Saved in:
Bibliographic Details
Main Author: Yan, Renao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.14176
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913898212360192
author Yan, Renao
author_facet Yan, Renao
contents Deep learning-based pathological image analysis presents unique challenges due to the practical constraints of network design. Most existing methods apply computer vision models directly to medical tasks, neglecting the distinct characteristics of pathological images. This mismatch often leads to computational inefficiencies, particularly in edge-computing scenarios. To address this, we propose a novel Network Similarity Directed Initialization (NSDI) strategy to improve the stability of neural architecture search (NAS). Furthermore, we introduce domain adaptation into one-shot NAS to better handle variations in staining and semantic scale across pathology datasets. Experiments on the BRACS dataset demonstrate that our method outperforms existing approaches, delivering both superior classification performance and clinically relevant feature localization.
format Preprint
id arxiv_https___arxiv_org_abs_2506_14176
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle One-Shot Neural Architecture Search with Network Similarity Directed Initialization for Pathological Image Classification
Yan, Renao
Computer Vision and Pattern Recognition
Deep learning-based pathological image analysis presents unique challenges due to the practical constraints of network design. Most existing methods apply computer vision models directly to medical tasks, neglecting the distinct characteristics of pathological images. This mismatch often leads to computational inefficiencies, particularly in edge-computing scenarios. To address this, we propose a novel Network Similarity Directed Initialization (NSDI) strategy to improve the stability of neural architecture search (NAS). Furthermore, we introduce domain adaptation into one-shot NAS to better handle variations in staining and semantic scale across pathology datasets. Experiments on the BRACS dataset demonstrate that our method outperforms existing approaches, delivering both superior classification performance and clinically relevant feature localization.
title One-Shot Neural Architecture Search with Network Similarity Directed Initialization for Pathological Image Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.14176