Many Americans have social and cultural hurdles that prevent them from getting timely cancer diagnoses and effective treatments. For example, lack of transportation can prevent women from being regularly screened for breast cancer, increasing the risk
of a late cancer diagnosis that can be difficult to treat. “Such social factors, also called social determinants of health, are well-known but have been difficult to analyze and interpret despite the use of powerful machine learning methods”
says Dr. Bhavnani, professor of biomedical informatics at UTMB, who co-leads the $1 million award from the National Institutes of Health with Dr. Hunter, clinical associate professor of pharmacy practice from Texas Southern University.
So what’s the problem? According to Dr. Bhavnani, there are three problems in analyzing social determinants of health. First, patients are similar and different in complex ways based on their social determinants of health. For example,
some patients lack transportation, health insurance, and steady employment, whereas others partially overlap with this group by also lacking steady employment, but additionally having language and communication problems with their providers. We need
methods that can not only identify such complex overlapping patterns to help design targeted solutions, but also measure their statistical significance and replicability, which are critical in the biomedical sciences.
A second methodological problem is that many AI and machine learning methods use complex mathematical formulas that transform the data in ways that are difficult to interpret by clinicians -- the so called “black box” AI problem. This problem
prevents researchers from inspecting whether the data and algorithms are biased against some groups. For example, if the data only includes individuals with health records, the results are biased because they exclude individuals who have never seen
a doctor due to low income. A third problem is that Black and Hispanic researchers are currently underrepresented in AI and machine learning research, which further increases the risk of biased data, analyses, and their interpretations.
The NIH award to Drs. Bhavnani and Hunter will directly address these three methodological hurdles by using an approach called Human-Centered AI. This approach will use graphical networks to automatically identify complex patterns in very large
datasets, while also visualizing the results at each step so ethicists, biostatisticians, and clinicians can inspect the data and interpret the results. The research will use data from the All of Us program which aims to contain data from
one million Americans with a focus on underrepresented populations. In addition, the collaboration with Dr. Hunter from Texas Southern University, a historically black college or university (HBCU), will leverage his rich experience in treating cancer
patients from disadvantaged communities to inspect the analysis and interpret the results.
Rodney Hunter, PharmD, BCOP, MPI and Clinical Associate Professor of Pharmacy Practice, TSU
"This partnership between TSU and UTMB on this NIH grant will begin a new era of leveraging AI and social determinants of health to help predict and address factors that have negatively impacted morbidity and mortality associated with cancers in minority patients, resulting in poorer outcomes compared to their majority counterparts.”
Kristen Peek, PhD, Dean of the School of Public and Population Health, UTMB
“We are very excited about this NIH award as it targets health inequity, while engaging researchers from multiple departments including biostatistics and data science, bioethics and humanities, and population health and health disparities.”
Randy Urban, MD, Chief Research Officer, and Director of the Institute for Translational Sciences, UTMB
“This NIH grant to Drs. Bhavnani and Hunter are part of an important regional alliance between UTMB and TSU, providing critical insights of how to incorporate social determinants of health into the care of patients with cancer.”
Sandra Hatch, MD, Chair of Radiation Oncology, UTMB
“Dr. Bhavnani’s human-centered AI approach will go a long way in reducing the hurdles faced by cancer patients in getting timely and effective cancer care.”
The NIH grant is from the Artificial Intelligence/Machine Learning Consortium to Advance Health Equity and Researcher Diversity (AIM-AHEAD).