Evaluating Sociodemographic Bias in an Artificial Intelligence Algorithm to Detect Cognitive Impairment in Electronic Health Records
Colin G Magdamo, Yignan He, John R Dickson, Tanish Tyagi, M Brandon Westover, Shibani Mukerji, Christine S Ritchie, Bradley T T Hyman, Deborah Blacker, Sudeshna Das · 2025
Read Evaluating Sociodemographic Bias in an Artificial Intelligence Algorithm to Detect Cognitive Impairment in Electronic Health Records by Colin G Magdamo, Yignan He, John R Dickson, Tanish Tyagi, M Brandon Westover, Shibani Mukerji, Christine S Ritchie, Bradley T T Hyman, Deborah Blacker, Sudeshna Das (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11712252.txt.
Description
Memory and thinking problems often go undiagnosed, which delays help and hides the true scale of dementia. Colin Magdamo and co-authors built a screening program that reads health records to flag likely hidden cases in this 2025 study. They then checked whether the program treated social groups fairly.
The program first uses a large language model to turn doctors free-text notes into coded features, then a second model scores each patient. It was tested on three years of records from front-line care patients, split by social traits such as age, sex, and background. Fairness checks matter because a biased tool could widen the very gaps it aims to close. The paper shows how record-based screening can be audited group by group before wide use.
Catalog record
| CREATOR | Colin G Magdamo, Yignan He, John R Dickson, Tanish Tyagi, M Brandon Westover, Shibani Mukerji, Christine S Ritchie, Bradley T T Hyman, Deborah Blacker, Sudeshna Das |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 3 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |