As and when people becomes older, the probability of development of dementia keeps increasing. In spite of the presence of medical imaging tests and cognitive function tests, the detection of dementia is not that advanced. Scientists throughout the world are in search of a right detection tool at an early stage of development.

A joint research team from Nara Institute of Science and Technology and Osaka University have found a technique to detect dementia via machine learning, where the machine analyses the speech and language of the patients while answering the questions provided by interactive computer avatars.

Varied features of dementia, such as the duration of delay in response to questions from avatars, articulation rate of the voice, intonation, and the percentage of verbs and nouns in utterance were taken into consideration.

The accuracy of this detection test was found to be 92% by answering around 6 random neuropsychological based questions which take around 2-3 minutes to answer per question.

Takashi Kudo, the senior author of this work, said "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."