Domain Adaptation Techniques for Natural and Medical Image Classification

2025-08-28

AI Summary

To address degraded generalization in natural and medical image classification caused by train/test distribution shifts, this work systematically evaluates domain adaptation (DA) techniques in clinical settings. We conduct 557 experiments across five natural-image and eight medical-image datasets, benchmarking seven state-of-the-art DA methods under challenging scenarios—including out-of-distribution generalization, dynamic data streams, and few-shot learning. Results show that Deep Subdomain Adaptation Network (DSAN) achieves superior performance on medical images: it yields a 6.7% accuracy gain under dynamic data streams and attains 91.2% accuracy on the COVID-19 dataset. Crucially, DSAN’s attention mechanism provides interpretable cross-domain alignment evidence. This study constitutes the first large-scale empirical validation of DSAN’s robustness and practicality across heterogeneous medical vision tasks. It establishes DSAN as a trustworthy, deployable DA solution for low-resource, high-bias medical image classification—bridging critical gaps between theoretical DA advances and real-world clinical deployment.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Medical and Biological Imaging
• Application Domains: Natural Sciences

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Domain adaptation (DA) techniques have the potential in machine learning to alleviate distribution differences between training and test sets by leveraging information from source domains. In image classification, most advances in DA have been made using natural images rather than medical data, which are harder to work with. Moreover, even for natural images, the use of mainstream datasets can lead to performance bias. {With the aim of better understanding the benefits of DA for both natural and medical images, this study performs 557 simulation studies using seven widely-used DA techniques for image classification in five natural and eight medical datasets that cover various scenarios, such as out-of-distribution, dynamic data streams, and limited training samples.} Our experiments yield detailed results and insightful observations highlighting the performance and medical applicability of these techniques. Notably, our results have shown the outstanding performance of the Deep Subdomain Adaptation Network (DSAN) algorithm. This algorithm achieved feasible classification accuracy (91.2%) in the COVID-19 dataset using Resnet50 and showed an important accuracy improvement in the dynamic data stream DA scenario (+6.7%) compared to the baseline. Our results also demonstrate that DSAN exhibits remarkable level of explainability when evaluated on COVID-19 and skin cancer datasets. These results contribute to the understanding of DA techniques and offer valuable insight into the effective adaptation of models to medical data.

Problem

• Evaluating domain adaptation techniques for natural and medical images
• Addressing performance bias in mainstream datasets for image classification
• Assessing model adaptation effectiveness in medical data scenarios

Innovation

• Deep Subdomain Adaptation Network algorithm
• Resnet50 for COVID-19 classification
• Dynamic data stream domain adaptation