Karl Helmer, PhD, maintains research interests in several areas: (1) the creation of data management and sharing infrastructure; (2) the development of imaging protocols for multi-site MR imaging-based studies; (3) the creation of quality control and assessment software for MRI data; and (4) development of terminologies/ontologies to describe neuroscience experiments and semantic-web-based tools to annotate data. He also has an ongoing research project to study the neural underpinnings of musical improvisation and expert performance.

Dr. Helmer is currently chiefly active in the field of biomedical informatics and is currently directing the data management efforts for multiple projects. He is currently the Director of the Data Core for the MarkVCID project. His role on this project is twofold: (1) design, contribute to and direct the building of a data management infrastructure for the project; and (2) work with project investigators to create harmonized imaging protocols for each biomarker that can be used across the sites involved in this consortium. In addition, he is designing, directing the development of, and contributing to a data-sharing system called Entrepôt™. This system is able to collect data from multiple sites, curate it, and make it available to users for reuse. It will mainly be used for medium-to-small projects, such as Pharma-sponsored drug trials. The system is currently being built to house the data from three funded projects. He and his team also investigate the use of Natural Language Processing to extract information from supporting material to better characterize and understand primary data.

Education

PhD in Physics, University of Rochester

Select Publications

1. Helmer KG, Dardzinski BJ, Sotak CH. The application of porous-media theory to the investigation of time-dependent diffusion in in vivo systems. NMR Biomed. 1995 Nov-Dec;8(7-8):297-306.

2. Helmer KG, Chou MC, Preciado RI, Gimi B, Rollins NK, Song A, Turner J, Mori S. Multi-site Study of Diffusion Metric Variability: Characterizing the Effects of Site, Vendor, Field Strength, and Echo Time using the Histogram Distance. Proc SPIE Int Soc Opt Eng. 2016 Feb 27;9788. pii: 97881G.

3. Warner GC, Helmer KG. Characterization of Diffusion Metric Map Similarity in Data From a Clinical Data Repository Using Histogram Distances. Front Neurosci. 2018 Mar 8;12:133.