Machine learning based model for the detection of multiple sclerosis from OCT-derived macular and optic disc retinal biomarkers
Muhmmad Aqib Anwar, Ghassan Mohamedsalih, Fatema Al Mannaei, Anant Pai, Tanvir Alam · 2026
Read Machine learning based model for the detection of multiple sclerosis from OCT-derived macular and optic disc retinal biomarkers by Muhmmad Aqib Anwar, Ghassan Mohamedsalih, Fatema Al Mannaei, Anant Pai, Tanvir Alam (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-13466523.txt.
Description
Anwar and colleagues trained a machine learning system in 2026 to spot multiple sclerosis from eye scans alone. MS damages nerves body-wide, and the retina thins measurably as it progresses. The team measured two retinal layers around the macula and optic disc in both eyes. Scans came from 74 patients with MS and 44 healthy controls. Models were tuned per patient, and explanation tools showed which features drove each call.
One layer alone performed strongly. Ganglion cell layer sectors reached an F1 score, a combined accuracy measure, of about 94 percent. The nerve fiber layer alone scored lower, near 88 percent. Combining both layers gave the best result: an F1 near 96 percent with precision above 97 percent. Every measured sector of both layers was thinner in MS patients, a highly significant pattern. The authors argue eye scans could offer a cheap, painless screening route for MS.
Catalog record
| CREATOR | Muhmmad Aqib Anwar, Ghassan Mohamedsalih, Fatema Al Mannaei, Anant Pai, Tanvir Alam |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 88 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |