Spanish researchers have developed an artificial-intelligence system to help doctors diagnose Alzheimer’s disease.
The system, based on brain-scan analysis, can also detect when a person suffers from the mild cognitive impairment that can signal that a person has Alzheimer’s.
Doctors can use the tool to make earlier diagnoses. Alzheimer’s-related processes start decades before memory loss becomes apparent. Early diagnosis would give researchers a chance to study factors that could one day prevent the progression of the disease.
The study, “Ensembles of Deep Learning Architectures for the Early Diagnosis of the Alzheimer’s Disease,” was published in the International Journal of Neural Systems.
White matter consists of nerve fibers that connect different brain areas. Deterioration of the networks causes loss of brain function, including memory.
Researchers based their diagnosis system on deep learning architecture. The self-learning algorithms at the heart of the architecture identified brain regions and connections that were abnormal in Alzheimer’s, automatically extracting the most relevant characteristics of numerous images.
“The study uses deep learning techniques to calculate brain function predictors and magnetic resonance imaging to prevent Alzheimer’s disease. To do this, we have used different neural networks with which to model each region of the brain to combine them afterwards,” the researchers said in a press release.
To test the validity of their computer-aided diagnosis method, the researchers looked at a sizable chunk of information from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) database.
The system could distinguish between Alzheimer’s and cognitively healthy people with 90 percent accuracy, the team found. It could also identify people with mild cognitive impairment at a high level of accuracy. And it could discriminate between mild cognitive impairment and Alzheimer’s, and between mild cognitive impairment and the lack of impairment seen in healthy controls.
Researchers said the system could be tweaked to help diagnose other types of dementia as well, including Parkinson’s Disease.