PulseAugur
EN
LIVE 14:34:39

New datasets and AI methods boost skin lesion classification

By PulseAugur Editorial · [3 sources] ·

Researchers have developed new datasets and methods to improve AI's ability to classify skin lesions from dermatoscopic images. One paper introduces IMA++, a large dataset with over 17,000 segmentation masks from nearly 15,000 images, including multiple annotations per image to aid research on annotator preferences. Another study, DerMAE, uses synthetic data generation and knowledge distillation to train lightweight models for efficient on-device skin lesion classification, addressing class imbalance issues. A third paper proposes a contrastive meta-domain adaptation strategy to enhance the robustness of skin lesion classification models across different clinical and acquisition conditions. AI

IMPACT Advances in AI for medical imaging could lead to earlier and more accurate skin cancer detection.

RANK_REASON Multiple research papers released on arXiv detailing new datasets and methods for skin lesion classification.

Read on arXiv cs.CV →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

COVERAGE [3]

  1. arXiv cs.CV TIER_1 Italiano(IT) · Kumar Abhishek, Jeremy Kawahara, Ghassan Hamarneh ·

    IMA++: ISIC Archive Multi-Annotator Dermoscopic Skin Lesion Segmentation Dataset

    arXiv:2512.21472v2 Announce Type: replace Abstract: Multi-annotator medical image segmentation is an important research problem, but requires annotated datasets that are expensive to collect. Dermoscopic skin lesion imaging allows human experts and AI systems to observe morpholog…

  2. arXiv cs.CV TIER_1 English(EN) · Francisco Filho, Kelvin Cunha, F\'abio Papais, Emanoel dos Santos, Rodrigo Mota, Thales Bezerra, Erico Medeiros, Paulo Borba, Tsang Ing Ren ·

    DerMAE: Improving skin lesion classification through conditioned latent diffusion and MAE distillation

    arXiv:2602.19848v2 Announce Type: replace Abstract: Skin lesion classification datasets often suffer from severe class imbalance, with malignant cases significantly underrepresented, leading to biased decision boundaries during deep learning training. We address this challenge us…

  3. arXiv cs.CV TIER_1 English(EN) · Rodrigo Mota, Kelvin Cunha, Emanoel dos Santos, F\'abio Papais, Francisco Filho, Thales Bezerra, Erico Medeiros, Paulo Borba, Tsang Ing Ren ·

    Contrastive meta-domain adaptation for robust skin lesion classification across clinical and acquisition conditions

    arXiv:2602.19857v2 Announce Type: replace Abstract: Deep learning models for dermatological image analysis remain sensitive to acquisition variability and domain-specific visual characteristics, leading to performance degradation when deployed in clinical settings. We investigate…

RELATED ENTITIES

RELATED TOPICS