Sarra Eddahiri '19 shares insights on the promising potential of AI in the effective diagnosis, treatment and prevention of mental illness.
It was during an undergraduate Global Health class at Elon when I first learned about the complexity of the global mental health crisis. Class discussions included the increasing prevalence of mental illness globally and the combination of biological, psychological and societal factors that contribute to it.
One focus in particular that caught my interest and stuck with me throughout the years was this question of how much of the increasing prevalence of mental illness is linked to the deterioration of our general well-being in society, and how much of it is linked to our improved ability to better capture these rates as a result of decreasing stigma, better understanding of the human brain and improvement of the methodologies and technologies used for diagnosis. Although there is currently no definitive answer to this question, there is no doubt that technological progress has contributed to our ability to better understand the human brain and mental health and will continue to do so.
The current relationship between technology and mental health is, understandably, mainly portrayed pejoratively. The word “technology” is often associated with personal devices, media and the internet and criticized for its direct impact on attention span, addiction and self-esteem as it provides an aperture on others’ lives enabling for constant comparison. While these criticisms are valid, I believe we are slowly entering a new and exciting era that will showcase an unprecedented positive, productive and progressive relationship between technology usage and mental health. The bridge for this relationship is artificial intelligence (AI).
When leveraged in a rigorous, ethical and creative way, machine learning has the great potential of addressing many of the most complex problems involving human behavior and emotions.
The effective diagnosis, treatment and prevention of mental illness is a burgeoning global health challenge ripe for the promising potential of AI. Like other medical fields, mental health care is plagued by subjectivity and human error, often due to reliance on self-reporting and interview-based approaches. Such methods put the burden on patients to proactively report and describe their own symptoms accurately, despite their critical conditions. Many of the unique challenges of mental illness diagnosis and treatment are related to the lack of objective monitoring and evaluation of symptoms that are difficult to achieve given the unpredictability and complexity of human behavior and emotions.
Fortunately, we are entering a new age of mental illness diagnosis. Machine learning technology has demonstrated unprecedented capabilities to process and learn from complex data patterns to drive highly reliable inferences. On the other hand, our relentless use of everyday technologies has yielded an abundance of data that captures our most detailed and precise behavioral patterns — a perfect data source for a data-hungry technology. These are the exact tools needed to build a powerful solution capable of automating mental illness diagnosis and monitoring, which minimizes the burden on patients and their health care providers.
When leveraged in a rigorous, ethical and creative way, machine learning has the great potential of addressing many of the most complex problems involving human behavior and emotions. The next few years will be defining for this revolutionary AI application that will enable the monitoring and forecasting of mental health fluctuations to provide help to those who need it the most at an early stage and before it is too late. The question then becomes, how can we ensure the safe and ethical application of such powerful AI technologies?
A 2019 public health studies graduate, Sarra Eddahiri’s career in the tech industry combines human-centered problem-solving, product safety and research.