Artificial intelligence on inflammatory dermatoses: where we are and where are we going?

BACKGROUND

Artificial intelligence (AI) is increasingly gaining ground in dermatology, with studies reporting accuracy equal to or greater than dermatologists for the diagnosis of skin lesions from clinical and dermoscopic images.1 AI has been developed and improved constantly for dermatology, however, the focus has been much more on neoplastic diseases, due to their high prevalence and high morbidity.

OBJECTIVES

Describe the possible applications of AI in inflammatory dermatoses.

METHODS

Articles published between 2013 and 2023 in Medline and Lilacs were retrieved after applying the inclusion and exclusion criteria 19 articles were selected. From each selected article, the necessary information was extracted and with this data, the present review was written.

RESULTS

The first studies on AI in dermatology focused on the diagnosis of neoplasms, especially melanoma, due to the ease of standardization of images, obtaining accuracy equivalent to that of a dermatologist in clinical and dermoscopic lesions. Actually, there are many studies on artificial intelligence in inflammatory dermatosis, such as psoriasis, helping to calculate the PASI, hidradenitis suppurativa, and atopic dematitis.

STUDY LIMITATIONS: The limitation of the study is that it is a literature review and because it is an innovative topic with a limited number of studies published in the literature.

CONCLUSIONS

Considerable of what is published in the literature is in computer science journals, but it is possible to perceive that there is an important interest in the area and that artificial intelligence will advance to assist dermatologists.