In the Name of Fairness: Assessing the Bias in Clinical Record De-identification. (arXiv:2305.11348v1 [cs.LG])

Data sharing is crucial for open science and reproducible research, but the
legal sharing of clinical data requires the removal of protected health
information from electronic health records. This process, known as
de-identification, is often achieved through the use of machine learning
algorithms by many commercial and open-source systems. While these systems have
shown compelling results on average, the variation in their performance across
different demographic groups has not been thoroughly examined. In this work, we
investigate the bias of …

algorithms bias commercial data data sharing electronic health records fairness health identification information legal machine machine learning machine learning algorithms name process protected health information research science sharing