Collaboration Stories

The purpose of this series is to tell the story of long-lasting, as well as emergent, stories of collaboration among clinicians and methodologists, content experts, and engineers, who have successfully worked on joint projects at the intersection of AI and health. Such collaborations are at the center of e-HAIL’s mission, and we offer these stories as a way of inspiring collaboration in other researchers.

LM*AI*C: How e-HAIL and the Center for Global Health Equity are Teaming Up to Fight Cancer Overseas

A cancer diagnosis is bad enough, but worse yet for some is a diagnosis that never comes or is late.

Too often in regions where there’s just a single pathologist for a million or more people, cancers remain undetected. It is why patients in low- and middle-income countries (LMICs) so often succumb to illnesses that are survivable in well-resourced places.

But a handful of faculty members at U-M are working to change that with help from e-HAIL and the U-M Center for Global Health Equity (CGHE). While much of e-HAIL’s work to-date has supported domestic research, recent projects have begun to look further afield, leveraging AI to improve timely access to cancer diagnosis and treatment.

“AI-enhanced diagnostic tools can allow us to catch patients much earlier in their cancer journey and get them into treatment pathways while their cancers are still highly treatable,” said Akbar Waljee, MD, MSc, the Lyle C. Roll Professor of Medicine, who has leadership roles in both e-HAIL and CGHE. “The challenge in many parts of the world is identifying patients in time, but we believe AI can help us bridge that gap.”

Waljee, along with Ulysses Balis, the A. James French Professor of Pathology Informatics, Arvind Rao, Professor of Radiation Oncology and Computational Medicine and Bioinformatics, as well as international partners, have received NIH support for their work on colorectal cancer (CRC) in Kenya, where data from Nairobi indicate that CRC increased 2.7-fold between 1993 and 2005. Their work focuses on using AI to help doctors diagnose this form of cancer more efficiently, thus potentially improving health outcomes. To do this, the currently available algorithms need to be trained on diverse datasets—not just on information from high-income countries. The newly established Diagnostic Medicine Consortium which Balis helps to lead aims to curate globally relevant data repositories so that AI tools can be just as reliable for practitioners and their patients in Africa as they are in America.

“Broadly sourced sets of primary diagnostic data are essential to build AI tools that perform reliably for all locales and populations,” said Balis, MD. “These validated data sets can and will be made freely available on a global basis. This knowledge can then be brought back to the individual patient to benefit their care.”

Likewise, Kamran Mirza, Godfrey D. Stobbe Professor of Pathology Education, is focused on ensuring that the tools and systems Balis and others are designing can be effectively implemented around the world. His work, including virtual medical education programs tailored to LMICs, addresses the human and systemic factors that determine whether technological innovations create lasting change.

A free PathElective Pathology Education website launched by Mirza and collaborators in 2020 offered virtual teaching modules after many in-person clinical rotations were canceled because of the COVID-19 pandemic. Within the first four months, the site had garnered some 35,000 visits and nearly 5,000 subscribers from 99 countries. It continues to be used all over the world as a resource to introduce medical students to the field.

“Sustainable impact requires building local capacity, addressing infrastructure challenges, and creating systems that work within existing constraints,” said Mirza, who, like Balis, is a member of both e-HAIL and CGHE.

Although advanced data science is rarely linked with under-resourced settings, Waljee sees it as exactly where the greatest opportunities for collaboration lie—and a sweet spot for continued e-HAIL-CGHE partnership.

“AI can help us reimagine pathology as a global resource, expanding the capability of providers to deliver timely, life-saving diagnoses regardless of the setting,” Waljee said.