When people hear the term Artificial Intelligence or AI – many think of a movie image of some AI character outsmarting society – eventually deciding to wipe the population off the face of the Earth. Luckily, this stereotypical doomsday scenario is now a thing of the past. Instead, the positive evolution of AI is in full effect across the worlds of business, retail, finance, and medicine and is expanding into other fields, including mental health.
One research project is now using artificial intelligence to detect behavioral signs of anxiety with more than 90 percent accuracy. What does this mean for clinicians? It suggests that AI has the necessary applications for addressing mental health and well-being. Below, we investigate this new research and discuss how behaviors were analyzed using deep learning algorithms. We also report on the latest ways mental healthcare professionals use AI as digital diagnostic tools and how this affects the future of the 16 trillion dollar mental health market.
Anxious? There’s an AI for That
In Neuroscience News, Simon Fraser University visiting professor and social psychologist Gulnaz Anjum states, “Since the onset of COVID-19 and one climate disaster after another, more and more people are experiencing anxiety. Our research shows that AI could provide a highly reliable measurement for recognizing the signs that someone is anxious.” During their study, first reported in the journal Pervasive and Mobile Computing, Anjum and team members collected data from adult participants for their Human Activity Recognition (HAR) research. In the field of Human Activity Recognition, human activities are recognized based on sensors’ streaming data. HAR has been utilized widely in areas where human behaviors are studied, such as healthcare, rehabilitation, and many other domains, including psychiatry.
In the new AI study, participants performed a series of activities in a specific order while wearing sensors that recorded their movements. The researchers created a dataset of typical anxiety-displaying behaviors for the sensors to detect, including idle sitting, nail-biting, knuckle cracking, and hand tapping. They produced this data using smartphone motion sensors and the Inertial Measurement Unit (IMU). Then, their behaviors were analyzed using deep learning algorithms and computational hybrid models. These measures were shown to perform better than other models and could recognize anxiety-related behaviors with over 92 percent accuracy. The researchers suggest this AI could help analyze, diagnose, treat, and monitor psychological disorders such as anxiety disorder.
AI as a Personalized Mental Health Diagnostic Tool
According to a recent Business Insider article, psychiatry researchers are also using artificial intelligence to create more effective and personalized treatment plans. As Dr. Ellen Lee, staff psychiatrist at the VA San Diego Healthcare System, tells Business Insider, “Psychiatry is a unique field because mental healthcare providers generally don’t have clear imaging findings indicating mental health pathology to make a diagnosis.” She adds, “Instead, practitioners primarily rely on the patient’s self-reported symptoms and medical history.” AI addresses this issue by helping researchers assess and diagnose the variability of mental health conditions more fully.
Case in point, Dr. Charles Marmar, a Department of Psychiatry chair at NYU Grossman School of Medicine, states in the article, “It [AI] can help us determine whether there is one kind of depression or seven kinds of depression.” Marmar and his colleagues have been using machine learning — a form of AI that uses computer algorithms to analyze large amounts of data — to evaluate the heterogeneity of psychiatric illnesses. The more data these algorithms process, the more accurate they become. More specifically, a machine learning method, random forests (RF), in conjunction with a clustering method, were used to identify subtypes derived from 16 self-report and clinician assessment scales, including the clinician-administered PTSD scale for DSM-IV.
The research team used this machine learning to pinpoint two forms of post-traumatic stress disorder (PTSD) in veterans, a mild form with relatively few symptoms and a chronic, severe form in which patients experienced high levels of depression and anxiety. The study was published in Translational Psychiatry. Using machine learning, Marmar plans to explore further and validate possible five PTSD subtypes: anxious and dissociative, depressed, cognitive functioning impaired, mild, and severe.
How AI is Filling Future Gaps in Mental Health Care
According to the University of Maryland Division of Research, University of Maryland researchers are creating a computerized framework that could lead to a system capable of a kind of “mental weather forecast.” This system will blend language and speech analysis with machine learning and clinical expertise to help mental health clinicians and patients connect and head off crises. It could also help reduce the 16 trillion dollar cost of treating mental health worldwide.
How will it work? Patients answer a series of questions about physical and emotional well-being. The system will use AI to analyze word choice and language use. It will also monitor the patient’s speech patterns, examining changes in the timing and degree of movement made by the lips and different tongue parts. It would then compare it to a baseline sample taken from healthy control subjects or when the participant was in remission. The system could live in an app on patients’ phones and automatically monitor their mental state to determine their level of need for clinical intervention and what resources are available to help.
According to Deanna Kelly, director of the Maryland Psychiatric Research Center’s Treatment Research Program, “The World Health Organization estimated that the cost of treating mental health issues between 2011 and 2030 would top $16 trillion worldwide, exceeding cardiovascular diseases. Pandemic stress has exacerbated an already high level of need, and in some cases resulted in breakdown conditions for the system.” With this in mind, Kelly adds the cost benefits of this computerized AI framework stating, “Serious mental illness makes up a large portion of health care costs here in the U.S. and around the world. Finding a way to assist clinicians in preventing relapses and keeping people well could dramatically improve people’s lives and save money.”
