QnAs with Donald Geman
With the proliferation of “omics” technologies, personalized medicine—which tailors treatment to an individual’s genomic profile—promised a revolution in care. That revolution, says applied mathematician Donald Geman, has been slow to arrive. Geman has spent nearly four decades devising statistical methods for a variety of applications. He recently teamed up with an interdisciplinary group of scientists at The Johns Hopkins University, where he is a professor of applied mathematics and holds appointments at the university’s Institute for Computational Medicine and Center for Imaging Science. Geman helped engineer an algorithm that reduces data complexity and may assist in differentiating between certain forms of cancer. This work builds on his earlier research in computer vision, leveraging his experience with pattern-recognition problems. PNAS recently spoke to Geman, who was elected to the National Academy of Sciences in 2015, about his current research.