- Location
- Room 306, York University Leonard G. Lumbers Building (LUM), 115 Ottawa Road, Toronto, ON
- Series/Type
- U of T Community Event
- Format
- Hybrid
- Dates
- April 29, 2024 from 3:00pm to 4:30pm
Links
Join us at the Statistical Sciences Applied Research and Education Seminar (ARES) with
Larissa Stanberry
Program Director
Data Science and Investigator Initiated Research
Minneapolis Heart Institute Foundation
Free Hybrid (In-person/Online) Event | Registration Required
Talk Title
Clinical Prediction Models – Signal or Noise? A Case of Heart Failure
Abstract
Medical field is abuzz with artificial intelligence, that is disrupting health care by transforming its many aspects from image analysis and drug discoveries to patient monitoring, to healthcare operations and public health initiatives. In clinical research, this impact is felt through the steady increase in the number of clinical prediction models, proclaiming novel predictors and promising superior accuracy, intuitive use, and drastic improvements in patient outcomes and resource allocation. This increase is due not only to growing availability of healthcare data and developments in analysis methodology, but also, and in no small part, to advances in modern software lowering the barriers to entry. The democratization of technology and the advancement of user-friendly tools are allowing researchers with varying skill sets to try their luck in developing clinical prediction models.
We conducted a systematic review of research publications in PubMed 2018 – 2023 in the field of heart failure that were presented as developing clinical prediction models by their authors. The abstracted data elements were based on those identified in PROBAST (Prediction model Risk of Bias ASsessment Tool) and TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis). We evaluate the methodological conduct of the studies and present the sentiment analysis to estimate the prevalence of subjective or promotional language in the abstract corpus.
Speaker Profile
Dr. Stanberry is an experienced statistician and a program director with professional focus on bridging the gap between biomedical research and clinical practice. She completed her PhD in Statistics at the University of Washington in Seattle. Dr. Stanberry leads a cardiovascular research program at the Minneapolis Heart Institute Foundation, a non-profit research institute in Minneapolis, Minnesota. Her professional focus is on advancing clinical research through rigorous statistical treatment of data. She has authored and contributed to many scientific publications. Dr Stanberry also serves as a statistical editor of top tier research journals in cardiovascular field (JACC Heart Failure and JACC Advances) and NASA Human Research Program.
