Thanks to a $7 million gift from an anonymous donor, Albert Einstein College of Medicine is launching an institute that will allow scientists and clinicians to harness vast amounts of biomedical data, an essential resource for healthcare research.

A key element of Einstein’s strategic plan, the data science institute will help ensure that the College of Medicine has the infrastructure and expertise it needs to thrive in a rapidly evolving data driven environment.

“We are in the midst of a revolution in the field of artificial intelligence and machine learning—it’s just exploding,” says Marla Keller, M.D., executive dean at Einstein. “Being able to effectively use ‘big data’ allows us to make giant leaps forward in advancing basic science for drug discovery and improving patient outcomes.”

Harnessing the Power of Biomedical Data

“We live in an increasingly data-rich world in which modern biomedical technologies generate massive datasets,” says Mimi Kim, Sc.D., professor and head of the division of biostatistics in the department of epidemiology & population health at Einstein and Montefiore and the Harold and Muriel Block Chair in Epidemiology & Population Health.

Researchers now have at their disposal data on tens of thousands of molecular features and more, which enables greater understanding of the biological mechanisms underlying complex diseases, says Dr. Kim, who is also the associate director of the Harold and Muriel Block Institute for Clinical and Translational Research at Einstein and Montefiore.

Huge volumes of data that can be mined for research are also available from millions of electronic health records, as well as personal health devices like smartphones and smartwatches that collect information on physical activity, sleep, heart rate, and other health parameters.

Having all that data is great, but handling it is a challenge. “We need data science methods to maximize what we can learn from all these sources of very big, complicated data to extract knowledge and insights that lead to scientific breakthroughs and new treatments,” says Dr. Kim.

Einstein and Montefiore had pockets of data expertise in different departments—but no academic entity focused on developing and coordinating data science resources, tools, and training. The new data science institute, which will be launched this year and headed by Dr. Kim, brings all those data assets under one roof.

“Now we can help investigators more easily find data scientists to collaborate with,” Dr. Kim says. “And we can help to satisfy the huge unmet demand for data science training—for medical and graduate students, postdoctoral researchers, and faculty.”

Using Data to Benefit Patients

Data science is already integral to Einstein research. Examples of where large sets of data are yielding meaningful insights include:

Cancer treatments. A cancer-associated fibroblast (CAF) is a type of connective tissue cell that can promote tumor growth, spread, and resist therapy. Until recently, it wasn’t clear whether CAFs played a role in glioblastomas, a form of brain cancer. But by analyzing transcriptomic data from thousands of cells, Einstein scientists linked higher CAF levels in glioblastomas with worse outcomes, opening the door to new potential treatments.

Personalized medicine. Powerful machine-learning algorithms can help predict outcomes in individual patients. For example, 20% of pregnancies in patients with the autoimmune disease lupus result in an adverse outcome, such as premature birth, low birth weight, or stillbirth. Using clinical and laboratory data available from national and international studies of lupus, Einstein researchers are developing a machine-learning-based risk calculator for estimating the likelihood of pregnancy complications in individual lupus patients.

Medical imaging. Einstein researchers and clinicians are applying artificial intelligence and machine learning to data from imaging technologies, improving early detection and diagnosis, as well as the ability to predict the course of a disease and its response to therapy.

Harnessing health-record data. By analyzing the health records of thousands of patients, Einstein researchers have been able to compare the effectiveness of different treatments for chronic diseases such as asthma, diabetes, and cardiovascular disease.

Dr. Keller already has her sights set on larger goals. “We hope that the initial funding for this will generate interest and excitement for additional financial support from foundations and government organizations interested in scientific discoveries using big data. We want to grow this institute,” she says. “We envision Einstein and Montefiore as the exemplar for the power of data to advance biomedical science.”