Home AI Entertainment AI in Mental Health: The Journey, Rewards, and Stumbling Blocks – A...

AI in Mental Health: The Journey, Rewards, and Stumbling Blocks – A Fresh WHO Research

Artificial intelligence (AI) presents exciting possibilities in mental health research and treatment. However, a recently published study sheds light on potential pitfalls that could point towards an over-hasty endorsement of nascent AI techniques still requiring real-world validation.

AI: A Potential Game-Changer in Mental Health Care

In 2021, a staggering 150 million individuals in the WHO's European Region were grappling with mental health issues. The COVID-19 pandemic exacerbated the situation, leading to heightened stress levels, limited access to care, and amplified vulnerability due to economic instability, conflict, and violence.

As the mental health crisis surges, the medical field is undergoing a transformation, spearheaded by AI. AI has emerged as a promising ally in strategizing mental health services, pinpointing, and tracking mental health conditions on an individual and community level. The use of AI enables automation of tasks, aids healthcare professionals, and provides deeper insights into intricate disorders, using digitized data in formats like electronic health records, medical images, and hand-written clinical notes.

In September 2022, WHO/Europe launched its “Regional Digital Health Action Plan for the WHO European Region 2023–2030”. This initiative acknowledged the pivotal role of big data and AI in pioneering predictive analytics for improved health outcomes.

“It is crucial to review the current application of AI in mental health research, given its increasing usage in healthcare. This will shed light on trends, existing gaps, potential opportunities, and inherent challenges,” asserts Dr. David Novillo-Ortiz, Regional Advisor on Data and Digital Health at WHO/Europe, and co-contributor to the study.

Unraveling the Challenges

The comprehensive review titled “Methodological and Quality Pitfalls in the Deployment of AI in Mental Health Research”, authored by leading minds from the Polytechnic University of Valencia, Spain, and WHO/Europe, scrutinizes the utilization of AI in mental health research from 2016 to 2021.

“We noticed a skewed application of AI in mental health research, heavily favoring the study of depressive disorders, schizophrenia, and other psychotic disorders. This uncovers a substantial gap in our comprehension of its application in studying diverse mental health conditions,” highlights Dr. Ledia Lazeri, Regional Advisor for Mental Health at WHO/Europe.

While AI promises innovative insights into crafting effective health policies and understanding the status of mental disorders, it also entails the complicated use of statistical analyses, mathematical methodologies, and high-dimensional data. This could potentially introduce bias, misinterpretation of results, and unwarranted optimism towards AI efficacy if not appropriately managed. The review detected major flaws in how AI processes statistics, inadequate data validation, and insufficient risk of bias assessment.

Furthermore, the absence of transparent reporting on AI models poses a significant problem, impairing their repeatability. The study revealed that data and models typically remain confidential, leading to a lack of cooperation among researchers.

“The lack of transparency and methodological lapses are alarming, as they impede the safe and pragmatic deployment of AI. Additionally, data engineering for AI models seems neglected or misconstrued, and data management is often subpar. These critical drawbacks might hint at an overzealous promotion of AI models without sufficient real-world viability evaluation,” explains Dr. Novillo-Ortiz.

“AI forms the backbone of the impending digital upheaval. In this review, we got a sneak peek into what the future holds, and it is a clarion call for healthcare systems to modify their infrastructures and processes to incorporate mental health services,” adds Antonio Martinez-Millana, Assistant Professor at the Polytechnic University of Valencia, and co-contributor to the study.

Key findings from the study were unveiled at a WHO/Europe organized event on 7 December 2022, titled “Big Data Analytics and AI in Mental Health.” This assembly brought together experts from across the European Region to discuss realistic ways of implementing AI models in planning mental health services, as well as safety and success factors, such as the involvement of individuals living with mental health conditions in the development process.

Exit mobile version