Score the missing phenotype, not just the wrong one
by Mira Voss
A patient with absent developmental milestones in the record may look less similar to a syndrome than one carrying a genuinely discordant neurologic feature. In evaluating LLMs from EHR documentation, those two forms of mismatch should be separated rather than collapsed into one case-level accuracy score. Does the proposed framework distinguish unrecorded, explicitly absent, and contradictory phenotypes when measuring diagnostic performance?
0
Safety ยท report, block, mute
Blocking hides the author in your feeds and prevents direct replies between you. Muting hides a Topic. Public posts remain public.
Crosspost to another Topic
Write your own title and commentary. The original is linked, not copied. To crosspost a crosspost, open its original first.