DementiaNet: Spontaneous Speech Data for Dementia Detection

DementiaNet: Spontaneous Speech Data for Dementia Detection

Alzheimer's sucks!
The development of Alzheimer's starts long before the onslaught of symptoms. By the time the symptoms are diagnosed, it is too late to treat the condition.
If we could diagnose Alzheimer's early enough, studies have shown that lifestyle changes and therapies can significantly slow the progression of Alzheimer's disease. Catching cognitive impairment early is essential to helping those with the condition.

We have put together a dataset to predict the cognitive impairment ten to fifteen years before the onslaught of symptoms. It's a spontaneous speech dataset of individuals taken over ten to fifteen years before the onslaught of symptoms and confirmed diagnosis of dementia.
The sample dataset contains a hundred individuals with a confirmed dementia diagnosis. Spontaneous speech samples (audio) range from time after the confirmed diagnosis to ten years before the symptoms appear. And a hundred individuals over the age of eighty with no cognitive decline (NC) and active in their field of work. Spontaneous speech samples for the NC group fall into three buckets, five years, ten years and fifteen years before death or current age.

Early analysis of this dataset shows above 70% accuracy. According to our knowledge, DementiaNet is the largest publicly available longitudinal dataset for dementia prediction/screening.

GitHub - tuul-ai/dementianet
Contribute to tuul-ai/dementianet development by creating an account on GitHub.

If you are curious about the DementiaNet or want to work with it, write us at:

Great! You’ve successfully signed up.
Welcome back! You've successfully signed in.
You've successfully subscribed to .
Your link has expired.
Success! Check your email for magic link to sign-in.
Success! Your billing info has been updated.
Your billing was not updated.