The study examines why dermatology AI models, largely trained on light‑skinned, cancer‑focused images, perform poorly when applied to diverse patient populations. By comparing a cancer‑trained baseline, two dermatology foundation models, and a general‑purpose vision model on tone‑stratified and disease‑shifted datasets, the authors find that disease‑distribution shift, rather than skin‑tone underrepresentation, is the primary cause of generalization failure. Representation analysis shows that cancer‑specialized features lack transferable structure, while dermatology‑pretrained features maintain stronger clustering, and lightweight adaptation with about ten labeled examples per category can recover most performance.
arXiv:2609.36400v1 Announce Type: cross Abstract: Deep learning classifiers for dermoscopic skin lesions often reach high in-distribution accuracy while quietly relying on spurious background cues su...
The paper evaluates how well generalist and dermatology-specific machine learning models perform on diverse skin lesion datasets, including dermoscopic images and smartphone photographs. It benchmarks a range of architectures—general-purpose vision-language models, foundation models, and task-specific dermatology classifiers—under conditions of distribution shift, modality change, and demographic variability. The study quantifies the performance gap between current state‑of‑the‑art models and the robustness needed for safe, equitable clinical deployment.
Accurate dermatological diagnosis naturally necessitates equitable performance across diverse populations, yet a systematic lack of expertly annotated images, especially for underrepresented skin tones and rare diseases, impedes progress toward measurably fair methods. We introduce cgDDI (Controllable Generation of Diverse Dermatological Imagery), a hybrid framework that (1) synthesizes realistic healthy skin samples without disturbing other input properties, (2) maps single-sample rare lesions onto novel skin-tones and locations non-parametrically, and (3) allows for efficient parametric generation with as few as 10 training samples.
arXiv:2610.08086v1 Announce Type: new Abstract: The clinical integration of AI systems in digital dermatology relies heavily on human trust. Clinically interpretable visual concepts can act as interm...
The paper introduces MIFR, a modality‑invariant and fair representation framework for skin disease classification that jointly processes clinical photographs and dermoscopic images using ViT‑based encoders. It employs a five‑component multi‑objective loss to balance classification accuracy, fairness across skin tones, class alignment, and modality invariance. Experiments on paired and external datasets demonstrate competitive predictive performance and fairness, with t‑SNE visualizations confirming alignment of embeddings from different modalities.
Skin diseases represent a major global public health burden, yet machine learning tools developed to assist in their diagnosis suffer from two critical limitations: reliance on only one modality for d...
Concept bottleneck models (CBMs) rely on reliable concept annotations to allow clinicians to inspect and correct skin disease diagnoses. An audit of the SkinCon dermatology concept dataset reveals a tone‑dependent pattern: erythema is annotated far less often on dark skin (odds ratio 0.17) while hyperpigmentation and hypopigmentation are annotated more often (odds ratios 6.2 and 4.5). An evidential concept model trained on these annotations reproduces the same bias, and three label‑error detectors perform worse than a disease‑only baseline at detecting erythema mentioned by dermatologists but absent in SkinCon labels. The study concludes that concept annotations must be audited within each subgroup before they can be trusted for explanations.
arXiv:2607. 26765v1 Announce Type: cross Abstract: Background/Objectives: Dermoscopic skin lesion classifiers often lose accuracy under domain shift across imaging devices, illumination, and capture artifacts.
GAMF-Net is a grouped availability‑aware multimodal fusion network designed for smartphone skin lesion classification. It divides ten clinical variables into three semantic groups—demographics, history‑related risk, and lesion presentation—and encodes each group with field values and availability masks. The image branch supplies base logits while each clinical group contributes image‑conditioned residual evidence; only available groups are fused with equal weight, and the model defaults to image‑only when no clinical data are present. Evaluated on PAD‑UFES‑20, GAMF-Net achieves an accuracy of 74.17 % and outperforms image‑only and other multimodal baselines, with ablation studies highlighting the importance of clinical information and constrained equal‑weight fusion.
Background/Objectives: Dermoscopic skin lesion classifiers often lose accuracy under domain shift across imaging devices, illumination, and capture artifacts. We study how data augmentation improves the robustness of a binary malignant-versus-non-malignant classifier, with emphasis on out-of-domain (OOD) generalization.
arXiv:2606. 13135v1 Announce Type: cross Abstract: Purpose.
The study compares five pre‑trained convolutional neural networks—ResNet50, VGG16, VGG19, MobileNet, and InceptionV3—for melanoma detection using dermatoscopic and histopathological image datasets. Accuracy varied across models and modalities, with ResNet50 achieving the highest scores (84% on HAM10000 and 83% on CR‑AI4SkIN) and InceptionV3 the lowest (71% on ISIC 2018). The results show that a model’s performance on dermatoscopic images does not necessarily predict its performance on histopathological images.