AI Diagnoses Alzheimer's With More Than 95 Percent Accuracy

Researchers from the Stevens Institute of Technology have developed an artificial intelligence (AI) tool that can diagnose Alzheimer's disease with more than 95 percent accuracy because it is able to explain its conclusions, reports Tech HQ . The algorithm is trained to identify subtle linguistic clues previously overlooked, by utilizing texts composed by both healthy subjects and known Alzheimer's patients. The researchers then rendered each sentence as a unique numerical sequence, or vector, representing a point in a 512-dimensional space. This enables complex sentences to be assigned a tangible numerical value, and simplifies analysis of structural and thematic relationships between sentences. The AI algorithm gradually learns to identify differences between sentences composed by either healthy or unhealthy individuals before ascertaining exactly how likely any given text was to have been produced by someone with Alzheimer's. "We're . . . making it far easier to explain to patients why the AI came to the conclusion that it did while diagnosing patients," said Stevens Institute of Artificial Intelligence founding director K.P. Subbalakshmi.