New Speech-Analysis Methods for Dementia Detection
Xiaohui Liang, an assistant professor of computer science at the University of Massachusetts Boston, discussed the study’s significance in a news release.1 Dr. Liang, who will lead the study, said the research team is tackling a significant and complicated data-science question: Whether gathering long-term speech patterns of individuals at home will help scientists to develop new speech-analysis methods for early detection of dementia. “Our team envisions that the changes in the speech patterns of individuals using the voice assistant systems may be sensitive to their decline in memory and function over time,” Dr. Liang explained. Past research has shown that memory loss is not the only symptom of dementia, and speech changes and loss can be a strong indicator. A large University of Wisconsin-Madison study has found that using pauses and filler words may be an early indicator of cognitive decline.2Solving the Problem of Late Dementia Diagnosis
In the upcoming four-year study, researchers will use machine and deep learning techniques to collect data on patient participants. The grant features an 18-month laboratory evaluation and a 28-month home evaluation to determine whether the voice assistant systems can indeed measure and predict an individual’s decline in a home setting. John Batsis, a geriatrician and an associate professor of medicine at the Geisel School of Medicine at Dartmouth, outlined the study’s possible outcome in an interview.3 Dr. Batsis said that ideally, if their study was successful, home-based tools such as Alexa and Google Home could help identify speech patterns associated with dementia. The researchers hope to develop a system that could be placed into people’s homes. Ultimately, the technology could help track patients and their caregivers, who may have concerns about their loved ones’ cognitive health. The researchers admit there may be challenges in developing a system that people feel comfortable using in their homes. Will these home-based voice assistants replace current clinical evaluations? No, but they could aid the push for earlier detection. Dr. Batsis says one of the biggest challenges in dementia is late diagnosis. “Our goal is to bring that timeline forward a bit, to give families opportunities to plan and think ahead of all the issues while the patient is still able to make decisions, and that’s really key.”Innovations On the Horizon
The field of “voice-first” technology is still pretty new, but now third-party developers are getting on the bandwagon. Here are just a few of the innovations on the horizon… Smart home technology – such as controlling lights and thermostats – can give people, especially those with mobility issues, a greater sense of control and independence. Another smart speaker app can send alerts to designated people. The product’s creator says too often people don’t call 911 until it’s too late. This technology is a way for people to reach out if they are scared or worried. I also found another nifty app that uses voice-recognition technology to help older adults follow their routines and offer health and wellness reminders including medication, hydration and exercise prompts. For more information on these devices, go to this site. I won’t lie, I have a love-hate relationship with technology. Still, I believe the four-year study could offer some help for earlier diagnosis of dementia. I’m also eager to learn more about technologies that can help older people age in their own homes and stay healthy.