GitHub Repository
Created On
Updated On
Demo for skin lesions detection using deep learning CNN model and explain predictions using LLMs for end users. The CNN model trained on MSLD v2.0 dataset for detecting: Monkeypox, Chickenpox, Cowpox, HFMD, Measles and healthy. This app demonstrates the potential of using LLMs to explain CNN analysis results in textual context.

More details

Use Cases Limitations Evidence Owner's Insight
  • This app can be used in English and French, can be used for detecting viral infectious skin lesions as Monkeypox, Chickenpox, Cowpox, HFMD, Measles and Healthy cases.
  • Integrate LLMs such as Llama-3 70B, Llama-3 8B, Mixtral 8x7B to generate detailed information about the skin lesion.

The current model is designed to classify six distinct categories: Monkeypox, Chickenpox, Cowpox, HFMD, Measles, and Healthy. We are actively exploring opportunities to enhance the model’s capabilities by incorporating additional categories in the future.

This application has not undergone real clinical trials and should not be used as a definitive diagnosis. Always consult a medical professional for accurate advice and assessment.

Ali, S. N., Ahmed, M. T., Jahan, T., Paul, J., Sani, Noor, N., Asma, A. N., & Hasan, T. (2023). A Web-based Mpox Skin Lesion Detection System Using State-of-the-art Deep Learning Models Considering Racial Diversity. ArXiv (Cornell University).



Warning: Not intended for clinical use. Assume outputs are unsafe and unvalidated. Use carefully.

  • Favorites: 0
  • Executions: 33

  • Diagnostics & Imaging


A Ahmed AL-Rufai

Member since