Why Emotion AI Is the Key to Mental Health Treatment
The last few years have seen promising changes in the realm of mental health. People are talking about it more, breaking the stigma, and seeking treatment. Although this trend is positive, the numbers of those in need of mental health treatment are still high.
- By Alex Potamianos, Shri Narayanan
- April 7, 2020
In the U.S. alone, approximately one in five adults -- 43.8 million in total -- will experience mental illness in a year. Sixteen million adults will experience at least one depressive episode. Suicide is one of the leading causes of death, claiming more than 47,000 lives in 2017. In the face of these incredible numbers, mental health professionals are looking at a revolutionary new technology to help combat mental illnesses.
Emotion artificial intelligence (emotion AI) is a promising technology that offers health care professionals another option for supporting their patients with resources and monitoring their well-being. This technology can analyze subtle cues in the voices, writing, and facial expressions of individuals and react accordingly -- just like a human would. This ability to adapt is not only incredible but extraordinarily useful, too. Above all, the technology can help connect and fill information gaps between encounters with human experts.
Emotion AI Possibilities and Promise in Mental Healthcare
The applications for emotion artificial intelligence are practically endless, including in the psychology and healthcare fields. Emotion AI can free up doctors to work more with their patients by analyzing patient records and generating reports based on the data, handling administrative tasks, and even assisting with diagnosis or intervention. Emotion AI can also help patients create awareness of their emotional states and regulate their emotions better under stressful or challenging conditions.
Artificial intelligence can help doctors and therapists increase emotional awareness for their patients, such as in expressing empathy, and deliver diagnoses more quickly and with more accuracy. It can also be an important tool to predict how patients will approach therapy and take steps to ensure they are successful and remain in treatment.
By supporting healthcare providers, emotion AI can help them provide better care, spend more time with patients, and reduce the costs of mental health treatment.
Emotion AI for Chatbots Delivering Personalized Therapy
Chatting with a computer may not be the first thing that springs to mind when you think of therapy, but for many patients, interacting with a chatbot or a robotic companion powered by emotion AI can have significant benefits. There are several companies developing avatars and assistive robots to provide therapeutic options for patients.
Chatbot technology can serve a number of functions when it comes to providing treatment for mental illnesses. It may not seem that a computer can take on the highly personal task of providing therapy, but it turns out that many people interact with avatars the way they would with another human.
Additionally, virtual therapists may encourage more in-depth sharing than a human therapist. Some patients might feel more comfortable sharing their struggles with a machine than they would with another person. For patients who have a difficult time feeling comfortable in a therapy setting, an emotion AI-powered chatbot can be freeing. The chatbots create an emotionally safe environment for sharing, similar to how pretend play with virtual friends creates a safe environment that is essential for the emotional and social development of children. Of course, although promising in its capabilities and in complementing human expert support, many of the initial reports haven't yet been fully vetted in well-controlled studies across conditions and population.
Most important, emotion AI technology automates talk therapy and accessibility, enabling broader access to mental health treatments, which is one of the most expensive and time-intensive forms of treatment. Now those who may struggle to afford traditional therapy sessions or have difficulty finding time to schedule sessions have an alternative solution to finding and receiving treatment. These options can be made available when patients need them rather than booking an appointment in advance. This feature can potentially make chatbots incredibly useful as a therapeutic aid that complements traditional means of therapeutic delivery. Chatbot-assisted behavioral and therapeutic intervention is still a while into the future; however, chatbots that can help us be more aware about our own feelings and more sociable are here to stay.
Using Emotion AI to Predict Suicide Risks
Although emotion AI offers treatment options, it also can play a valuable role in predicting patient behavior, such as individuals' risk of suicidal behavior.
Facebook is one company using emotion AI to monitor users' posts, look for content that could signal that a user is suicidal, and alert local authorities. This process was developed after a popular feature allowed users to report posts that potentially signaled suicidal ideation and directed human moderators to look at the post and provide resources.
Developers are working to create emotion AI that can assist doctors with predicting suicide risk and clinical management. Although still in progress, this technology can be helpful in understanding what risk factors drive suicidal behavior. Researchers at Vanderbilt University created an AI model that used patient health records to predict suicide risk with 84 to 92 percent accuracy within one week of a suicide event and 80 to 86 percent within two years.
It's still early in the process of using emotion AI to predict suicide risk and support direct healthcare providers in preventing suicides, but it's exciting to watch the development of this promising technology-facilitated capability.
The Future of Emotion AI and Mental Health Treatments
It's clear that emotion AI is here to stay in the healthcare and mental health field. This promising technology has a lot to offer patients and care providers alike, and it's continuing to improve as devoted professionals pour their expertise into developing better AI models with the goal of helping people -- to broaden access, improve quality of care, and reduce costs.
There are still significant challenges to tackle if AI is to be effectively used to understand and help treat mental health conditions. The main challenge is how to integrate the AI into the therapist-patient loop of detection, awareness, and, especially, treatment. Privacy and patient data concerns also need to be addressed because AI models often rely on sensitive patient data to make decisions and predictions.
Because emotion AI capabilities are still developing, it's important to ensure that it is overseen by humans who are ready to take a second look at AI chats and treatments to verify that patients are receiving what they need. Emotion AI models are getting increasingly good at understanding patient emotions, but expert human supervision is necessary to prevent patients from being harmed by incorrect care during machine-assisted intervention.
Emotion AI is looking forward to a bright future working together with care providers to address the needs of patients. In the future, AI models will be able to assist in providing therapy and care for many patients who would otherwise not have access to care, whether due to time constraints or cost. This technology stands on the edge of a revolutionary change in the way we understand and treat mental illness.
Developers still face many complexities, but with continuing advances in emotion AI capabilities, the future of technology support for mental healthcare looks bright.
About the Authors
Alex Potamianos, a rare mix of scientist and businessman, is an innovator in the field of speech and natural language processing, interactive voice response systems, and behavioral informatics. He has over 20 years of leadership experience in the corporate and entrepreneurial sides of the business. His background includes working at AT&T Labs-Research, Bell Labs, and Lucent Technologies and is a co-founder of Alesman. His academic achievements have gone hand in hand with his extensive research work, receiving his M.Sc. and Ph.D. degrees in engineering sciences from Harvard University and later his MBA from the Stern School of Business, NYU.
Shri Narayanan is a pioneer in human-centered information processing and communication technologies, with a special emphasis on behavioral signal processing and informatics. Shri brings his vision to the team, encouraging new ways of thinking. He holds M.S. engineering and Ph.D. degrees in electrical engineering from the University of California Los Angeles (UCLA) and is an Andrew J. Viterbi Professor of Engineering at the University of Southern California (USC). Shri holds appointments as Professor of Electrical Engineering, Computer Science, Linguistics, Psychology, Neuroscience and Pediatrics and is the founding director of the Ming Hsieh Institute. Prior to USC, Shri was with AT&T Bell Labs and AT&T Research from 1995 to 2000. At USC, he founded and currently directs the Signal Analysis and Interpretation Laboratory (SAIL).