Artificial Intelligence in Dermatology

A topical collection in Dermato (ISSN 2673-6179).

Viewed by 9689

Editor


E-Mail Website
Guest Editor
Department of Dermatology and Allergy, University Hospital Basel, Basel, Switzerland
Interests: machine learning; telemedicine; inflammation; psoriasis; organoids; neutrophils; genetics

Topical Collection Information

Our new journal “Dermato” is proud to announce a Topical Collection on “Artificial Intelligence in Dermatology”. Much has been written about the subject in recent years, and new advances have been made that seemingly make it possible to replace the diagnostic powers of dermatologists in the near future. Many of us are wondering, however, how long this will take. As of today, not a single diagnostic AI algorithm has been approved by the FDA in dermatology, in contrast to radiology and cardiology that have double-digit numbers of such approvals already. In this issue, we are going to explore the challenges of dermatology to reach the threshold of having clinically useful algorithms.

One important difference between dermatology and radiology or cardiology is the lack of standardized data sources. Radiology has well-standardized protocols with patient posture, time of exposure, etc. that allow for reproducible images all over the world. The dermatologist, however, can make a clinical photograph any way they prefer it. This matters little for communication with human colleagues, but it is a great difficulty for training machine learning algorithms. One important exception is dermoscopy; hence, it may not surprise that the greatest AI-based classification advances in dermatology were made with dermoscopic images. This issue will also demonstrate how AI can, even now, be used as decision support tool for clinicians. As active dermatologists, we may not embrace being replaced completely by a machine―but additional support of our diagnostic capabilities will surely come in handy. Indeed, some clinical tools are offering a few such features already.

The centerpiece of this Topical Collection will be general clinical advances with AI in dermatology, and we have collected a fine selection of articles that demonstrate feasible approaches of integrating trained algorithms for the benefit of our patients. 

Prof. Dr. Alexander Navarini-Meury
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the collection website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Dermato is an international peer-reviewed open access quarterly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1000 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (3 papers)

2024

Jump to: 2022

12 pages, 2384 KiB  
Article
Dermatological Knowledge and Image Analysis Performance of Large Language Models Based on Specialty Certificate Examination in Dermatology
by Ka Siu Fan and Ka Hay Fan
Dermato 2024, 4(4), 124-135; https://doi.org/10.3390/dermato4040013 - 30 Sep 2024
Viewed by 694
Abstract
Large language models (LLMs) are trained using large datasets and may be applied to language-based tasks. Studies have demonstrated their ability to perform and pass postgraduate medical examinations, and with the increasingly sophisticated deep learning algorithms and incorporation of image-analysis capabilities, they may [...] Read more.
Large language models (LLMs) are trained using large datasets and may be applied to language-based tasks. Studies have demonstrated their ability to perform and pass postgraduate medical examinations, and with the increasingly sophisticated deep learning algorithms and incorporation of image-analysis capabilities, they may also be applied to the Specialty Certificate Examination (SCE) in Dermatology. The Dermatology SCE sample questions were used to assess the performance of five freely available and high-performance LLMs. The LLMs’ performances were recorded by comparing their output on multiple-choice questions against the sample answers. One hundred questions, four of which included photographs, were entered into the LLMs. The responses were recorded and analysed, with the pass mark set at 77%. The accuracies for Claude-3.5 Sonnet, Copilot, Gemini, ChatGPT-4o, and Perplexity were 87, 88, 75, 90, and 87, respectively (p = 0.023). The LLMs were generally capable of interpreting and providing reasoned responses to clinical scenarios and clinical data. This continues to demonstrate the potential of LLMs in both medical education and clinical settings. Full article
Show Figures

Figure 1

15 pages, 5539 KiB  
Article
Development of an AI-Based Skin Cancer Recognition Model and Its Application in Enabling Patients to Self-Triage Their Lesions with Smartphone Pictures
by Aline Lissa Okita, Raquel Machado de Sousa, Eddy Jens Rivero-Zavala, Karina Lumy Okita, Luisa Juliatto Molina Tinoco, Luis Eduardo Pedigoni Bulisani and Andre Pires dos Santos
Dermato 2024, 4(3), 97-111; https://doi.org/10.3390/dermato4030011 - 16 Aug 2024
Viewed by 1067
Abstract
Artificial intelligence (AI) based on convolutional neural networks (CNNs) has recently made great advances in dermatology with respect to the classification and malignancy prediction of skin diseases. In this article, we demonstrate how we have used a similar technique to build a mobile [...] Read more.
Artificial intelligence (AI) based on convolutional neural networks (CNNs) has recently made great advances in dermatology with respect to the classification and malignancy prediction of skin diseases. In this article, we demonstrate how we have used a similar technique to build a mobile application to classify skin diseases captured by patients with their personal smartphone cameras. We used a CNN classifier to distinguish four subtypes of dermatological diseases the patients might have (“pigmentation changes and superficial infections”, “inflammatory diseases and eczemas”, “benign tumors, cysts, scars and callous”, and “suspected lesions”) and their severity in terms of morbidity and mortality risks, as well as the kind of medical consultation the patient should seek. The dataset used in this research was collected by the Department of Telemedicine of Albert Einstein Hospital in Sao Paulo and consisted of 146.277 skin images. In this paper, we show that our CNN models with an overall average classification accuracy of 79% and a sensibility of above 80% implemented in personal smartphones have the potential to lower the frequency of skin diseases and serve as an advanced tracking tool for a patient’s skin-lesion history. Full article
Show Figures

Figure 1

2022

Jump to: 2024

7 pages, 906 KiB  
Review
Augmented and Virtual Reality in Dermatology—Where Do We Stand and What Comes Next?
by Mathias Bonmarin, Severin Läuchli and Alexander Navarini
Dermato 2022, 2(1), 1-7; https://doi.org/10.3390/dermato2010001 - 25 Jan 2022
Cited by 6 | Viewed by 6325
Abstract
As the skin is an accessible organ and many dermatological diagnostics still rely on the visual examination and palpation of the lesions, dermatology could be dramatically impacted by augmented and virtual reality technologies. If the emergence of such tools raised enormous interest in [...] Read more.
As the skin is an accessible organ and many dermatological diagnostics still rely on the visual examination and palpation of the lesions, dermatology could be dramatically impacted by augmented and virtual reality technologies. If the emergence of such tools raised enormous interest in the dermatological community, we must admit that augmented and virtual reality have not experienced the same breakthrough in dermatology as they have in surgery. In this article, we investigate the status of such technologies in dermatology and review their current use in education, diagnostics, and dermatologic surgery; additionally, we try to predict how it might evolve in the near future. Full article
Show Figures

Figure 1

Back to TopTop