One-Shot Neural Architecture Search with Network Similarity Directed Initialization for Pathological Image Classification

2025/06/17 by Renao Yan, Yan, Renao
Computer Science · Engineering · #AI in cancer detection #Computer Vision and Pattern Recognition (cs.CV) #Digital Imaging for Blood Diseases #FOS: Computer and information sciences #Medical Imaging and Analysis

paper · pdf · doi:10.48550/arxiv.2506.14176

openalex publication_date 2025/06/17 · openalex created_date 2025/10/18 · openalex updated_date 2026/07/28

Abstract

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.

Citations

• MergeUp-augmented Semi-Weakly Supervised Learning for WSI Classification
• Task-oriented Embedding Counts: Heuristic Clustering-driven Feature Fine-tuning for Whole Slide Image Classification
• Shapley Values-enabled Progressive Pseudo Bag Augmentation for Whole Slide Image Classification
• Stain-invariant self supervised learning for histopathology image analysis
• Differentiable Zooming for Multiple Instance Learning on Whole-Slide Images
• BRACS: A Dataset for BReAst Carcinoma Subtyping in H&E Histology Images
• Probeable DARTS with Application to Computational Pathology
• Hierarchical Graph Representations in Digital Pathology
• Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries
• Training data-efficient image transformers & distillation through attention
• AutoBSS: An Efficient Algorithm for Block Stacking Style Search
• MS-NAS: Multi-Scale Neural Architecture Search for Medical Image Segmentation
• FBNetV2: Differentiable Neural Architecture Search for Spatial and Channel Dimensions
• Multinomial Distribution Learning for Effective Neural Architecture Search
• Single Path One-Shot Neural Architecture Search with Uniform Sampling
• FBNet: Hardware-Aware Efficient ConvNet Design via Differentiable Neural Architecture Search
• ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
• DARTS: Differentiable Architecture Search
• Regularized Evolution for Image Classifier Architecture Search
• A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets
• Learning Transferable Architectures for Scalable Image Recognition
• Neural Architecture Search with Reinforcement Learning
• Learning Deep Features for Discriminative Localization
• Graph drawing by force‐directed placement

Related

• A Lightweight Neural Architecture Search Model for Medical Image Classification
• Probeable DARTS with Application to Computational Pathology
• Meta-Learning of NAS for Few-shot Learning in Medical Image Applications
• Dual-channel Prototype Network for few-shot Classification of Pathological Images
• MedNNS: Supernet-based Medical Task-Adaptive Neural Network Search