Hasil untuk "Dermatology"

Enno Schmidt, M. Sticherling, Miklós Sárdy et al.

(1) Department of Dermatology, University of Lübeck, Lübeck, Germany (2) Lübeck Institute of Experimental Dermatology, University of Lübeck, Lübeck, Germany (3) Department of Dermatology, University Hospital Erlangen, Erlangen, Germany (4) Department of Dermatology and Venereology, University Hospital Munich (LMU), Munich, Germany (5) Department of Dermatology and Allergology, Philipps-Universität Marburg, Marburg, Germany (6) Department of Dermatology, Venereology, and Allergology, University Hospital Würzburg, Würzburg, Germany (7) Helios University Hospital Wuppertal, Department of Dermatology, Allergology and Dermatosurgery, Wuppertal, Germany (8) Department of Dermatology and Venereology, University of Cologne, Cologne, Germany (9) Department of Dermatology, Medical Center, University of Freiburg, Freiburg, Germany (10) Dermatologist in Private Practice, Fulda, Germany (11) Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Department of Dermatology, Venereology und Allergy, Division of Evidence based Medicine (dEBM), Berlin, Germany (12) Gilead Sciences GmbH, Martinsried, Germany (13) Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Institut für Klinische Pharmakologie und Toxikologie, Berlin, Germany (14) Department of Dermatology and Allergology, University Hospital Ulm, Ulm, Germany (15) Pediatric Immunology and Rheumatology, University Hospital and Outpatient Clinic for Pediatrics, Leipzig, Germany (16) Charité – Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität zu Berlin, and Berlin Institute of Health, Department of Dermatology, Venereology und Allergy, Allergy Center, Berlin, Germany

D. Boas, A. Dunn

Areez Muhammed Shabu, Mohammad Samar Ansari, Asra Aslam

Skin cancer is one of the most common cancers worldwide and early detection is critical for effective treatment. However, current AI diagnostic tools are often trained on datasets dominated by lighter skin tones, leading to reduced accuracy and fairness for people with darker skin. The International Skin Imaging Collaboration (ISIC) dataset, one of the most widely used benchmarks, contains over 70% light skin images while dark skins fewer than 8%. This imbalance poses a significant barrier to equitable healthcare delivery and highlights the urgent need for methods that address demographic diversity in medical imaging. This paper addresses this challenge of skin tone imbalance in automated skin cancer detection using dermoscopic images. To overcome this, we present a generative augmentation pipeline that fine-tunes a pre-trained Stable Diffusion model using Low-Rank Adaptation (LoRA) on the image dark-skin subset of the ISIC dataset and generates synthetic dermoscopic images conditioned on lesion type and skin tone. In this study, we investigated the utility of these images on two downstream tasks: lesion segmentation and binary classification. For segmentation, models trained on the augmented dataset and evaluated on held-out real images show consistent improvements in IoU, Dice coefficient, and boundary accuracy. These evalutions provides the verification of Generated dataset. For classification, an EfficientNet-B0 model trained on the augmented dataset achieved 92.14% accuracy. This paper demonstrates that synthetic data augmentation with Generative AI integration can substantially reduce bias with increase fairness in conventional dermatological diagnostics and open challenges for future directions.

Qiuping Chen, Jifeng Liu

Enzo Errichetti, Noemi Plozner, Stinco Giuseppe

Not required

Alexa G. Ries, BS, Padma Mohandas, MBBS, Arucha L. Ekeowa-Anderson, MBBS et al.

Asif Mohammed Saad, Umme Niraj Mahi

Hair artifacts in dermoscopic images present significant challenges for accurate skin lesion analysis, potentially obscuring critical diagnostic features in dermatological assessments. This work introduces a fine-tuned SegFormer model augmented with dropout regularization to achieve precise hair mask segmentation. The proposed SegformerWithDropout architecture leverages the MiT-B2 encoder, pretrained on ImageNet, with an in-channel count of 3 and 2 output classes, incorporating a dropout probability of 0.3 in the segmentation head to prevent overfitting. Training is conducted on a specialized dataset of 500 dermoscopic skin lesion images with fine-grained hair mask annotations, employing 10-fold cross-validation, AdamW optimization with a learning rate of 0.001, and cross-entropy loss. Early stopping is applied based on validation loss, with a patience of 3 epochs and a maximum of 20 epochs per fold. Performance is evaluated using a comprehensive suite of metrics, including Intersection over Union (IoU), Dice coefficient, Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), and Learned Perceptual Image Patch Similarity (LPIPS). Experimental results from the cross-validation demonstrate robust performance, with average Dice coefficients reaching approximately 0.96 and IoU values of 0.93, alongside favorable PSNR (around 34 dB), SSIM (0.97), and low LPIPS (0.06), highlighting the model's effectiveness in accurate hair artifact segmentation and its potential to enhance preprocessing for downstream skin cancer detection tasks.

Pouyan Rezapoor, Aleksi Tamminen, Juha Ala-Laurinaho et al.

Terahertz (THz) imaging has emerged as a promising technology in medical diagnostics due to its non-ionizing radiation and high sensitivity to water content. However, conventional THz imaging systems face limitations such as slow mechanical scanning, restricted field of view, and poor telecentricity. To overcome these challenges, we introduce the Telecentric Offset Reflective Imaging System (TORIS), a novel dual-mirror scanning design optimized for high-speed, distortion-free imaging. The system employs a telecentric f-theta lens and is validated using ray tracing and physical optics simulations. It achieves uniform resolution across a 50 mm x 50 mm field of view without the need for mechanical translation stages. Broadband spectral imaging of a USAF resolution test target across WR-2.2 (325-500 GHz) and WR-1.5 (500-700 GHz) frequency bands demonstrates consistent beam focus and minimal distortion, with a maximum deviation of 2.7 degrees from normal incidence and a beam waist of 2.1 lambda at the field edge in the WR-1.5 band. The system's sensitivity to hydration dynamics is further validated through imaging of wet tissue paper, capturing temporal changes in water content. In vivo imaging of human skin after capsaicin patch application reveals localized hydration variations due to biochemical response and adhesive removal. These findings confirm the system's potential for real-time hydration sensing and dermatological evaluation. TORIS sets a new benchmark in THz imaging, with applications in clinical diagnostics, wound assessment, and material characterization.

Junyu Yan, Feng Chen, Yuyang Xue et al.

Recent studies have shown that Machine Learning (ML) models can exhibit bias in real-world scenarios, posing significant challenges in ethically sensitive domains such as healthcare. Such bias can negatively affect model fairness, model generalization abilities and further risks amplifying social discrimination. There is a need to remove biases from trained models. Existing debiasing approaches often necessitate access to original training data and need extensive model retraining; they also typically exhibit trade-offs between model fairness and discriminative performance. To address these challenges, we propose Soft-Mask Weight Fine-Tuning (SWiFT), a debiasing framework that efficiently improves fairness while preserving discriminative performance with much less debiasing costs. Notably, SWiFT requires only a small external dataset and only a few epochs of model fine-tuning. The idea behind SWiFT is to first find the relative, and yet distinct, contributions of model parameters to both bias and predictive performance. Then, a two-step fine-tuning process updates each parameter with different gradient flows defined by its contribution. Extensive experiments with three bias sensitive attributes (gender, skin tone, and age) across four dermatological and two chest X-ray datasets demonstrate that SWiFT can consistently reduce model bias while achieving competitive or even superior diagnostic accuracy under common fairness and accuracy metrics, compared to the state-of-the-art. Specifically, we demonstrate improved model generalization ability as evidenced by superior performance on several out-of-distribution (OOD) datasets.

Asif Newaz, Asif Ur Rahman Adib, Rajit Sahil et al.

Background: Arsenicosis is a serious public health concern in South and Southeast Asia, primarily caused by long-term consumption of arsenic-contaminated water. Its early cutaneous manifestations are clinically significant but often underdiagnosed, particularly in rural areas with limited access to dermatologists. Automated, image-based diagnostic solutions can support early detection and timely interventions. Methods: In this study, we propose an end-to-end framework for arsenicosis diagnosis using mobile phone-captured skin images. A dataset comprising 20 classes and over 11000 images of arsenic-induced and other dermatological conditions was curated. Multiple deep learning architectures, including convolutional neural networks (CNNs) and Transformer-based models, were benchmarked for arsenicosis detection. Model interpretability was integrated via LIME and Grad-CAM, while deployment feasibility was demonstrated through a web-based diagnostic tool. Results: Transformer-based models significantly outperformed CNNs, with the Swin Transformer achieving the best results (86\\% accuracy). LIME and Grad-CAM visualizations confirmed that the models attended to lesion-relevant regions, increasing clinical transparency and aiding in error analysis. The framework also demonstrated strong performance on external validation samples, confirming its ability to generalize beyond the curated dataset. Conclusion: The proposed framework demonstrates the potential of deep learning for non-invasive, accessible, and explainable diagnosis of arsenicosis from mobile-acquired images. By enabling reliable image-based screening, it can serve as a practical diagnostic aid in rural and resource-limited communities, where access to dermatologists is scarce, thereby supporting early detection and timely intervention.

Weiren Zhao, Feng Wang, Yanran Wang et al.

Recent advancements have highlighted the Mamba framework, a state-space model known for its efficiency in capturing long-range dependencies with linear computational complexity. While Mamba has shown competitive performance in medical image segmentation, it encounters difficulties in modeling local features due to the sporadic nature of traditional location-based scanning methods and the complex, ambiguous boundaries often present in medical images. To overcome these challenges, we propose Uncertainty-Driven Mamba (UD-Mamba), which redefines the pixel-order scanning process by incorporating channel uncertainty into the scanning mechanism. UD-Mamba introduces two key scanning techniques: 1) sequential scanning, which prioritizes regions with high uncertainty by scanning in a row-by-row fashion, and 2) skip scanning, which processes columns vertically, moving from high-to-low or low-to-high uncertainty at fixed intervals. Sequential scanning efficiently clusters high-uncertainty regions, such as boundaries and foreground objects, to improve segmentation precision, while skip scanning enhances the interaction between background and foreground regions, allowing for timely integration of background information to support more accurate foreground inference. Recognizing the advantages of scanning from certain to uncertain areas, we introduce four learnable parameters to balance the importance of features extracted from different scanning methods. Additionally, a cosine consistency loss is employed to mitigate the drawbacks of transitioning between uncertain and certain regions during the scanning process. Our method demonstrates robust segmentation performance, validated across three distinct medical imaging datasets involving pathology, dermatological lesions, and cardiac tasks.

Felix Wagner, Pramit Saha, Harry Anthony et al.

Safe deployment of machine learning (ML) models in safety-critical domains such as medical imaging requires detecting inputs with characteristics not seen during training, known as out-of-distribution (OOD) detection, to prevent unreliable predictions. Effective OOD detection after deployment could benefit from access to the training data, enabling direct comparison between test samples and the training data distribution to identify differences. State-of-the-art OOD detection methods, however, either discard the training data after deployment or assume that test samples and training data are centrally stored together, an assumption that rarely holds in real-world settings. This is because shipping the training data with the deployed model is usually impossible due to the size of training databases, as well as proprietary or privacy constraints. We introduce the Isolation Network, an OOD detection framework that quantifies the difficulty of separating a target test sample from the training data by solving a binary classification task. We then propose Decentralized Isolation Networks (DIsoN), which enables the comparison of training and test data when data-sharing is impossible, by exchanging only model parameters between the remote computational nodes of training and deployment. We further extend DIsoN with class-conditioning, comparing a target sample solely with training data of its predicted class. We evaluate DIsoN on four medical imaging datasets (dermatology, chest X-ray, breast ultrasound, histopathology) across 12 OOD detection tasks. DIsoN performs favorably against existing methods while respecting data-privacy. This decentralized OOD detection framework opens the way for a new type of service that ML developers could provide along with their models: providing remote, secure utilization of their training data for OOD detection services. Code: https://github.com/FelixWag/DIsoN

Wibisono Nugraha, Muhammad Eko Irawanto, Moerbono Mochtar et al.

Background: Keloid is a growth of fibrous tissue in the wound tissue of susceptible individuals. This tissue extends beyond the boundaries of the previous wound. The site of keloids commonly appears on a high-tension area, such as the chest, shoulders, and neck. Histopathologically keloids show thickened the epidermis and the vascularization and infiltration of inflammatory cells in the dermis. Purpose: This study aims to determine the relationship between the site of keloids and the histopathological appearance of keloids. Methods: An analytical observational study was conducted on keloid patients visiting the Dermatovenereology outpatient clinic of Dr.Moewardi Hospital. The sample collection used consecutive sampling techniques Result: The majority of keloid patients are >30 years old (46.4%). Most patients with keloids were female (53.6%). Keloids were mostly found on the chest (25.0%). Tounge-like appearance of the epidermis at the edges of the lesions was mostly on the shoulders and chest (33.3% each, p=0.048); flattened appearance was found on the middle epidermis of the lesions, which were mostly on the ears, shoulders, and upper extremities (22.7% each, p=0.011). Increased vascularity was found in the dermis at the edges of the lesions, especially in the ear and chest areas (31.3% each, p=0.046). Moderate-severe inflammatory infiltrates in the dermis at the edges of the lesions were commonly found on the chest (p=0.04). Conclusion: There is a significant relationship between the site of the lesion and the histopathological appearance of the keloid in epidermal as well as dermal layers.

Joana Revés, Catarina Bexiga, Alexandra Chaveiro et al.

Yutong Zhang, Yi Pan, Tianyang Zhong et al.

Medical images and radiology reports are crucial for diagnosing medical conditions, highlighting the importance of quantitative analysis for clinical decision-making. However, the diversity and cross-source heterogeneity of these data challenge the generalizability of current data-mining methods. Multimodal large language models (MLLMs) have recently transformed many domains, significantly affecting the medical field. Notably, Gemini-Vision-series (Gemini) and GPT-4-series (GPT-4) models have epitomized a paradigm shift in Artificial General Intelligence (AGI) for computer vision, showcasing their potential in the biomedical domain. In this study, we evaluated the performance of the Gemini, GPT-4, and 4 popular large models for an exhaustive evaluation across 14 medical imaging datasets, including 5 medical imaging categories (dermatology, radiology, dentistry, ophthalmology, and endoscopy), and 3 radiology report datasets. The investigated tasks encompass disease classification, lesion segmentation, anatomical localization, disease diagnosis, report generation, and lesion detection. Our experimental results demonstrated that Gemini-series models excelled in report generation and lesion detection but faces challenges in disease classification and anatomical localization. Conversely, GPT-series models exhibited proficiency in lesion segmentation and anatomical localization but encountered difficulties in disease diagnosis and lesion detection. Additionally, both the Gemini series and GPT series contain models that have demonstrated commendable generation efficiency. While both models hold promise in reducing physician workload, alleviating pressure on limited healthcare resources, and fostering collaboration between clinical practitioners and artificial intelligence technologies, substantial enhancements and comprehensive validations remain imperative before clinical deployment.

Harry Anthony, Konstantinos Kamnitsas

Reliable use of deep neural networks (DNNs) for medical image analysis requires methods to identify inputs that differ significantly from the training data, called out-of-distribution (OOD), to prevent erroneous predictions. OOD detection methods can be categorised as either confidence-based (using the model's output layer for OOD detection) or feature-based (not using the output layer). We created two new OOD benchmarks by dividing the D7P (dermatology) and BreastMNIST (ultrasound) datasets into subsets which either contain or don't contain an artefact (rulers or annotations respectively). Models were trained with artefact-free images, and images with the artefacts were used as OOD test sets. For each OOD image, we created a counterfactual by manually removing the artefact via image processing, to assess the artefact's impact on the model's predictions. We show that OOD artefacts can boost a model's softmax confidence in its predictions, due to correlations in training data among other factors. This contradicts the common assumption that OOD artefacts should lead to more uncertain outputs, an assumption on which most confidence-based methods rely. We use this to explain why feature-based methods (e.g. Mahalanobis score) typically have greater OOD detection performance than confidence-based methods (e.g. MCP). However, we also show that feature-based methods typically perform worse at distinguishing between inputs that lead to correct and incorrect predictions (for both OOD and ID data). Following from these insights, we argue that a combination of feature-based and confidence-based methods should be used within DNN pipelines to mitigate their respective weaknesses. These project's code and OOD benchmarks are available at: https://github.com/HarryAnthony/Evaluating_OOD_detection.

Richard Bindernagel, DO, MBA, Krina Chavda, DO

Jocić Ivana, Daković Dragana, Kandolf-Sekulović Lidija et al.

Introduction. Mucous membrane pemphigoid (MMP) is a rare autoimmune, chronic inflammatory disease that affects mucous membranes, most commonly the eyes and mouth, with or without skin involvement. It is a complex disease with several complications, including scarring, especially on conjunctival mucosa, that can lead to visual loss. Case report. We report four patients (two men and two women) with MMP. In all patients, the disease started between seventy and eighty years of age. The diagnosis was confirmed based on clinical appearance, histology, direct and indirect immunofluorescence studies, indirect split skin technique, and enzyme-linked immunosorbent assay (ELISA) test. The majority of lesions were on the gums and buccal mucosa; one patient had laryngeal involvement and a lesion on the umbilicus. No ocular involvement and no malignancy were detected. Direct immunofluorescence tests revealed continuous linear IgG deposition in the basal membrane zone in two patients, and they were treated with oral nicotinamide and tetracycline hydrochloride. In two patients, we detected IgG along with IgA linear deposition; they received treatment with methylprednisolone. Complete remission was achieved in all patients. Conclusion. Early diagnosis and an adequate therapeutic approach are necessary for the MMP treatment in long-term disease control and reduction of disease-related complications.

Aref Moshayedi, Stephen J. Malachowski, Justin Haught

Abstract Reactive infectious mucocutaneous eruption (RIME) is a newly proposed clinical entity characterised by post‐infectious mucositis involving two or more mucous membranes. The term expands the previously described Mycoplasma pneumoniae‐induced rash and mucositis to include additional infectious agents. We report a case of RIME secondary to SARS‐CoV2 and Influenza A coinfection with subsequent reactivation of varicella zoster virus on the lips. RIME can have significant clinical overlap with Stevens‐Johnson Syndrome and differentiation is key in limiting unnecessary future medication restrictions on patients. This report serves to increase awareness of RIME, including to coinfections and possible reactivation of Human Herpes Viruses.

Sepehr Sadafi, Ali Ebrahimi, Masoud Sadeghi et al.

Objective: Psoriasis is a disease with an immunogenetic background in which cytokines have important effects on its prevalence and incidence. The present meta-analysis evaluated the relationship between tumor necrosis factor-alpha (TNF-α) polymorphisms (rs361525, rs1800629, rs1799724, 1800630, and rs1799964) and psoriasis risk in studies following Hardy-Weinberg equilibrium (HWE). Materials and methods: Four databases were searched to retrieve relevant studies reporting the distributions of TNF-α polymorphisms in psoriasis cases compared to controls. The effect sizes were the 95% confidence intervals (CIs) and odds ratios (ORs). Subgroup analysis, sensitivity analyses, publication bias, trial sequential analysis (TSA), and meta-regression were performed on the initial pooled results of TNF-α polymorphisms. Results: Thirty-six articles with 71 studies were included in the meta-analysis (twenty-six: rs361525, twenty-seven: rs1800629, nine: rs1799724, four: 1800630, and five: rs1799964). The pooled ORs for −238 G/A rs361525 polymorphism were 2.33 (p < 0.00001), 2.79 (p < 0.0001), 2.35 (p < 0.00001), 2.44 (p < 0.00001), and 2.45 (p < 0.00001), as well as 1.57 (p < 0.00001), 1.98 (p = 0.01), 1.61 (p < 0.00001), 1.64 (p < 0.00001), and 1.79 (p < 0.00001) for −857 C/T rs1799724 polymorphism in allelic, homozygous, heterozygous, dominant, and recessive models, respectively. Ethnicity, psoriasis type, and sample size affected the pooled results of rs361525, rs1800629, and rs1799724 polymorphisms. Based on TSA, there were just sufficient cases for −238 G/A rs361525 polymorphism in five genetic models and −857C/T rs1799724 polymorphism in allelic, heterozygous, and dominant models. Conclusions: The A allele and GA and GG genotypes of −238 G/A rs361525 polymorphism and T allele, TT and CT genotypes of −857C/T rs1799724 polymorphism were related to increased risks in psoriasis cases. Well-designed studies (with no deviation from HWE in controls) with more cases are recommended in the future.