This talk is part of the CCBBI seminar series and will start at 1pm in this zoom room: 

 https://osu.zoom.us/j/91663959745?pwd=VVZKaE81Ty8zZjBUSlZnNjRoOWNVQT09

Title: Better biomarkers based on quality, reproducible and open science

Abstract: Mental health is a major public health challenge currently costing trillions of dollars, for which neuroscientific approaches are key to advancing the understanding of mechanisms, risk factors, biomarkers and to improve treatments. The rapid adoption of data sharing and open science practices present an unprecedented opportunity to offer better care at a lower cost. However, the quality and efficacy of potential biomarkers and treatments is critically dependent on the quality of multiple stages of data science. These stages include but are not limited to preprocessing, quality control, feature extraction, model building, and performance evaluation. Often overlooked, inaccuracies at these stages can get multiplied to produce suboptimal or irreproducible final results (so-called “garbage-in, garbage-out” and butterfly effects). Much research so far targeted only one of the aforementioned specific issues. Such studies require statistical comparison of relative efficacy to decide among many available options (e.g., biomarkers or treatment plans). However, due to the quality issues at each of the aforementioned stages, we haven’t fully leveraged the benefits from deep integration and proper application of machine learning and data science. In this talk, I present an outline of the research program at the Open MINDS lab to solve these challenges in two key areas: neuroimaging quality control and biomarker performance evaluation. I encourage you to visit the following websites to learn more: niQC SIG and crossinvalidation.com.