At the University of Washington, eScience Data Science Fellow and Research Assistant Professor of Psychology Ariel Rokem and UW Data Science Postdoctoral Fellow Adam Richie-Halford have created a way for the general public to help an algorithm learn to read MRI scans. Fibr utilizes the vast dataset of the Healthy Brain Network to better understand how mental health disorders are first diagnosed in childhood and adolescence. But in order for the algorithm to differentiate between scans that show long-range fiber connections in the brain and those that don’t, it must first learn what to look for. Regardless of scientific training, anyone who wants to participate can view a short tutorial and start guiding Fibr towards new innovations in neuroscience and beyond.

With thousands of MRI scans to analyze, the individual attention that works on a small scale becomes impractical on a much larger one. In order to train a computer algorithm to correctly identify poor quality data or failures in the preprocessing pipeline, “we need labelled data to train the quality control algorithms,” said Richie-Halford. When you log into the Fibr website, you have the opportunity to view the anonymized MRI data and swipe left or right, depending on whether the MRI scan displayed shows the right pattern of  long-range fiber connections. This public input provides the training data that helps guide the quality control algorithm to identify good or bad dMRI data. Rokem and Richie-Halford have formatted Fibr as a game with a simple tutorial so that anyone can participate, regardless of scientific training or previous experience.

Read the full article on the eScience Institute website here.