Deep learning–assisted malaria microscopy with sensitivity-aware threshold optimization
Fatma Dehbi, Reda Mohamed Hamou, Menaouer Brahami, Sahmoud Shaaban, Abdelhamid Ghoul, Ameur Latreche, Selman Djeffal · 2026
Read Deep learning–assisted malaria microscopy with sensitivity-aware threshold optimization by Fatma Dehbi, Reda Mohamed Hamou, Menaouer Brahami, Sahmoud Shaaban, Abdelhamid Ghoul, Ameur Latreche, Selman Djeffal (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/pmc-13553222.txt.
Description
Fatma Dehbi and co-authors build an AI tool to read malaria blood slides, tuned to miss as few infected cells as possible. Missing a parasite cell can put a patient at risk, even when overall accuracy looks high. Their 2026 study trains two neural networks on a balanced public set of more than twenty-seven thousand cell images.
At the standard cutoff, the smaller network gave the best plain accuracy. After retuning the cutoff to prize recall, the deeper network caught ninety-seven percent of infected cells on held-out tests. Missed infected cells fell from one hundred seventy-seven to sixty-one. The authors warn that images, not patients, were split for testing, so clinic trials at many centers are still needed.
Catalog record
| CREATOR | Fatma Dehbi, Reda Mohamed Hamou, Menaouer Brahami, Sahmoud Shaaban, Abdelhamid Ghoul, Ameur Latreche, Selman Djeffal |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |