Alzheimer's Disease Prediction

Artificial Intelligence is now helping to predict the eventual onset of Alzheimer's disease in healthy people. The research conducted by IBM Research and Pfizer uses short standardised speech tests with better results than current methods. 

Alzheimer's is a disease of the brain. Those afflicted by it, typically people over 60 years of age, slowly lose their ability to think, ability to recall from memory and eventually the ability to carry out simple tasks.

Prediction of the onset of the disease can lead to the development of a simple, straightforward and accessible metric to help clinicians assess the risk of Alzheimer’s disease in an individual, leading to earlier intervention. There is no effective cure or prevention of this disease but the best way to delay onset and slow progression is to intervene early which may be possible one day. 

AI techniques were used to train models using short language samples from the Framingham Heart Study. Sample were selected based on age-groups and labelled based on whether those samples before 85 years were eventually diagnosed with Alzheimer's or not. Psycholinguistic Analyses and Non-Linguistic Variables were used in a multi-model approach. For training predictive models, the researchers experimented with linear SVM, logistic regression and Naïve Bayes classifiers. The hyperparameters of the classifiers were set using nested cross-validation.

Here's a short video describing the research:

The research was much more involved and detailed than has been described above. To get the full depth of the study and understand this advancement look inside the research paper for this advancement.