Hello! I'm Eric Bridgeford. I am a postdoctoral scholar at Stanford University working towards developing statistical methods for neuroscience research. I am interested in end-to-end decision making in biological datasets, summarized in part by the below figure:
In many fields such as neuroimaging with expansive datasets, translating insights into clinical practice has proven problematic at best. I believe that disparities that arise across data collection sites in the form of batch effects, difficulties tuning analyses to properly adjust for potential confounding biases, difficulties identifying best practices for the construction of data derivatives (data pre-processing), and difficulties understanding how insights will translate when other pre-processing steps are taken, may be a potential source of this problem.
I completed my graduate education at Johns Hopkins University in the Department of Biostatistics. I was be supervised by Dr. Brian Caffo and Dr. Joshua Vogelstein. My work focused on statistical methods for understanding measurement error in connectomes.
0%: Beginner, 50%: Proficient, 100%: Expert
With Dr. Bassett and Dr. Muldoon, I aided in the development of a novel network statistic, the Small World Propensity. Our statistic provides an extension of small worldness to weighted graphs, a common element of real-world networks, and robustness to changes in edge density.
With the NeuroData Team, I worked with Dr. Vogelstein to develop statistics for quantifying reliability in multi-scan scientific data and a processing pipeline for functional MRI connectomics to extend our existing pipeline for diffusion connectomics.
With Dr. Vogelstein, Dr. Caffo, and Dr. Priebe, I extended our MRI connectomics pipeline for hyperparallelizability on AWS architecture and developed theoretical methods/software packages for manifold embedding techniques, independence testing, and graph statistics.
As the Vice President of Engineering, I was responsible for contributing technological vision, coauthoring technical documents and patents, and managing the software and data analytics teams. I also managed the delivery of internal technical projects. The company has since re-focused as Trimwire by Atana, LLC.
With Dr. Vogelstein, Dr. Caffo, and Dr. Priebe, I extended our MRI connectomics pipeline for hyperparallelizability on AWS architecture and developed theoretical methods/software packages for manifold embedding techniques, independence testing, and graph statistics.
In my free time, I enjoy playing guitar in 80s hair metal bands, building and tinkering with guitar-related equipment, hiking, and rock climbing.