Developing a speech analysis platform based on AI models to early detect mental health conditions

What is the Challenge?

Dementia, especially dementia caused by Alzheimer's disease, is considered one of the most severe health problems of our time. Currently more than 55 million people live with dementia worldwide, and there are nearly 10 million new cases every year. Today’s screening tests by physicians have limitations and AI technology can be used to spot patterns or changes to memory and thinking. It is known that the disease begins many years before clinical symptoms appear, so how can we use the power of AI to detect early symptoms of dementia?

What is the Solution?

Spanish-based startup Accexible is developing a speech analysis platform based on AI models for early detection of diseases such as dementia, Parkinson and coronary disease. Accexible’s automated speech analysis captures a patient’s voice and uses AI to analyze speech patterns and detect the early signs of dementia from tone, pauses, repetition, and sentiment analysis via so-called vocal biomarkers. In any device with a microphone, a user speaks for 30 seconds, and in less than one minute the platform gives an accurate result to the health professional. Accexible detects dementia with an accuracy rate of over 91 percent. The solution enables a rapid, digital, cost-effective tool to diagnose and monitor diseases, reducing the burden from the healthcare system and caregivers and allows them to better respond to patient’s individual needs.