Online, 6/11/2023.
iToBoS project was presented at Fairness of AI in Medical Imaging, a Medical Imaging Workshop that took place virtually on 6th November 2023.
Research on fairness, equity, and accountability in the context of machine learning has extensively demonstrated ethical risks in the deployment of such systems, including for medical image analysis. In a series of interdisciplinary events, the aim of FAIMI workshops is to advance the discourse around fairness issues in the medical image analysis community.
Frederik Pahde was invited as a speaker to present the paper titled “Reveal to revise: An explainable ai life cycle for iterative bias correction of deep models”, published at MICCAI 2023. This work introduces a framework allowing for iterative bias correction and has been developed in the iToBoS project, to unlearn various biases caused by data artifacts present in public benchmark datasets for skin cancer detection (ISIC 2019), such as skin markers and band-aids.
Find out more at https://faimi-workshop.github.io/2023-online/