Workshop: Sharing "Scruffy" Data
We know that bad data produces unusable results, but gone are the days of using only pristine clinical trials data to draw conclusions and make decisions. 'Scruffy' data can come in many forms: real-world data, data from wearable devices, and other non-clinical trial data sets. Even if each data set is well organized and characterized, combining data sets can create disorder. In addition, there is increased interest in sharing data with non-academic partners such as citizen scientists, patient organizations, and patients themselves. Is there a benefit to embracing 'scruffy' data, or is the goal to bring order to the data and data sharing process? What is the next frontier of dealing with the complexities of data sharing and aggregation to make sure we are obtaining meaningful results?
Director, National Library of Medicine; Interim Associate Director for Data Science, National Institutes of Health
CEO and Co-Founder, Cyft; Assistant Professor, Harvard Medical School
Harold H. Hines, Jr. Professor of Medicine, Yale School of Medicine; Director, Center for Outcomes Research and Evaluation, Yale New Haven Hospital
Associate Professor of Pathology, Brigham and Women's Hospital and Harvard Medical School; Director, Partners Laboratory for Molecular Medicine