09.14.17

Voice Analysis to Detect Alzheimer’s Disease within Seconds

Meet Ludwig, a two-foot-tall robot that can determine whether people are showing symptoms of Alzheimer’s disease in less than one minute simply by listening to their voices. Developed by Canadian firm WinterLight Labs, Ludwig uses proprietary voice analysis algorithms to pick up on subtle cues that reveal signs of Alzheimer’s, as reported by The Telegraph. The company claims its solution operates at an accuracy rate between 82 percent and 100 percent.

“Our platform can analyze natural speech to detect and monitor dementia, aphasia, and various cognitive conditions,” WinterLight stated on its website. “Using a short one-minute sample of speech, WinterLight can characterize the speaker’s cognitive, acoustic and linguistic state, including lexical diversity, syntactic complexity, semantic content, and articulation.”

WinterLight tests patients by showing them a photo of a familiar subject and then having them describe what they are seeing for 45 seconds. A recording of the patient’s comments is then sent to a cloud-based system that automatically transcribes their words. Based on this transcription, WinterLight’s algorithms evaluate 400 distinct variables to evaluate speech, language, and cognition.

Based on this evaluation, the voice analysis system generates a mini-mental state examination (MMSE) score, which is a commonly used test for screening cognitive function.

The technology is attuned to detecting telltale signs of cognitive impairment in speech.

“People with Alzheimer’s have word finding difficulties which results in longer pauses between words, more hesitation, using more pronouns and fewer nouns (e.g. “her” instead of “Rita”) and using less complex words (e.g. “car” instead of “sedan”),” WinterLight noted. “In contrast, a person with Parkinson’s disease or Multiple Sclerosis sounds different but might use the same language as a neurologically healthy senior.”

The use of Ludwig the robot apparently provides a friendlier way to interact with the company’s algorithms, compared to a more utilitarian interface, such as a speaker and microphone. Ludwig resembles a ventriloquist’s dummy, with a human-like visage that’s easy to relate to.

WinterLight’s technology promises to provide quicker diagnoses of Alzheimer’s, potentially leading to faster detection and more effective treatment of a condition that affects millions. More than 5 million Americans now live with Alzheimer’s, a total that can rise to more than 16 million by 2050, according to the Alzheimer’s Association.

Total medical payments in 2017 for all individuals with Alzheimer’s and other forms of dementia are estimated at $259 billion, the association reports. Early diagnosis and treatment of Alzheimer’s can prevent an irrevocable disability.

Tyler Schulze is vice president, strategy & development at Veritone.  He serves as general manager for developer partnerships, cognitive engine ecosystem, and media ingestion for the Veritone platform. Learn more about our platform and join the Veritone developer ecosystem today.