Researchers from IBM and Pfizer have published details on a new AI model that interprets written speech, which they claim can predict whether a person will develop Alzheimer's seven years before they show symptoms. The idea is attractive for its simplicity: The model's only input is a written sample from the "cookie-theft picture description task," a common cognitive test that asks participants to describe what's happening in a drawing (three guesses what the drawing is of). Researchers trained the AI to pore over participants' responses, picking up on hints of cognitive decline like repetition, misspellings, two-word sentences, and limited vocabulary. Now, hold your applause: The model is in early days, and it isn't any better than current cognitive assessments. The initial study--based on data gathered from just 270 Americans over the course of four decades--showed the AI could predict a future Alzheimer's diagnosis 70% of the time.

The types of information assessed by the artificial intelligence includes delays in responses, intonation, the percentage of verbs and nouns used, voice articulation and recall. Also looked at is the'gaze' of the patient, in terms of how attentive they are. These algorithms were developed based on an assessment of features of speech, language, and faces taken from recorded dialogues with elderly participants. Some of the participants had dementia, at various stages, and some did not. In trials, the artificial intelligence has successfully learnt how to differentiate for dementia without need to be programmed, at a rate of 90 percent success based on asking just six questions. The typical time taken for the assessment was just two to three minutes. Commenting on this, lead researcher Takashi Kudo stated, as quoted by European Pharmaceutical review: "If this technology is further developed, it will become possible to know whether or not an elderly individual is in the early stages of dementia through conversations with computer avatars at home on a daily basis." The test has been reported to the journal IEEE Journal of Translational Engineering in Health and Medicine. The research is titled "Detecting Dementia through Interactive Computer Avatars."