Advancements in Medical Image Classification Through Fine-tuning Natural Domain Foundation Models
Year: 2025
Type: article
Abstract: Using massive datasets, foundation models are large-scale, pre-trained models that perform a wide range of tasks. These models have shown consistently improved results with the introduction of new methods. It is crucial to analyze how these trends impact the medical field and determine if these advancements can influence the medical domain. This study investigates the application of re- cent state-of-the-art foundation models—DINOv2, MAE, VMamba, CoCa, SAM 2, and AIMv2—for medical image classification. We explore their effectiveness on datasets including CBIS-DDSM for mammography, ISIC2019 for skin lesions, APTOS2019 for diabetic retinopathy, and CHEXPERT for chest radiographs. By fine-tuning these models and evaluating their configurations, we aim to understand the potential of these advancements in medical image classification. The results indicate that these advanced models significantly enhance classification outcomes, demonstrating robust performance despite limited labeled data. Based on our results, AIMv2, the DINOv2 and SAM2 models outperformed others, demonstrating that progress in natural domain training has positively impacted the medical domain and improved classification outcomes.
Source: IEEE SIGPORT
DOI: 10.60864/rczg-zw25
Authors: Mobina Mansoori, Sajjad Shahabodini, Farnoush Bayatmakou, Jamshid Abouei, Konstantinos N. Plataniotis, and Arash Mohammadi
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6958803266
FWCI:
Cites:
Cited by:
Related to: