Dermoscopy of early-stage melanoma can be challenging, and repeated examination at three-month intervals may disclose subtle changes. In patients with atypical nevus syndrome or more than 50 nevi, repetitive excision of benign lesions does not guarantee that melanomas will be identified at an early stage and exposes patients to potentially disfiguring surgery. We present the case of a high-risk patient where repeated dermoscopy of an in situ melanoma showed that part of the pigment network had coarsened, even though the lesion had not changed macroscopically.
Download full-text PDF |
Source |
|---|
Publication Analysis
Top Keywords
Similar Publications
Assessment of image quality on the diagnostic performance of clinicians and deep learning models: Cross-sectional comparative reader study.
J Eur Acad Dermatol Venereol
December 2024
School of Public Health and Preventive Medicine, Monash University, Melbourne, Victoria, Australia.
Article Synopsis
- Skin cancer diagnosis relies heavily on image quality, particularly the impact of HDR-enhanced images, on clinicians' and AI models' accuracy.
- A study with 18 dermatology clinicians assessed 303 skin lesion images classified into low, high, and enhanced quality, finding the best diagnostic performance with high-quality images.
- While clinicians outperformed a CNN model in diagnosing low and enhanced quality images, the CNN excelled in multiclass classification with high-quality images, showing the varying effectiveness of image quality in diagnostics.
Comprehensive assessment of intradermal responses to hyaluronic acid-based skin injection fillers through multi-pathway dynamic synergies.
Colloids Surf B Biointerfaces
February 2025
National Engineering Research Center for Biomaterials, Sichuan University, Chengdu, Sichuan 610064, China; College of Biomedical Engineering, Sichuan University, Chengdu 610064, China.
Article Synopsis
- Skin injection fillers need thorough evaluation for market entry, focusing on intradermal reactions to ensure safety and compatibility.
- The study introduces a detailed evaluation method, integrating various indicators like dermoscopy, moisture content, and ultrasound to assess the effects of sodium hyaluronate gel.
- Findings reveal that while the gel showed minimal irritation and acceptable safety levels, existing assessment standards might not adequately measure its long-term effects, suggesting the need for improved evaluation methods.
MASDF-Net: A Multi-Attention Codec Network with Selective and Dynamic Fusion for Skin Lesion Segmentation.
Sensors (Basel)
August 2024
School of Electronics and Information Engineering, Sichuan University, Chengdu 610065, China.
Automated segmentation algorithms for dermoscopic images serve as effective tools that assist dermatologists in clinical diagnosis. While existing deep learning-based skin lesion segmentation algorithms have achieved certain success, challenges remain in accurately delineating the boundaries of lesion regions in dermoscopic images with irregular shapes, blurry edges, and occlusions by artifacts. To address these issues, a multi-attention codec network with selective and dynamic fusion (MASDF-Net) is proposed for skin lesion segmentation in this study.
View Article and Find Full Text PDFDynamic optical coherence tomography unveils subclinical, vascular differences across actinic keratosis grades I-III.
Exp Dermatol
August 2024
Department of Dermatology, Copenhagen University Hospital, Copenhagen, Denmark.
Article Synopsis
- * The study involved 47 patients, revealing that D-OCT identified distinct vascular differences between AK grades and photodamaged skin, especially at a depth of 300μm, and showed significant differences in vessel arrangement and patterns across the various grades of AK.
- * The combined use of D-OCT and dermatoscopy proved effective in differentiating between AK grades, with results indicating increased vascularization and vessel disorganization in higher-grade AK lesions.
Adaptive neighborhood triplet loss: enhanced segmentation of dermoscopy datasets by mining pixel information.
Int J Comput Assist Radiol Surg
August 2024
Fraunhofer Institute for Toxicology and Experimental Medicine, Hannover, Germany.
Purpose: The integration of deep learning in image segmentation technology markedly improves the automation capabilities of medical diagnostic systems, reducing the dependence on the clinical expertise of medical professionals. However, the accuracy of image segmentation is still impacted by various interference factors encountered during image acquisition.
Methods: To address this challenge, this paper proposes a loss function designed to mine specific pixel information which dynamically changes during training process.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!