Background: With the implementation of the Mammography Quality Standards Act in 1992, mammography technologist positioning training became widely available with strong support from the ACR and other organizations. Since the advent of FFDM in 2001 and then DBT in 2011, very little emphasis has been given to a modify training techniques to adapt to these new modalities. In 2016, the FDA reported that poor positioning techniques decrease the sensitivity in mammography by 18.1% . At the same time, hands-on positioning training became scarcer with the introduction of internet learning. Evaluation of image quality has been subjective, which can be widely variable. Mammography positioning techniques continue to be the primary reason for ACR accreditation failure1. Consistent and reproducible methods of positioning have been taught privately since that time with marked improvement noted . However, there have not been any large, prospective studies that demonstrate the impact of positioning training techniques on mammographic image quality utilizing objective and automated measurement methods, which are now widely available through the commercialization of AI. These techniques, combined with technologist-specific performance data to target training, shows great promise to improve and maintain high levels of mammographic image quality
Learning Objectives: 1. Demonstrate how standardized techniques can be customized by technologist-specific performance data and how resulting image quality can be automatically and objectively measured to validate improvement. 2. Evaluate the impact of targeted positioning training on mammographic image quality at an individual and overall facility level. 3. Based on these results, recommend a new methodology for a targeted positioning training approach based on objectively measured image quality data for dissemination at facilities throughout the country.
Abstract Content/Results: We plan to demonstrate this new approach using AI driven performance data combined with technologist-specific training plans and standardized techniques. To evaluate the impact of this approach on mammographic image quality, we will measure the change in positioning quality scores, pre and post targeted training. These scores will include measures of overall quality as well as ACR positioning metrics, i.e., visualization of the IMF and increase in the PNL measurement.
Conclusion: This work intends to validate that a targeted positioning training approach can result in measurable mammographic image quality improvements. As this approach is based on automated and objective measures and standard techniques, it lends itself to become a nationwide standard.
