AI System Works with Physicians to Identify the Most Helpful Treatments for People Diagnosed with Depression
When searching for an exact diagnosis through a myriad of complex and serious factors spread across millions of data points, it helps to have a powerful magnifying glass — and another expert point of view.
Enter Arjun P. Athreya, Ph.D., an engineer by trade who has a knack for bringing together data science algorithms and computing technology to predict events relating to potential disasters lurking within mission-critical cybersecurity infrastructure.
Dr. Athreya is the inaugural doctoral graduate of the Mayo Clinic and University of Illinois Alliance for Technology-Based Healthcare Fellowship program. Now, as a new Mayo Clinic faculty and staff member within Molecular Pharmacology and Experimental Therapeutics, the electrical and computer engineer has his sights set on health care.
Dr. Athreya developed a system for Mayo Clinic that uses artificial intelligence approaches ranging from machine learning to probabilistic graphs to better indicate treatment prognoses in people diagnosed with depression — the leading cause of medical disability worldwide. The system works by identifying patterns within patient history and other relevant data to predict which treatment option is best for the patient’s condition.
“In working with physicians, I learned that finding a diagnosis or treatment prognosis for a complex condition using the huge volumes of data generated from each patient can be like searching for a needle in the haystack,” Dr. Athreya says. “I try to create a magnifying glass to narrow the possibilities down and support the physician’s medical expertise.”
How do AI and machine learning fit into health care? Mathematical formulations of AI methodologies can discover patterns in a patient’s data — such as genome, microbiome and imaging data — that can explain unique characteristics of the specific patient, allowing for the right treatment to be chosen at the right time and right dose to achieve the therapeutic benefit.
“When people hear AI, they usually feel like they’re going to be replaced by machines,” he says. “First, no doctor will be replaced, as I argued in my doctoral dissertation. Instead, I want to show that AI-based tools serve as an interactive companion to the physician, a technological innovation that assists clinicians in their patient care delivery.”
William V. Bobo, M.D., is one of the physicians who collaborates with Dr. Athreya as part of a research team that is looking to personalize the treatment of major depression. A variety of treatment methods exist, but because of the timeline required to monitor effects — sometimes spanning several years — patients often grow weary of the process.
“Precision is of utmost importance in medicine,” Dr. Bobo says. “We work to give patients a more accurate diagnosis earlier, to spare them the suffering of their symptoms and the frustration they have.”
By combining the two mutually exclusive fields of direct patient care and data analysis, the team is able to answer that need. Dr. Bobo thinks this unique spirit of collaboration at Mayo Clinic is what makes the approach successful.