Chest imaging plays a pivotal role in screening and monitoring patients, and various predictive artificial intelligence (AI) models have been developed in support of this. However, little is known about the effect of decreasing the radiation dose and, thus, image quality on AI performance. This study aims to design a low-dose simulation and evaluate the effect of this simulation on the performance of CNNs in plain chest radiography. Seven pathology labels and corresponding images from Medical Information Mart for Intensive Care datasets were used to train AI models at two spatial resolutions. These 14 models were tested using the original images, 50% and 75% low-dose simulations. We compared the area under the receiver operator characteristic (AUROC) of the original images and both simulations using DeLong testing. The average absolute change in AUROC related to simulated dose reduction for both resolutions was <0.005, and none exceeded a change of 0.014. Of the 28 test sets, 6 were significantly different. An assessment of predictions, performed through the splitting of the data by gender and patient positioning, showed a similar trend. The effect of simulated dose reductions on CNN performance, although significant in 6 of 28 cases, has minimal clinical impact. The effect of patient positioning exceeds that of dose reduction.
LINK
The seventh ACL Research Retreat was held March 19–21, 2015, in Greensboro, North Carolina. The retreat brought together clinicians and researchers to present and discuss the most recent advances in anterior cruciate ligament (ACL) injury epidemiology, risk factor identification, and injury risk screening and prevention strategies. Subsequently, our goal was to identify important unknowns and future research directions. The ACL Research Retreat VII was attended by 64 clinicians and researchers from Australia, Canada, India, Ireland, the Netherlands, South Africa, the United States, and the United Kingdom. The meeting featured 3 keynote and 29 podium presentations highlighting recent research. Keynotes were delivered by Bruce Beynnon, PhD (Univer- sity of Vermont), Charles ‘‘Buz’’ Swanik, PhD, ATC (University of Delaware), and Mark Paterno, PhD, PT, ATC, SCS (Cincinnati Children’s Hospital Medical Cen- ter), addressing their ongoing work related to sex-specific multivariate risk factor models for ACL injury,1 the role of the brain in noncontact ACL injury,2 and the incidence and predictors of a second ACL injury after primary ACL reconstruction and return to sport,3 respectively. Podium and poster presentations were organized into thematic sessions of prospective and case-control risk factor studies, anatomical and hormonal risk factors, neuromuscular and biomechanical risk factors, injury risk assessment after ACL injury, and injury-prevention strategies. Time was provided for group discussion throughout the conference. At the end of the meeting, attendees participated in 1 of 3 breakout sessions on the topics of genetic, hormonal, and anatomical risk factors; neuromechanical contributions to ACL injury; and risk factor screening and prevention. From these discussions, we updated the 2012 consensus state- ment4 to reflect the most recent advances in the field and to revise the important unknowns and future directions necessary to enhance our understanding of ACL injury. Following are the updated consensus statement, keynote presentation summaries, and free communication abstracts organized by topic and presentation order.
