Automated Image Analysis
We recently published a joint work with the Department of Rheumatology, University Hospital of Zurich, the Department of Quantitative Biomedicine, and the Department of Rheumatology and Immunology, University Hospital Bern. The study assessed the capacity of AI to detect patterns of microangiopathy in nailfold capillaroscopy (NFC) images of patients with Systemic Sclerosis (SSc). We used local registers from the University Hospital Zurich and ‘off-the-shelf’ artificial intelligence models (ViT).
In their editorial Maurizio Cutolo, Emanuele Gotelli and Vanessa Smith conclude, “the ViT is welcome and seems to represent a further valid system for an early and fast reading of the NFC images/morphological biomarkers in SSc, reaching for the first time the fusion of EULAR-validated algorithms for the delineation of the scleroderma pattern from the non-scleroderma pattern and artificial intelligence.”
Based on our recent publication we have developed a presentation where you
- get a overview about SSc and the classification of the disease,
- can check your skills to classify capillaroscopy images,
- can play with an AI simulator.
Alexandru Garaiman, Farhad Nooralahzadeh, Carina Mihai, Nicolas Perez Gonzalez, Nikitas Gkikopoulos, Mike Oliver Becker, Oliver Distler, Michael Krauthammer, Britta Maurer, Vision transformer assisting rheumatologists in screening for capillaroscopy changes in systemic sclerosis: an artificial intelligence model, Rheumatology, 2022, https://doi.org/10.1093/rheumatology/keac541