Despite significant advancements in the automatic classification of skin lesions using artificial intelligence (AI) algorithms, skepticism among physicians persists. This reluctance is primarily due to the lack of transparency and explainability inherent in these models, which hinders their widespread acceptance in clinical settings. The primary objective of this study is to develop a highly accurate AI-based algorithm for skin lesion classification that also provides visual explanations to foster trust and confidence in these novel diagnostic tools. By improving transparency, the study seeks to contribute to earlier and more reliable diagnoses. Additionally, the research investigates the impact of Test Time Augmentation (TTA) on the performance of six Convolutional Neural Network (CNN) architectures, which include models from the EfficientNet, ResNet (Residual Network), and ResNeXt (an enhanced variant of ResNet) families. To improve the interpretability of the models’ decision-making processes, techniques such as t-distributed Stochastic Neighbor Embedding (t-SNE) and Gradient-weighted Class Activation Mapping (Grad-CAM) are employed. t-SNE is utilized to visualize the high-dimensional latent features of the CNNs in a two-dimensional space, providing insights into how the models group different skin lesion classes. Grad-CAM is used to generate heatmaps that highlight the regions of input images that influence the model’s predictions. Our findings reveal that Test Time Augmentation enhances the balanced multi-class accuracy of CNN models by up to 0.3%, achieving a balanced accuracy rate of 97.58% on the International Skin Imaging Collaboration (ISIC 2019) dataset. This performance is comparable to, or marginally better than, more complex approaches such as Vision Transformers (ViTs), demonstrating the efficacy of our methodology.

A case of a severe necrotizing vasculopathic skin lesions occurred in a 43 year old women affected by multiple sclerosis (MS) submitted to IFNb-lb has been described. A fter two months of therapy the patient presented, in injection sites of the abdomen, arms and legs, numerous ulcers. A biopsy of the lesions was performed and evidenced confluent necro sis of the superficial and deep skin tissue with mild infiltration by inflammatory cells and thrombosis in deep blood vessels. The IFNb-lb was immediately discontinued and therapy with corticosteroids was started. A fter 12 months from the onset of the adverse reaction, the skin vasculopathic lesions cicatrised leaving sclerotic areas on the abdomen. Neutralizing antibodies against IFNb-lb (NA Bs) were strongly positive at the onset of the skin ulcers and slowly decreased until the recovery. A possible role of NA Bs in the development of the skin lesions has been considered.

Dermoscopy is widely used for the diagnosis of skin cancers and it increases the accuracy of basal cell carcinoma (BCC) detection. The dermoscopic criteria of BCC have recently been updated and they can be divided into three categories: vascular, pigment-related, and nonvascular/nonpigmentrelated. The objective of our multicentre retrospective study was to report the validity of a new dermoscopic pigment-related feature utilized to detect pigmented BCC (pBCC) (i.e., brown homogeneous blotches (BHB)). Cases of pBCC were collected from the databases of IDI-IRCCS of Rome and from three Italian private dermatology centres. BHB were defined as the presence of circumscribed areas of brown homogeneous pigmentation which are devoid of any dermoscopic specific arrangement (net, fat fingers, etc.) and of other internal dermoscopic structures, except for occasional vascular ones, such as arborizing vessels or globules/dots. Controls included melanocytic and non-melanocytic lesions. We examined images of 270 pigmented lesions (female 145; 51.8%) including 90 histopathological proven pBCC and 180 control cases (i.e., 90 melanocytic and 90 non-melanocytic lesions). BHB were present in 61 cases of 90 pBCC. These results showed a sensibility of 67.8 and a specificity of 93.3 with a positive predictive value of 83.6, negative predictive value of 85.3, posLR 10.2, negLR 0.3; OR 29.4, P value<0.001. Our multicentre retrospective study showed that the BHB may represent a new dermoscopic criterion for the diagnosis of pBCC.

The integration of artificial intelligence (AI), particularly convolutional neural networks (CNNs), into dermatological diagnosis demonstrates substantial clinical potential. While the existing literature predominantly benchmarks algorithmic performance against human experts, our study adopts a novel perspective by investigating the intrinsic complexity of dermatoscopic images. Through rigorous experimentation with multiple CNN architectures, we isolated a subset of images systematically misclassified across all models—a phenomenon statistically proven to exceed random chance. To determine whether these failures stem from algorithmic biases or inherent visual ambiguity, expert dermatologists independently evaluated these challenging cases alongside a control group. The results revealed a collapse in human diagnostic performance on the AI-misclassified images. First, agreement with ground-truth labels plummeted, with Cohen’s kappa dropping to a mere 0.08 for this subset, compared to 0.61 for the control group. Second, we observed a severe deterioration in expert consensus; inter-rater reliability among physicians fell from moderate concordance (Fleiss’ kappa = 0.456) on control images to only modest agreement (Fleiss’ kappa = 0.275) on the misclassified subset. We identified image quality as a primary driver of these dual systematic failures. To promote transparency and reproducibility, all data, code, and trained models have been made publicly available.