Areas of Research

Biomedical Informatics research covers a broad spectrum of inquiry - from the analysis of genomic microarray datasets to the evaluation of hospital organizations during the adoption of new technology. This spectrum reflects the many facets of Biomedical Informatics, which can be defined as “the scientific field that deals with biomedical information, data, and knowledge – their storage, retrieval, and optimal use for problem-solving and decision-making.” (Shortliffe & Blois, 2001). Below we present our faculty’s major funded research areas. We welcome your ideas for collaboration and invite individuals interested in training with us to contact us for further information.

Biostatistics

Over many years, Roger S. Day, ScD has studied how computational and modeling tools could help people create better biological understanding, then apply it to better individual treatment decisions. The technical developments that he works on to support this goal constitute a collection of topics which all related to understanding cancer treatment better.  These topics are knowledge representation, software architecture for comprehensive modeling and validation, multi-scale modeling in cancer, strategies for overcoming drug resistance in cancer, and how pharmaceutical and biological interactions should be statistically modeled.

A major thrust of this effort is combining biomathematical models with research results and other knowledge, to better understand the natural history of cancer and develop individualized treatment strategies for cancer.  The OncoTCap provides a platform for identifying and solving problems along this... More

Clinical Informatics

Steven Handler, MD, PhD, Assistant Professor with a primary appointment in the Department of Biomedical Informatics and secondary appointments in Geriatric Medicine, and Clinical and Translational Research focuses on clinical and translational informatics in the long-term care setting.  His interests include developing and testing active medication monitoring systems to enhance the detection and response to potential adverse drug events, improving medication safety and adherence during care transitions, and the application of information technology to improve the quality, efficiency, and cost of nursing home care.

Clinical Predictive Modeling

Clinical care involves making many predictions under uncertainty, including risk assessment, diagnosis, prognosis and therapeutic management. The better those predictions can be made, the better clinical care is likely to be. The increasing availability and richness of electronic health records (EHRs) are increasing the opportunities for developing and deploying computer-based clinical prediction methods. Such methods can serve as key components of computer-based decision support systems. The data in EHRs can be used to construct prediction models using machine learning methods. Individual patient data from EHRs can also serve as input to the predictions models.

Gregory F. Cooper, M.D., Ph.D. and Shyam Visweswaran, M.D., Ph.D. are leading projects to apply artificial intelligence, machine learning, and Bayesian modeling to develop clinical prediction models from data. These projects... More

Comparative Effectiveness Research (CER)

The IOM defines CER as the generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat and monitor a clinical condition, or to improve the delivery of care. The purpose of CER is to assist consumers, clinicians, purchasers, and policy makers to make informed decisions that will improve health care at both the individual and population levels.

Dr. Richard Boyce is one of four scholars in the AHRQ-funded University of Pittsburgh's Comparative Effectiveness Research Program Scholars Program. As part of his training, he is conducting research to compare the effectiveness and safety of antidepressants used to treat elderly adults residing in nursing homes. The aims of the study include a systematic review of antidepressant... More

Data-Driven Modeling of Usual Clinical Care

Clinical care is complex and often fast paced. Preventable medical errors can and do occur, as has been well documented in recent years. Clinical guidelines and rule-based alerts provide clinical decision support that is intended to reduce medical errors. These methods are driven by expert knowledge. As such, they tend to focus on high impact areas in which medical errors are either prevalent, serious, or both. However, the coverage of such methods is relatively narrow.

We are investigating data-driven methods for helping avoid medical errors. Machine-learning methods are applied to electronic health record (EHR) data to derive computer-based probabilistic models of usual care. These models, which can be complex and time-oriented, represent the probability of various types of care being given for different types of medical conditions. The care of a current patient, as revealed by his or her EHR, is automatically compared to the model of usual care that has been... More

Datawarehouses and Repositories

Michael J. Becich MD, PhD, professor and chair of the Department of Biomedical Informatics, focuses on developing datawarehouses and data mining strategies for genomic and proteomic data derived from high throughput methodologies such as expression microarrays and tissue microarrays. His interests also include tissue bank information systems, clinical information systems and imaging repositories that are currently operating in the Pathology Department at University of Pittsburgh. He is also the leader for the University of Pittsburgh’s Cancer Biomedical Informatics Grid (caBIG) projects and the Informatics Codirector of Pitt’s Clinical and Translational Science Institute. Becich currently serves as PI or Co-PI on eight funded grants, including grants from the... More

