Neural Imaging for Alzheimer’s Prediction using AI: Exploring CNNs and LIME Explanations
Abraham Varghese, Vinu Sherimon, Xavier C. Raja, Ben George Ephrem, Prasanth Gouda · 2025
Read Neural Imaging for Alzheimer’s Prediction using AI: Exploring CNNs and LIME Explanations by Abraham Varghese, Vinu Sherimon, Xavier C. Raja, Ben George Ephrem, Prasanth Gouda (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-11713524.txt.
Description
Abraham Varghese and four co-authors trained a neural network to predict Alzheimer's disease from brain scans, in a 2025 paper. Their data came from a large public imaging initiative, with scans sorted into classes from healthy cognition through stages of mild impairment to Alzheimer's. The team cleaned the images first, focusing on brain regions, then built a convolutional network with two feature layers and a stack of dense layers.
Training accuracy reached nearly ninety-nine percent, and a small test set of sixty-five images was classified without error. To open the model's reasoning, they applied LIME, a tool that highlights which image regions pushed each decision. The highlighted zones matched the predicted class in the healthy example they detail. The authors argue that pairing accuracy with such explanations is the path to trustworthy diagnostic AI.
Catalog record
| CREATOR | Abraham Varghese, Vinu Sherimon, Xavier C. Raja, Ben George Ephrem, Prasanth Gouda |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 3 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |