In case you missed it, catch the reply of this Data Citizens: A Distinguished Lecture Series talk with Joel Saltz, MD, PhD. The talk was held on Monday, February 13, 2023, from 3:30 to 4:30 PM EST at Allen Hall, on the University of Miami Coral Gables campus. TALK TITLE: “Artificial Intelligence, Multi-Modal Analysis, Digital Health, and Clinical Informatics.”

Please NOTE: We apologize for some low audio issues at the start and the Q&A audio was not salvagable.

 

For Joel Saltz, M.D., Ph.D., data science tools can improve patient care by providing a better understanding of cancer, heart disease and other challenging medical conditions. “We are looking for new ways to identify high-risk patients at an early stage of the disease, select the right personalized treatment and predict the outcome,” said Dr. Saltz, chair and distinguished professor, Department of Biomedical Informatics; vice president for clinical informatics at the Renaissance School of Medicine; professor of pathology and the Cherith Endowed Chair at Stony Brook University in New York.

A pioneer in establishing the field of biomedical informatics, Dr. Saltz delivered a February 14 talk on “Artificial Intelligence, Multi-Modal Analysis, Digital Health and Clinical Informatics” at the Richter Library through the Data Citizens Distinguished Lecture Series sponsored by the University of Miami Institute for Data Science and Computing (IDSC) and Miami Clinical and Translational Science Institute.

Dr. Saltz has developed innovative pathology informatics methods, including the first published whole slide virtual microscope system, and computer-aided pathology diagnosis techniques. His projects at Stony Brook include developing population health visual analytics, applying ML/AI tools for patient populations, using bots for healthcare advice and natural language processing (NLP) applications to screen for aortic aneurysm clinic patients

In his Data Citizens talk, Dr. Saltz highlighted the use of digital imaging data and algorithms to identify cellular patterns of tumor infiltrating lymphocytes (TILs) in prostate cancer patients. “We can look at the molecular data, get spatial descriptions and compare with patient outcomes,” he said. “So far, we have been able to link these TIL patterns to about 220 tumors.”

Dr. Saltz added that these types of computation-based pathology biomarkers can be more precise than a pathologist’s visual observations through a microscope. To refine the AI-based algorithms, Dr. Saltz and his team are studying visual attention – where pathologists are looking when examining slide images of cancer cells. “By modeling the relationship between human attention and pathologist classification, we can improve the precision of AI tools,” he said.

Along with cellular research, Dr. Saltz is also bringing data science tools to bear on population health issues in several collaborative studies. “Education is another important aspect of our program,” he added. “Both researchers and clinicians need to understand how biomedical informatics and digital health tools can inform their decisions.”

 

Story by Richard Westlund

 

---

 

Abstract

This talk will survey rapidly emerging health-related artificial intelligence methods, AI-driven clinical informatics projects and end with a description of how health AI applications can be used to spur Computer Science research. Dr. Saltz will describe the evolution of AI methods that simultaneously analyze many types of complementary information to guide medical and population health decision-making and to create new predictive biomarkers. One key component of these methods lies in the development of methods that are able to resolve image data—Radiology, Pathology, and geospatial—into meaningfully labeled objects. Dr. Saltz will also describe progress made in this area and the application of these methods in the context of clinical data analyses. In addition, Dr. Saltz will discuss a number of AI-driven clinical informatics efforts: these encompass analysis and modeling of human attention patterns in Pathology image interpretation, aortic aneurysm detection and management, prediction of health system patient flow and emergency department congestion, algorithms to predict hospital readmission, and the development of methods for normalizing data in multi-million patient real word data studies. Finally, Dr. Saltz will address the use of digital health and biomedical informatics challenges as driving applications that can spur innovative computer science research, and he will discuss how to build synergistic multi-disciplinary teams.

About Joel Saltz, MD, PhD

Dr. Joel Saltz is a leader in research on advanced information technologies for large scale data science and biomedical/scientific research. He has developed innovative pathology informatics methods, including: the first published whole slide virtual microscope system; pioneering pathology computer-aided diagnosis techniques; and methods for decomposing pathology images into features and linking those features to cancer “omics”, response to treatment and outcome. He has broken new ground in big data through development of the filter-stream based DataCutter system, the map-reduce style Active Data Repository and the inspector-executor runtime compiler framework. He has also been an active contributor in clinical informatics, having developed predictive models for hospital readmissions, point of care laboratory testing quality assurance systems, decision support systems for electrophoresis interpretation and graphical user interfaces to support clinical data warehouse queries. Dr. Saltz has been a pioneer in establishing the field of biomedical informatics; he founded and built two highly successful departments of biomedical informatics, one at Ohio State University and one at Emory University. In 2013, he came to Stony Brook as Vice President for Clinical Informatics and Founding Department Chair of Biomedical Informatics—to create a living laboratory for biomedical informatics and to create a third unique biomedical informatics department dually housed in the School of Medicine and the College of Engineering. Dr. Saltz is trained both as a computer scientist and as a physician through the MSTP program at Duke University. He has deep experience in computer science, having served on the computer science faculties at Yale University and the University of Maryland. He completed his residency in clinical pathology at Johns Hopkins University and he is a practicing, board-certified clinical pathologist.

About Data Citizens

Data Citizens: A Distinguished Lecture Series is an ongoing course of in-depth talks by experts in the field of data science on a wide variety of topics including data visualization, big data, AI, and predictive analytics. The Data Citizens lecture series is co-sponsored by the Miami Clinical and Translational Science Institute (CTSI) and is free and open to the public.