The Oddly Reassuring Ways 'AI' Is Changing Mental Health Diagnosis & Treatment
Tech advances enable earlier diagnosis of mental illness.
My Agency / Shutterstock A form of technology originally recognized in 1956, artificial intelligence (AI), has begun to drive a potentially profound shift in how mental illness is detected, diagnosed and treated. The recent boom in the use of telemedicine has accelerated AI's influence.
However, while several prominent sectors of society are ready to embrace the potential of AI, caution remains prevalent in medicine, including psychiatry, evidenced by recent headlines in the news media like this one from the New York Times: "Warnings of a Dark Side to AI in Health Care."
Yet, regardless of apparent concerns, AI applications in medicine are steadily increasing. That includes the field of mental health, as well.
The big question now is, do AI techniques trump the skill and experience of clinical psychologists and other mental health caregivers?
Does artificial intelligence (AI) detect mental health issues better than clinicians?
In just over a decade, nearly all sections of the medical industry have been ushered into a new generation of digital care delivery.
Upended by the crisis of COVID-19, the healthcare sector has had to quickly find new methods of providing quality care for patients.
As a result, most of them turned to a digital world of healthcare solutions such as telehealth.
Telehealth services have given a new ray of hope to patients suffering from anxiety, depression and other mental health issues.
"The onset of COVID-19 dramatically increased the use of telehealth," said Zac Imel, Ph.D., professor, and director at the University of Utah, of clinical training in the department of educational psychology.
He went on: "Mental health is one area of healthcare that can be efficiently and effectively delivered via telehealth. Not everything is lost, and at least some results can be achieved. My program on telehealth has seen a major shift from 0 to 95% students opting for this training."
In fact, it was reported that more mental health services were being looked for online during the pandemic than other medical ailments during the pandemic.
The pandemic has thus compelled scientists to improvise Artificial Intelligence (AI) technology for its application in the mental health sector.
In the past, various tools such as chatbots and virtual assistants have tested the waters of AI.
However, since the demand for counseling services increased exponentially during the pandemic, the industry had to turn to more-efficient ways of meeting the shortage by expanding their use of AI tools.
In a paper published by Nida Saddaf Khan, Muhammad Sayeed Ghani, and Gulnaz Anjum, titled, ADAM-sense: Anxiety-displaying Activities Recognition by Motion Sensors, a Human Activity Recognition-based method is proposed to detect some behaviors pertaining to anxiety display.
A novel dataset of anxious behaviors is also created with unique features using motion sensors of smartphones and the Inertial Measurement Unit (IMU) to make such a model.
Human Activity Recognition systems recognize human activities based on the streamed data of sensors. This approach has found its application in the field of psychiatry, where the patients suffer from mental, emotional, and behavioral disorders.
The most common mental disorder is Anxiety Disorder (AD), according to the American Psychiatric Association (APA). AD affects about 30 percent of the adult population at some point in their lives.
Researchers have created and compared several deep learning-based models.
A deep model consisting of Convolution Neural Network (CNN) and Long-Short Term Memory (LSTM) has outperformed other models and could recognize anxiety-related behaviors with over 92-percent accuracy.
Effect of AI on Psychiatry
From the academic study, "Artificial Intelligence for Mental Health and Mental Illnesses: An Overview," published by the National Center for Biotechnology Information: "If we leverage today’s available technologies, we can obtain continuous, long-term monitoring of the unique bio-psycho-social profiles of patients that affect their mental health."
What this means is the field of mental health diagnosis and treatment appears to be on the verge of a technological shift that could increase accessibility to an enormous amount of data related to human behavior. Unfortunately, the resulting amount of complex data is too much for a human mind to process efficiently. However, AI is well suited to this task.
Benefits of AI in Psychiatry
The physician's clinical hours are limited and cannot compensate for the growing needs. In addition, clinicians are heavily laden with increased documentation requirements and poor, inefficient technology to support it.
These problems are specifically cumbersome for mental health counselors who are compelled to rely on their skills to foster therapeutic rapport with their patients and formulate customized treatment plans.
In addition to improving the detection and diagnosis of mental illnesses, AI technology offers many other benefits, including:
- Drawing meaning from large and varied data sources
- Enabling a better understanding of the population-level prevalence of mental illnesses
- Uncovering biological mechanisms or risks, or protective factors
- Offering technology to monitor treatment progress and medication adherence
- Delivering remote therapeutic sessions
- Providing intelligent self-assessment to determine the severity of mental illness
As AI techniques improve, defining mental illnesses can become more objective than the current DSM-5 classification schema. The obvious benefit is that identification of mental illnesses could happen much sooner, when interventions may be more effective.
Sidhharrth S. Kumaar is the Founder of NumroVani and a registered pharmacist turned Astro Numerologist. For more information, visit his website.