An AI-Driven Diagnostic System for Early Skin Disease Detection

Skin diseases are among the most prevalent health conditions worldwide, necessitating timely and accurate diagnostic tools. This paper presents a comprehensive AI-driven diagnostic system for early skin disease detection, implemented as a web-based application (Ai Skin) that leverages the OpenAI GPT-4o multimodal vision model. The proposed system supports two primary analysis modes: single-image skin analysis and dual-image comparative analysis for tracking disease progression. It integrates structured prompt engineering to extract clinically meaningful outputs including disease classification, severity scoring, skin type assessment, and contextual diagnostic recommendations. To evaluate system performance, controlled experiments were conducted using 200 annotated skin images across five disease categories (Acne, Eczema, Psoriasis, Rosacea, and Fungal Infection), sourced from the publicly available HAM10000 and DermNet datasets. The system achieved an overall accuracy of 91.5%, with a macro-average precision of 90.8%, recall of 89.6%, and F1-score of 90.2%. These results underscore the effectiveness of vision-language models in supporting dermatological decision-making and highlight the practical utility of the Ai Skin system as a non-invasive, accessible screening tool.

Artificial Intelligence; Skin Disease Detection; Deep Learning; Vision-Language Models

1. World Health Organization, "Skin Diseases," Global Health Estimates, 2022. [Online]. Available: https://www.who.int [Google Scholar] [Crossref]

2. Kassi, F. Diallo, and K. Ouattara, "Barriers to dermatological care in sub-Saharan Africa: A systematic review," International Journal of Dermatology, vol. 59, no. 4, pp. 401–409, 2020. [Google Scholar] [Crossref]

3. OpenAI, "GPT-4 Technical Report," arXiv preprint arXiv:2303.08774, 2023. [Google Scholar] [Crossref]

4. M. Combalia et al., "BCN20000: Dermoscopic lesions in the wild," arXiv preprint arXiv:1908.02288, 2019. [Google Scholar] [Crossref]

5. V. Rotemberg et al., "A patient-centric dataset of images and metadata for identifying melanomas using clinical context," Scientific Data, vol. 8, no. 1, p. 34, 2021. [Google Scholar] [Crossref]

6. Esteva, B. Kuprel, R. A. Novoa, J. Ko, S. M. Swetter, H. M. Blau, and S. Thrun, "Dermatologist-level classification of skin cancer with deep neural networks," Nature, vol. 542, no. 7639, pp. 115–118, 2017. [Google Scholar] [Crossref]

7. S. S. Han, I. Park, S. E. Chang, W. Lim, M. S. Kim, G. H. Park, and J. H. Kim, "Deep neural networks show an equivalent and often superior performance to dermatologists in onychomycosis diagnosis," Journal of Investigative Dermatology, vol. 138, no. 10, pp. 2200–2206, 2018. [Google Scholar] [Crossref]

8. T. J. Brinker, A. Hekler, A. H. Enk, J. Klode, A. Hauschild, C. Berking, and J. S. Utikal, "Deep learning outperformed 136 of 157 dermatologists in a head-to-head dermoscopic melanoma image classification task," European Journal of Cancer, vol. 113, pp. 47–54, 2019. [Google Scholar] [Crossref]

9. P. Tschandl, C. Rosendahl, and H. Kittler, "The HAM10000 dataset: A large collection of multi-source dermatoscopic images of common pigmented skin lesions," Scientific Data, vol. 5, Art. no. 180161, 2018. [Google Scholar] [Crossref]

10. DermNet NZ, "DermNet: The Dermatology Resource," 2024. [Online]. Available: https://dermnetnz.org [Google Scholar] [Crossref]

11. Dosovitskiy et al., "An image is worth 16x16 words: Transformers for image recognition at scale," in Proc. ICLR, 2021. [Google Scholar] [Crossref]

12. J. Kawahara, S. Daneshvar, G. Argenziano, and G. Hamarneh, "Seven-point checklist and skin lesion classification using multitask multimodal neural networks," IEEE Journal of Biomedical and Health Informatics, vol. 23, no. 2, pp. 538–546, 2019. [Google Scholar] [Crossref]

• The Impact of Ownership Structure on Dividend Payout Policy of Listed Plantation Companies in Sri Lanka
• Urban Sustainability in North-East India: A Study through the lens of NER-SDG index
• Performance Assessment of Predictive Forecasting Techniques for Enhancing Hospital Supply Chain Efficiency in Healthcare Logistics
• The Fractured Self in Julian Barnes' Postmodern Fiction: Identity Crisis and Deflation in Metroland and the Sense of an Ending
• Impact of Flood on the Employment, Labour Productivity and Migration of Agricultural Labour in North Bihar