A Machine Learning Approach for the Differential Diagnosis of Alzheimer and Vascular Dementia Fed by MRI Selected Features
Problem
Alzheimer's disease and vascular dementia share symptoms and MRI signs, making clinical differential diagnosis difficult and sometimes controversial.
Why it matters
Misdiagnosis between AD and VD leads to wrong treatment decisions for a growing dementia patient population.
What the AI did
Machine learning algorithms (ANN, SVM, ANFIS) combined with advanced MRI features were tested to classify VD versus AD and predict the prevalent disease in unclear cases.
Results
not stated in source
Impact
The study investigated ML as a supportive diagnostic tool, but the abstract gives no accuracy figures.
Record
| Venue | Frontiers in Neuroinformatics |
|---|---|
| Published | 2020 |
| Work type | article |
| Authors | Gloria Castellazzi, Maria Giovanna Cuzzoni, Matteo Cotta Ramusino, Daniele Martinelli, Federica M. Denaro, Antonio Ricciardi, Paolo Vitali, Nicoletta Anzalone et al. |
| Citations | 133 (OpenAlex, at import time) |
| References | 61 |
| Model or method | ANN, SVM, ANFIS |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3389/fninf.2020.00025 |
| Open access | gold |
| Keywords | Alzheimer disease, vascular dementia, MRI, machine learning, differential diagnosis, classification |
Importance rating
significant — thousands to millions affected
Differential diagnosis of AD vs VD affects treatment for a frequent dementia population; abstract calls them two of the most frequent dementia-like diseases but gives no numbers or accuracy results.