Artificial intelligence-based prediction of diseases among homeless populations in Bogotá: Implications for targeted interventions
Hugo Ordoñez, Cristian Camilo Ordoñez, Juan-Sebastian Gonzalez-Sanabria · 2026
Read Artificial intelligence-based prediction of diseases among homeless populations in Bogotá: Implications for targeted interventions by Hugo Ordoñez, Cristian Camilo Ordoñez, Juan-Sebastian Gonzalez-Sanabria (2026) 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/plos-pone-0352268.txt.
Description
Hugo Ordoñez and two colleagues build an artificial intelligence model to predict disease among Bogotá's homeless residents. Their 2026 study uses the city's 2024 homeless census. Over ten thousand records carried forty-six demographic, clinical, and social variables. This population suffers heavy burdens of hypertension, diabetes, tuberculosis, and HIV. It is also the population surveillance systems see worst. A gradient boosting model was trained to flag disease occurrence, with explanation tools exposing its reasoning.
The model reached about ninety-one percent overall accuracy. Its sensitivity was more modest, catching a little over half of true cases. The balance struck a workable F1 score for screening purposes. Explanation values showed officials which factors drive each risk profile. Prediction is not the end goal here. Targeted outreach is: sending scarce clinics, testing, and treatment to the street locations where risk concentrates. For invisible populations, a defensible model beats no map at all.
Catalog record
| CREATOR | Hugo Ordoñez, Cristian Camilo Ordoñez, Juan-Sebastian Gonzalez-Sanabria |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 66445 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |