Unequal medicine harms: reflections on the experiences of an intersex physician

Predicting Patient Demographics From Chest Radiographs With Deep Learning

“…The accuracy of labels is challenging to quantify because a "gold standard" set of labels was not available, and although the labels are believed to be accurate, their accuracy cannot be guaranteed. Creating a system in which patients are assigned as either reportedly male or female does not allow for patients with different gender identities [28][29][30], in the process of transitioning identities, or with phenotypical features of both male and female traits [31]. Recent meta-analyses report approximately 0.37% of Americans as transgender and 0.02% to 1.7% as intersex [32,33].…”