Artifical Intelligence Working Group

Artificial Intelligence Working Group

Goal

The AI working group's goal is to promote the development of accurate and effective computerized systems to aid melanoma diagnosis and clinical decision support by creating resources for and engaging the dermatology and computer science communities.

We engage our stakeholder communities through meetings, publications and conferences. We also host AI grand and live challenges.

Accomplishments

Our primary focus is the development of the ISIC Archive, a large and expanding open source public access archive of annotated skin images. The Archive currently contains over 100,000 images.
To date we have hosted 5 ISIC challenges, which engaged thousands of participants from the computer science communities.
The ISIC workshops and conferences have achieved significant participation and recognition and have included the most prestigious and prominent computer science and dermatology conferences across the globe.
Based on our AI challenges, over 1,000 scientific papers were written by ISIC members and other inspired individuals from various communities.

Publications our members were involved in include:

Codella, Noel CF, et al. "Skin lesion analysis toward melanoma detection: A challenge at the 2017 international symposium on biomedical imaging (isbi), hosted by the international skin imaging collaboration (isic)." 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018). IEEE, 2018.

Gutman, David, et al. "Skin lesion analysis toward melanoma detection: A challenge at the international symposium on biomedical imaging (ISBI) 2016, hosted by the international skin imaging collaboration (ISIC)." arXiv preprint arXiv:1605.01397 (2016).

Codella, Noel, et al. "Skin lesion analysis toward melanoma detection 2018: A challenge hosted by the international skin imaging collaboration (isic)." arXiv preprint arXiv:1902.03368 (2019).

Veronica, Rotemberg, et al. "A patient-centric dataset of images and metadata for identifying melanomas using clinical context." Scientific Data 8.1 (2021).

Marchetti, Michael A., et al. "Results of the 2016 International Skin Imaging Collaboration International Symposium on Biomedical Imaging challenge: Comparison of the accuracy of computer algorithms to dermatologists for the diagnosis of melanoma from dermoscopic images." Journal of the American Academy of Dermatology 78.2 (2018): 270-277.

Tschandl, Philipp, et al. "Comparison of the accuracy of human readers versus machine-learning algorithms for pigmented skin lesion classification: an open, web-based, international, diagnostic study." The Lancet Oncology 20.7 (2019): 938-947.

Tschandl, Philipp, et al. "Human–computer collaboration for skin cancer recognition." Nature Medicine 26.8 (2020): 1229-1234.

Rotemberg, Veronica, et al. "The role of public challenges and data sets towards algorithm development, trust, and use in clinical practice." Seminars in cutaneous medicine and surgery. Vol. 38. No. 1. 2019.

Next Steps

Continue to develop additional functionality in the Archive platform and scale the size and complexity of the Archive.
Add clinical (standard photography) close-up images to the current collection of mostly dermoscopic images and increase the ethnic/racial and diagnosis diversity of the Archive.
Continue to crowd source the annotations of the images in the Archive for dermoscopy attributes, in partnership with the Education Working Group and with Centaur Labs.
Support and host Grand Challenges and Live Challenges.
- Efforts are underway to assemble an annotated collection of images of lesions captured sequentially to permit inclusion of change-over-time as an additional diagnostic variable.
Explore multi-modal images (e.g., dermoscopy together with clinical +/- RCM +/- histology).
Engage the relevant communities with conferences and publications.
- Among these are an upcoming workshop at CVPR and organizing special issues on skin imaging for Elsevier's Medical Image Analysis (MIA) journal and the Journal of Biomedical and Health Informatics.

Working Group Members

Leader

Veronica Rotemberg, MD, PhD

Memorial Sloan Kettering Cancer Center

New York, NY, USA