A team of scientists at the Institute for Creative Technologies (ICT) at the University of California have developed an innovative new way of detecting the signs of depression and post-traumatic stress disorder (PTSD).
Ellie is an on-screen avatar. She’s welcoming, has a soft, encouraging voice and she asks probing questions in a non-judgmental way. With encouraging body language and a face that is interested and responsive, she has been highly trained and is very skilful and picking up clues.
She starts by asking a series of questions that can pick up on the signs of depression and PTSD. At the same time, a sensor and webcam is scanning your facial and body movements as well as your tone of voice. This information is then fed to Ellie so she knows the best way to interact with you.
The concept has been met with mixed reviews but the co-creator of the project, Professor Louis-Philippe Morency has stressed that the machine is not a substitute for a real therapist. She has instead been designed to help gather information and that a diagnoses and treatment decisions will still be made by human clinicians.
The machine is said to work by detecting minute facial and body movements that a human might not be able to pick up on. Anxious people are known to fidget with their hands which is an obvious movement but whilst distressed people may smile as often as happy people, their smiles are shorter and less intense which is something that may be a bit harder to recognise.
Do you think that Ellie could be a valuable tool in detecting mental illnesses? Maybe people will feel less anxious talking to a computer as opposed to an actual person? Alternatively, do you think that there’s no substitute for human interaction? Feel free to share your thoughts with us on our Facebook page.