Po-Hao Chen MD, MBA, Presenter: Nothing to Disclose

Alexander T. Ruutiainen MD, Abstract Co-Author: Nothing to Disclose

Howard Lee Roth MD, Abstract Co-Author: Nothing to Disclose

Tessa S. Cook MD, PhD, Abstract Co-Author: Nothing to Disclose

The premise of rapid reinforcement learning theory posits that education is a feedback loop consisting of rapid, repetitive performance paired with immediate, unambiguous feedback.  Software design literature cites this mechanism as the source of enjoyment one receives from otherwise banal activities such as matching tiles of arbitrary shape into straight lines (also known as Tetris) or propelling wingless avians against predetermined wooden structures (Angry Birds).   Identifying basic abnormalities such as neuroforaminal narrowing on MRI or architectural distortions on mammography are skills that may be best improved by practice.  We created a web-based platform and API to allow radiologists to easily create or use modules using game-like mechanics to encourage trainees to practice these basic skills.

Our web-based application is designed in PHP, with visual effects written using JavaScript and is compatible with a variety of browsers, including iOS- or Android-powered products.  Users can choose to complete a learning module of their choice. Each module contains a focused learning objective such as 'degenerative disc disease on MRI.'  Each module contains a set of brief cases, each designed to take 15 seconds or less to complete, immediately followed by visual feedback.  Mechanics such as a countdown timer, score, achievement, and leaderboard are available as motivational tools.  Educators can design rapid-reinforcement modules and choose to either publically share or keep them private. Users may consent to provide anonymous data for quality metrics on each module.

Our application should provides a safe virtual environment where trainees are encouraged to learn basic identification skills by immediate feedback.    Its mobile-friendly nature, along with the simplicity of each module's objective, is amenable to short bursts of learning on-the-go as well as dedicated learning sessions.  The data-recording mechanism provides feedback for educator-researchers on the efficiency and quality of their modules.

A cross-platform, open-source, web-based application can be used to motivate radiology trainees in honing basic image interpretation skills.

http://abstract.rsna.org/uploads/2014/14006064/14006064_ikyf.jpg

Chen, P, Ruutiainen, A, Roth, H, Cook, T, A Web-based Open Source Platform for Radiology Education using Rapid Reinforced Learning Mechanics.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14006064.html