Dental Informatics and Oral Health Translational Research

The Center for Informatics in Oral Health Translational Research (CIOHTR) at the School of Dental Medicine, University of Pittsburgh, is affiliated with DBMI. Dr. Heiko Spallek, Executive Director, and Dr. Tanja Bekhuis, Director of Oral Health Translational Methods, are members of the DBMI Training Program Faculty. They encourage multidisciplinary collaboration among trainees and faculty interested in dental informatics and translational research.

The overarching mission of the CIOHTR is to support research and education aimed at improving delivery of dental care and patient outcomes, as well as treatment of oral and maxillofacial conditions, particularly those related to systemic health. The focus is on strategies to improve the uptake and application of high quality evidence in patient care through development and testing of best practices derived from dissemination and implementation science. If you are interested in learning more, please visit... More

Federated Data Sharing for Translational Research

Advances in cancer research and personalized medicine will require significant new bridging infrastructures, including more robust biorepositories that link human tissue to clinical phenotypes and outcomes. Over the past 15 years, Dr. Rebecca Jacobson’s laboratory has created a robust and mature clinical natural language processing platform. The open-source TIES system is already deployed at numerous institutions where it is used to automatically annotate millions of clinical documents, creating rich research repositories that support the work of clinical and translational scientists. The TIES Cancer Research Network is a federated data sharing network that currently includes five institutions and is expected to grow substantially over the next several years. The TIES Cancer Research Network is poised to create a foundational informatics infrastructure linking many of the nation’s cancer centers.

Read More

Genomic and Proteomic Data: Analysis and Data Mining

Vanathi Gopalakrishnan, PhD is exploring the application of technology to the analysis of datasets from biological studies. She is fundamentally interested in technologies for data mining and discovery that allow incorporation of prior knowledge. Her research interests for the past decade have focused on the development, application, and evaluation of symbolic, probabilistic and hybrid machine learning methods to the modeling and analysis of high-dimensional, sparsely-populated biomedical datasets, particularly from proteomic profiling studies for early detection of disease. Her current research projects involve the study of novel variants of rule learning techniques for biomarker discovery, prediction and monitoring of neurodegenerative diseases, lung and breast cancers from molecular profiling studies. Methods for incorporating prior knowledge that are being researched in her laboratory include text mining and ontology construction... More

Health Informatics for the Underserved

Gerald Douglas, PhD is leading research efforts that will contribute to the elimination of health disparities, improve health care quality, encourage the adoption of personal wellness strategies, and provide support for the development and advancement of underserved communities. Dr. Douglas implemented several innovative approaches to using technology in health care, including the use of a low-power, robust and inexpensive touch screen workstation for clinicians at the point of care in Malawi, Africa. This system guides low-skilled health care workers with little or no computer experience through the diagnosis and treatment of patients according to national protocols.

Sample of Related Publications:

Landis Lewis Z, Douglas GP, Monaco V. Touchscreen task efficiency and learnability in an electronic medical record at the point-of-care. Medinfo 2010. In Press. PMID:... More

Human Computer Interaction and Evaluation

Friedman's "fundamental theorem" of biomedical informatics states that the combination of human intelligence and information resources is more powerful than human intelligence alone. Realizing the potential benefits of this combination requires careful attention to the design and evaluation of user interfaces that will help users maximize the utility of information resources. Harry Hochheiser, PhD uses techniques of contextual inquiry and prototyping to develop bioinformatics interfaces for the FaceBase project. In collaboration with Steve Handler, MD, PhD, Dr. Hochheiser is exploring human-computer interaction issues related to geriatric care, including evaluation of drug delivery systems (with support of the PA Department of Aging) and the re-engineering of rounding reports.  Claudia Mello-Thoms, PhD uses eye-tracking and cognitive... More

Image Perception Research

Medical images contain a significant amount of useful information for the diagnosis of a given patient, but they also contain even greater amounts of distractions. How do physicians separate the signal from the noise? Which type of information do they use in their decision making process? These are the primary focuses of the research of Claudia Mello-Thoms, MS, PhD, which combines principles from, among others, Cognitive Psychology, Mathematical Modeling and Machine Learning Systems to (1) create an Internet-based tutoring system that will teach novice and general radiologists the early signs of breast cancer (K01 grant, AHRQ); (2) simulate the learning behavior of Pathology residents as they read digital slides (R01 grant, NIH/NLM, PI: Crowley) and (3) develop statistical models for visual... More

Natural Language Processing and Deep Phenotyping

Precise phenotype information is needed to unravel the effects of genetic, epigenetic, and other factors on tumor behavior and responsiveness. Current models for correlating EMR data with –omics data largely ignore the clinical text, which remains one of the most important sources of phenotype information for cancer patients. Unlocking the value of clinical text has the potential to enable new insights about cancer initiation, progression, metastasis, and response to treatment.

In collaboration with Dr. Guergana Savova, Dr. Harry Hochheiser’s work will produce novel methods for extracting detailed phenotype information directly from EMR data. Dissemination of the software will enhance the ability of cancer researchers to abstract meaningful clinical data for translational research. If successful, systematic capture and representation of these phenotypes from EMR data could later be used to drive clinical genomic decision support.

Sample of Related Publications:... More

Pharmacoepidemiology

Pharmacoepidemiology may be defined as the study of the utilization and effects of drugs in large numbers of people. To accomplish this study, pharmacoepidemiology borrows from both pharmacology and epidemiology. Thus, pharmacoepidemiology can be called a bridge science spanning both pharmacology and epidemiology. (ISPE Website)

In collaboration with Dr Joseph Hanlon (Division of Geriatrics), Drs. Richard Boyce, Steve Handler, and Roger Day are investigating if there is a clinically relevant association between pharmacokinetic drug-drug  interactions (DDIs) involving psychotropic drugs and one of the most common causes of injury and morbidity among elderly nursing home (NH)  residents—falls.  The study, "The effect of known metabolic drug-drug interactions involving psychotropic drugs... More

Public Health Informatics and Biosurveillance

Several faculty members, including  Rich Tsui, PhD and Greg Cooper, MD, PhD, investigate methods for real-time detection and assessment of disease outbreaks within the Realtime Outbreak and Disease Surveillance (RODS) Laboratory. Founded in part by Michael Wagner, MD, PhD (funded by a R01 grant) and Rich Tsui PhD, and Jeremy Espino, MD, the RODS Laboratory is a biosurveillance research laboratory that is home to three large projects that work with health departments to create surveillance systems: the RODS Open Source Project, Pennsylvania RODS, and the National Retail Data Monitor (NRDM).

These projects benefit the public and also benefit the research by grounding our work in actual public health practice and by collecting surveillance data for algorithm validation and investigations into... More

Systems Biology and Computational Biology

Understanding the molecular basis of diseases from a network perspective is a research focus of Panayiotis (Takis) Benos, PhD, and Madhavi Ganapathiraju, PhD.

Dr. Benos is an associate professor in the Department of Computational and Systems Biologyand co-director of the Joint CMU-Pitt PhD Program in Computational Biology.  In particular, Dr. Benos’ group is developing algorithms to infer regulatory gene modules of proteins and non-coding RNA genes (microRNAs) that are critically associated with lung diseases, such as idiopathic pulmonary fibrosis (IPF), and tissue and organ development.

Dr. Ganapathiraju is an assistant professor in the Department of Biomedical Informatics, Intelligent Systems Program, the... More

Translational Bioinformatics

Translational bioinformatics is an emerging area in informatics focused on “the development of storage, analytic, and interpretive methods to optimize the transformation of increasingly voluminous biomedical data, and genomic data in particular, into proactive, predictive, preventive, and participatory health” (from the American Medical Informatics Association Strategic Plan http://www.amia.org/inside/stratplan/).  Within DBMI,  Xinghua Lu, MD, PhD,  Vanathi Gopalakrishnan, PhD, Gregory F. Cooper, MD, PhD, Xia Jiang, PhD and Shyam Visweswaran MD, PhD apply machine learning and statistical methods i) for biomarker discovery in high-dimensional ‘omic data such as genomic, transcriptomic and proteomic data, and ii)... More

^