TL;DR
Researchers at Mass General Brigham University of Health Professions are analyzing fine-grained speech features to see whether they can help identify Alzheimer’s risk before cognitive symptoms emerge. Preliminary results suggest the measures performed better than cognitive testing in cognitively unimpaired older adults, but the findings do not establish a validated prediction tool or show that speech changes diagnose dementia.
A research team at Mass General Brigham University of Health Professions is studying whether tiny changes in speech can help identify Alzheimer’s risk before cognitive symptoms appear. In preliminary results, the team said speech-related data performed better than cognitive testing at identifying risk among cognitively unimpaired older adults; the findings are not yet evidence that the approach can diagnose or reliably predict dementia.
Get workout gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The project measures about 130 speech-related features, including aspects of how participants move their mouths and produce sounds. In one session described in a Nov. 11 report by The New York Times, a 76-year-old participant repeated syllables such as “ga ba da” and “ka pa ta” while sensors attached to her tongue, forehead, chin, jaw and lips recorded muscle movements.
Researcher Marziye Eshghi and her team feed the resulting data into artificial intelligence algorithms and compare it with known biological risk factors for Alzheimer’s. The preliminary comparison with cognitive testing concerned identification of Alzheimer’s risk in older participants without evident cognitive impairment. The report did not provide sample size, accuracy figures, or details of how the risk designation was established.
The work focuses on detecting patterns that may be too subtle for a listener to notice. Eshghi said the team sees differences that people may not detect by ear, potentially years before cognitive decline. That is the researcher’s characterization of the findings, not confirmation that every such speech difference signals future dementia.
Speech Could Add an Early Risk Signal
If further research supports the early findings, speech measurements could offer another way to study changes associated with Alzheimer’s risk, including among people who do not yet show impairment on standard cognitive tests. The appeal is that speech can be recorded and analyzed for patterns beyond what a brief conversation or a person’s own recollection may reveal.
For readers, the distinction between a research signal and a clinical test matters. This project does not show that a change in someone’s speech means they will develop dementia, nor does it establish that the method is ready for routine screening. A tool that flags risk could have consequences for people’s health decisions and anxiety, so its accuracy and appropriate use would need to be established before it could guide care.
Products Worth Considering
Levoit EverestAir-P Smart Air Purifier for Large Rooms
Oral-B iO5 Rechargeable Electric Toothbrush with 5 Cleaning Modes and Travel Case, White
Oral-B iO Series 3 Rechargeable Electric Toothbrush for Adults, Black, with 2 Brush Heads and Travel Case
As an affiliate, we earn on qualifying purchases.
How the Speech Study Works
The researchers are examining speech and speech-related movements, rather than relying only on the meaning of what a participant says. Repeating short syllables at speed gives the team a controlled way to record details of speech production. Sensors capture movement; algorithms then analyze many features and compare the data with biological risk factors for Alzheimer’s.
The report also describes research interest in other movements connected with speaking, including chewing and swallowing. These are part of a broader effort to investigate whether small changes in movement and communication may reflect shifts in brain health. The available account does not establish that those other measures are part of Eshghi’s specific 130-feature analysis.
The reported result is an early comparison within the study, not a completed clinical validation. The source material does not include a full paper or enough methodological detail to independently assess the analysis. It also does not say that the team has tracked participants long enough to show that the measured speech patterns predict who later develops dementia.
“We are seeing differences that humans can’t see or can’t really detect by their ears, but it’s there years and years before cognitive decline.”
— Marziye Eshghi, the researcher leading the project
Prediction Claims Need Testing
Several details needed to judge the result are not available in the report excerpt. These include the number and characteristics of participants, the performance measures used, and how much better the speech analysis did than cognitive testing. Without those details, readers cannot assess the size, reliability or clinical meaning of the reported difference.
It is also unclear whether the speech patterns predict future dementia, identify existing biological risk, or reflect other factors that can affect speech. The report does not establish a validated threshold, explain how the algorithm would perform across different populations, or show that it has been tested as a routine clinical screening tool. The findings are preliminary, and a speech measure alone should not be treated as a diagnosis.
More Evidence Before Clinical Use
The study’s ongoing work will need to establish whether the speech features consistently correspond to Alzheimer’s risk and whether they can predict later cognitive change. Larger and longer-term studies, with clear reporting of accuracy and participant groups, would help show whether the early comparison holds up beyond the initial analysis.
For now, the approach remains a research project. The reported findings do not establish a timeline for clinical availability or indicate that people should use speech changes to assess their personal dementia risk. Any future use would depend on further validation and evidence that the results can be interpreted safely alongside other clinical information.
Key Questions
Can speech changes diagnose dementia?
No. The reported research is preliminary and concerns identifying Alzheimer’s risk, not diagnosing dementia from speech. The source does not describe a validated clinical test.
What did the researchers measure?
The team measured about 130 speech-related features, using sensors to record fine movements of the tongue, face and mouth while participants spoke or repeated syllables.
What did the preliminary results show?
The researchers reported that speech-related data performed better than cognitive testing at identifying Alzheimer’s risk in cognitively unimpaired older people. The report excerpt does not give accuracy figures or other details needed to gauge the difference.
Should people be concerned about small changes in their speech?
The study does not show that an individual speech change predicts dementia. It describes an experimental analysis of many recorded features; it does not provide a basis for self-diagnosis or personal risk assessment.
Source: nytimes.com
