NIH Grant Supports Machine Learning to Improve HIV Patient Outcomes

 The National Institutes of Health awarded Brown University grants totaling $4.6 million to support the use of machine learning to improve outcomes for patients with HIV.  

Working collaboratively with Moi University in Kenya, Brown University will create, test, and launch data-driven tools to enhance the effectiveness of HIV care programs.  

Over the past four decades of treating HIV/AIDS, researchers have learned that HIV-positive patients need to be put and remain on effective treatment as soon as possible.  

Maintaining treatment can help turn HIV into a chronic but manageable disease, according to Joseph Hogan, a professor of public health and of biostatistics at Brown University, who has been researching HIV/AIDS for 25 years. 

Hogan is one of the primary investigators on the two recently awarded grants from the National Institutes of Health. The nearly $4.6 million grant will support the creation and utilization of data-driven machine learning tools to assist care programs in Kenya to meet treatment goals.  

“If the system works as designed, then we have confidence that we’ll improve the health outcomes of people with HIV,” co-director of the biostatistics program for Academic Model Providing Access to Healthcare (AMPATH), Hogan said in a press release. 

The first part of the project will use data science to understand the HIV care cascade. Collaborating with associate professor of biostatistics at Moi University, Ann Mwangi, Hogan will use AMPATH-developed electronic health record databases to develop algorithm-based machine learning tools.  

The tools will work to predict when and why patients might drop out of care when their viral load levels show they are at risk of treatment failure.  

According to Hogan, these algorithms will be integrated into the electronic health record system to deliver the data at the point of care through tablets physicians can use in the exam room with the patient.  

Working with experts in user interface design, the team will assess and test the best ways to communicate the results of the algorithm to the providers to assist them in deciding on patient care.  

The predictive modeling system the research team is developing will notice a physician to red flags in the patient’s treatment plan at the point of care, allowing for medical professionals to make changes as needed. 

“The idea is that the physician will be able to use the results of the algorithm to see at the point of care which patients are at risk, and then, be able to take preventive actions to avoid the negative outcomes, rather than respond to negative outcomes after they happen,” Hogan said. 

“With this project, we hope to bring the promise of A.I. and machine learning to the patient and clinic level and evaluate the developed tools that are going to have a measurable impact on patient outcomes.”