Virtual Avatar Aids Dementia Detection

Researchers have developed a machine learning algorithm that can diagnose dementia based on conversations between elderly individuals and virtual avatars.

AsianScientist (Sep. 20, 2018) – A team of scientists in Japan has demonstrated that it is possible to detect dementia from verbal exchanges between humans and a virtual avatar. Their findings are published in the IEEE Journal of Translational Engineering in Health and Medicine.

As populations around the world age, neurocognitive decline in the form of dementia is becoming increasingly common. However, current methods to detect dementia involve medical imaging systems or neuropsychological questionnaires. The former is expensive and non-portable, while the latter may be inaccurate with repeated use on the same patient as he or she becomes familiar with the questions asked.

To overcome these challenges in diagnosing dementia, a group of researchers from Osaka University and Nara Institute of Science and Technology in Japan developed machine learning algorithms for detecting signs of dementia in its early stages.

The team prepared fixed questions based on neuropsychological tests and random questions. They then recorded interactions between virtual avatars and 12 individuals diagnosed with dementia, as well as with 12 healthy controls. Subsequently, the researchers extracted speech, language and image features from the recorded data, creating a model for detecting dementia.

They found that delayed responses to questions from avatars, altered intonation, changes in articulation rate and differences in frequency of noun and verb usage could be used to diagnose dementia in human patients. Using their machine learning algorithm, the researchers were able to distinguish individuals with dementia from healthy controls at a rate of 90 percent.

“If this technology is further developed, it will become possible to know whether or not an elderly individual is in the early stages of dementia through conversation with computer avatars at home on a daily basis. It will encourage them to seek medical help, leading to early diagnosis,” said Dr. Takashi Kudo of Osaka University.

The article can be found at: Tanaka et al. (2017) Detecting Dementia Through Interactive Computer Avatars.


Source: Osaka University; Photo: Shutterstock.
Disclaimer: This article does not necessarily reflect the views of AsianScientist or its staff.

Asian Scientist Magazine is an award-winning science and technology magazine that highlights R&D news stories from Asia to a global audience. The magazine is published by Singapore-headquartered Wildtype Media Group.

Related Stories from Asian Scientist