Researchers at Texas A&M University have developed a smart-device based digital platform which can monitor hyperarousal, a key sign of psychiatric distress. The objective is to aid patients manage their mental wellness.
According to the researchers, this advanced technology can read facial cues, examine voice patterns and integrate readings from in-built vital signs sensors on smartwatches to determine if a patient is stressed. They also note that it could provide feedback and alert care teams in case of an abrupt deterioration in the mental health of a patient.
‘Mental health can change very rapidly, and a lot of these changes remain hidden from providers or counselors,’ said Farzan Sasangohar, Assistant Professor in the WM Michael Barnes ’64 Department of Industrial and Systems Engineering. ‘Our technology will give providers and counselors continuous access to patient variables and patient status, and I think it’s going to have a lifesaving implication because they can reach out to patients when they need it. Plus, it will empower patients to manage their mental health better.’
This digital platform of the researchers is described in the Journal of Psychiatric Practice.
This platform may tackle the challenge of lacking information on a patient’s mental health and would give unexpected deterioration a chance to be addressed. This way, health care professionals could know about their patients’ ongoing struggle with mental health and provide them with the appropriate care.
Patient-reported results between visits to the health care professional are very important in the design of effective health care interventions for mental health. To bridge this gap, Sasangohar’s team collaborated with clinicians and researchers in the Department of Psychiatry at Houston Methodist Hospital to develop a smart digital platform to assess the mental wellbeing of a patient.
‘The hospital has the largest inpatient psychiatry clinic in the Houston area,’ Sasangohar said. ‘With this collaboration, we could include thousands of patients that had given consent for psychiatric monitoring.’
The clinicians at Houston Methodist Hospital were already using an off-the-shelf patient navigation tool called CareSense. This software sends reminders and monitors questions to patients to better analyze their wellbeing. Individuals at risk of self-harm could be prompted to take questionnaires for major depressive disorder periodically.
Instead of relying solely on the patient’s subjective assessment, Sasangohar and his team developed a suite of software for automatized hyperarousal analysis easily installable on smartphones and smartwatches. These programs collect input from face and voice recognition applications and sensors already built into smartwatches, like pedometers and heart rate sensors. The data from these sources trains the machine learning algorithms to identify patterns aligned with the normal state of the arousal. Once it is trained, the algorithms continuously checks readings from the sensors and recognition applications to determine if an individual is in an elevated state of arousal.
‘The key here is triangulation,’ Sasangohar said. ‘Each of these methods on their own, say facial sentiment analysis, show promise to detect the mental state, albeit with limitations. But when you combine that information with the voice sentiment analysis, as well as physiological indicators of distress, the diagnosis and inference become much more powerful and clearer.’
According to Sasangohar, the subjective evaluation of mental state and the objective evaluation from the machine learning algorithms are integrated to give a final assessment of the individual’s state of arousal.
Although the prototype of their technology is ready, the researchers insist they still need to improve the battery life of smartphones carrying the software because the algorithms absorb a lot of power. They also have to address usability issues like difficulty in navigating the application.
‘Because of the stigmatization that surrounds mental illness, we wanted to build a mental health monitoring device that was very discreet,’ Sasangohar said. ‘So, we chose off-the-shelf products, like smartphones, and then build sophisticated applications that operate within these devices to make monitoring mental health discreet.’
Some other contributors to the study include Dr. Christopher Fowler and Dr. Alok Madan from the University of Texas McGovern School of Medicine and Baylor College of Medicine; Courtenay Bruce and Dr. Stephen Jones from the Houston Methodist Institute for Academic Medicine; Dr. Christopher Frueh from the University of Texas McGovern School of Medicine and the University of Hawaii; and Bita Kash from the Methodist Institute for Academic Medicine and Texas A&M.
The research is funded by the Texas A&M University President’s Excellence Grant (X-Grant)
By Marvellous Iwendi.
Source: Texas A&M Today