Virtual Human Clinician Helps with Mental Health Diagnoses
Millions of soldiers around the world carry a high risk of suffering Post Traumatic Stress Disorder (PTSD). PTSD can lead to depression and suicide. This is why it is vital to enable a new generation of clinical decision support tools and interactive virtual agent-based healthcare dissemination/delivery systems that are able to recognize and identify psychological distress from multimodal signals. These tools aim to provide military personnel and their families’ better awareness and access to care while reducing the stigma of seeking help. For example, the system’s early identification of a patient’s high or low distress state could generate the appropriate information to help a clinician diagnose a potential stress disorder. User-state sensing can also be used to create long-term patient profiles that help assess change over time.
When Canadian soldier Jody Mitic was sent to Afghanistan, he suffered a devastating landmine injury that severed both his feet. Undeterred, Mitic rebuilt his life and started a family and a business back in Canada. Like millions of soldiers around the world, Mitic carried a high risk of suffering Post Traumatic Stress Disorder (PTSD). PTSD can lead to depression and suicide. The US Veterans Administration says that veteran suicides have jumped 32% since 2001 and veterans between 18 and 29 years of age were six times more likely to take their own lives than civilians. PTSD is difficult to diagnose, which is why the team at the University of Southern California’s Institute for Creative Technologies (ICT) developed SimSensei – artificial intelligence with advanced emotional intelligence and facial recognition software called Multisense in the shape of a virtual clinician called ‘Ellie’. Multisense automatically analyses over 60 facial expressions in real time and over the long term, recording a wide range of symptoms associated with depression, anxiety and PTSD. The results can then be reviewed by a human therapist to help inform treatment. ICT has now conducted sessions with more than 600 subjects, most of them soldiers returning from deployment. The results have been staggering: 88% of subjects agreed or strongly agreed that they were willing to share information with the completely automated Ellie. Many said they felt more comfortable speaking to a virtual human because they did not fear judgement, even when they knew researchers would review the results. When Mitic spoke to Ellie, he was able to recall a traumatic experience where he felt unable to help a civilian woman with a baby. Thankfully, his speech pattern, facial expression and tone showed that he was able to deal with the trauma and did not require treatment.
Sierra Briscoe - https://www.linkedin.com/in/sierrabriscoe/