution to extend the convolutional kernel size, thereby addressing the issue of long-range dependencies. Experiments demonstrate that our model can achieve high results on small medical datasets, with Dice scores of 90.20% on the GlaS dataset, 80.43% on the MoNuSeg dataset, and without pre-training. The code will be available at https://github.com/thanhthu152/AC-MLP.
format Article
id doaj-art-221d48bd385449efb8e9c4e7c18cdfff
institution OA Journals
issn 1859-1531
language English
publishDate 2024-12-01
publisher The University of Danang
record_format Article
series Tạp chí Khoa học và Công nghệ
spelling doaj-art-221d48bd385449efb8e9c4e7c18cdfff2025-08-20T02:14:32ZengThe University of DanangTạp chí Khoa học và Công nghệ1859-15312024-12-01646910.31130/ud-jst.2024.332E9321AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological imagesNguyen Thanh Thu0Dinh Binh Duong1Tran Thi Thao2Pham Van Truong3School of Electrical and Electronic Engineering, Hanoi University of Science and Technology, VietnamSchool of Electrical and Electronic Engineering, Hanoi University of Science and Technology, VietnamSchool of Electrical and Electronic Engineering, Hanoi University of Science and Technology, VietnamSchool of Electrical and Electronic Engineering, Hanoi University of Science and Technology, VietnamRecent MLP-Mixer has a good ability to handle long-range dependencies, however, to have a good performance, one requires huge data and expensive infrastructures for the pre-training process. In this study, we proposed a novel model for nuclei image segmentation namely Axial Convolutional-MLP Mixer, by replacing the token mixer of MLP-Mixer with a new operator, Axial Convolutional Token Mix. Specifically, in the Axial Convolutional Token Mix, we inherited the idea of axial depthwise convolution to create a flexible receptive field. We also proposed a Long-range Attention module that uses dilated convolution to extend the convolutional kernel size, thereby addressing the issue of long-range dependencies. Experiments demonstrate that our model can achieve high results on small medical datasets, with Dice scores of 90.20% on the GlaS dataset, 80.43% on the MoNuSeg dataset, and without pre-training. The code will be available at https://github.com/thanhthu152/AC-MLP.https://jst-ud.vn/jst-ud/article/view/9327depthwise convolutionmlp-mixernuclei segmentationtoken mixing
spellingShingle Nguyen Thanh Thu
Dinh Binh Duong
Tran Thi Thao
Pham Van Truong
AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images
Tạp chí Khoa học và Công nghệ
depthwise convolution
mlp-mixer
nuclei segmentation
token mixing
title AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images
title_full AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images
title_fullStr AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images
title_full_unstemmed AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images
title_short AC-MLP: Axial Convolution-MLP Mixer for nuclei segmentation in histopathological images
title_sort ac mlp axial convolution mlp mixer for nuclei segmentation in histopathological images
topic depthwise convolution
mlp-mixer
nuclei segmentation
token mixing
url https://jst-ud.vn/jst-ud/article/view/9327
work_keys_str_mv AT nguyenthanhthu acmlpaxialconvolutionmlpmixerfornucleisegmentationinhistopathologicalimages
AT dinhbinhduong acmlpaxialconvolutionmlpmixerfornucleisegmentationinhistopathologicalimages
AT tranthithao acmlpaxialconvolutionmlpmixerfornucleisegmentationinhistopathologicalimages
AT phamvantruong acmlpaxialconvolutionmlpmixerfornucleisegmentationinhistopathologicalimages