Canadian researchers are working to make early diagnosis of Alzheimer’s disease possible with a machine learning model that could one day be turned into a simple screening tool that anyone can use with a smartphone.
The model developed by researchers at the University of Alberta, Canada, has the potential to be easily accessible via a smartphone and can distinguish between Alzheimer’s patients and healthy people with 70-75% accuracy. Even though the scientists focused on English and Greek speakers, the technology has the potential to be used in other languages as well.
By focusing on speech patterns rather than content, the tool could provide early indicators, which will help initiate treatment earlier and, implicitly, slow the progression of the condition. Although it is not a substitute for specialist doctors, it could support diagnosis in the early stages.
The new method of early diagnosis would be less expensive
The model created was able to distinguish Alzheimer’s patients from healthy people with an accuracy between 70 and 75%. Alzheimer’s can be difficult to detect in its early stages because the symptoms are subtle and can be mistaken for memory impairment typical of old age. The sooner potential problems are detected, the sooner action can be taken.
You may also be interested: “People become immortal in eight years!” He who said this has also predicted the future
“Until now, laboratory analyzes and imaging investigations were needed to detect brain changes, which required time, was expensive, and no one is tested this early,” he told Neurosciencenews
Eleni Stroulia, involved in the creation of the machine learning model.
Characteristics of people with Alzheimer’s, detectable with a smartphone
The Canadian research group analyzed the language used by Alzheimer’s patients. They examined acoustic and speech characteristics rather than specific words. The researchers started with speech characteristics that doctors noticed were common in Alzheimer’s patients.
See also: An artwork created by an artificial intelligence program sold for more than $400,000
These patients tended to speak more slowly, with more pauses or interruptions in speech. They usually used shorter and often unintelligible words. The researchers found ways to translate these features into speech features that the model could analyze.
PHOTO: Pixabay
2023-06-13 14:02:12
#Artificial #intelligence #detect #signs #Alzheimers #smartphone #Click #women
