BMI conducts research on techniques, tools, and systems that synthesize information to obtain a deep understanding of mechanisms underlying health and disease. BMI faculty engage in cutting-edge projects designed to contribute to integrative biomedical and translational research to enhance healthcare. These projects are carried out through synergistic activities with biomedical researchers and healthcare teams at Stony Brook and elsewhere in the U.S. and abroad. Research Areas include:
Computational and Systems Biology
This research area takes an interdisciplinary approach to the study of computation and systems-level analysis to resolve critical biological challenges. Researchers in this area focus on several computational and biologic components involving network analysis and reconstruction, modeling, algorithm development and genomic influence.
Faculty: Daifeng Wang
Faculty in this area develop methods and tools for the analysis, visualization, and translation of genomic data. Projects range from developing predictive models of response to therapy in human cancer to understanding gene regulation in model organisms.
Faculty: Daifeng Wang
The Electronic Health Record (EHR) is the primary means for managing and storing a patient’s data. There has been a growing push to utilize data collected from EHR systems for secondary use, including automated computation of clinical phenotypes to improve accuracy of records, generation of real-time alerts to improve patient care, and analysis of records to advance our understanding of disease and healthcare and to improve the quality of care.
Faculty: Janos Hajagos, Elinor Schoenfeld, Mary Saltz, Jonas Almeida, Joel Saltz
With the ever rise of available information often denoted as “big data”, this research takes a mathematical approach to provide actionable insights and our faculty work on the intersection of statistics, mathematics, and computer science. Another focus of our research is to develop and evaluate a suite of novel data and processing abstractions and optimizations for analysis of extremely large low-dimensional spatio-temporal data for scientific, biomedical, and clinical research. We develop and use quantitative models and methods to assist related research areas that are cross-platform and used on a variety applications.
Faculty: Daifeng Wang, Tahsin Kurc, Jonas Almeida, Joel Saltz
Imaging data is employed in care guidelines and clinical settings for virtually all cancer disease sites. With widespread adoption of digitizing microscope platforms, tissue images obtained from whole slide tissues or tissue microarrays are increasingly employed in clinical and research settings. Specific areas of research and development include deep learning methods for image classification and segmentation, image segmentation and feature computation pipelines, software for management and interrogation of image data and imaging features, high performance and Cloud computing methods and tools for image analysis and data management.
Faculty: Tahsin Kurc, Fusheng Wang, Joel Saltz
Population Healthcare Informatics
We are developing an integrated data gathering, integration, predictive-modeling and information-dissemination infrastructure for large scale population based studies, with the goals to reduce the risk of undesirable events like potentially avoidable emergency department visits and hospital-acquired infections, and to improve patient quality indicators. We also work on large scale integrative big spatial analytics for public health studies, by using healthcare data from open government initiatives, social media, and public data sources, with a goal for community or neighborhood level research. In particular, we work on big data driven studies for opioid epidemic.
Faculty: Elinor Schoenfeld, Mary Saltz, Jonas Almeida, Janos Hajagos, Fusheng Wang, Joel Saltz
This research involves developing control theoretics and theory towards medial imaging systems. An example of this research includes the study of human computer interaction as it pertains to incorporate physicians (pathologists) in existing semi-automated imaging tasks in order to reduce workload reduction, increase efficiency, and bridge potential information gaps.
Faculty: Daifeng Wang