As breast imaging continues shifting toward measurable, outcomes-based care, leading organizations are turning to advanced AI tools to ensure consistent performance. Derry Imaging, with seven locations in New Hampshire, has embraced this approach by integrating Lunit® AI as a cornerstone of its breast imaging quality strategy.
This has resulted in a highly standardized, data-driven program designed to optimize clinical performance, improve diagnostic accuracy, and elevate patient care.
Maintaining consistent image quality across multiple locations remains one of breast imaging's greatest challenges. Their staff consists of 16 members of the Breast Health team and seven radiologists. Rather than relying solely on subjective image review, Derry Imaging implemented Lunit AI and built a comprehensive quality program that combines on-the-job technologist feedback, objective performance analytics, and standardized breast density assessment. The result is a continuous quality improvement process spanning image acquisition through interpretation.
To create a consistent quality program, Derry Imaging focused on three key stages of breast imaging:
Derry Imaging was first in the world to incorporate AI insights on-the-job for their technologists with Volpara Live. Each image is analyzed, and if positioning or compression can be improved, technologists can make adjustments during image acquisition. Marisa Larouchelle, Director of Clinical Operations at Derry Imaging states, “Live allows you to fix what’s needed while the patient's there and helps avoid the worry and inconvenience of a technical call-back. Adding Volpara Live has nearly eliminated technical callbacks. Since its implementation, we’ve seen a progressive decline, with technical callbacks now accounting for less than 0.5% of exams.”
For the imaging center, reducing technical recalls increases capacity by freeing appointment slots that would otherwise be used for repeat exams. “This allows us to serve more patients, improve access to care, and make more effective use of our imaging resources.”
Analytics grades each image using PGMI quality assessment metrics while providing the objective data behind each score. This helps both technologists and managers track performance and address opportunities for improvement. Analytics enables Derry Imaging to benchmark performance at both the individual and organizational level. By establishing clear standards and providing measurable feedback, the platform supports a more uniform approach to breast imaging.
This level of standardization is essential in reducing variability, which is directly linked to improved diagnostic confidence and more reliable patient outcomes.
Consistent use of Volpara Live and Analytics has produced measurable quality improvements. Lunit can compare users of their Analytics software worldwide and rank performance of individuals and institutions across the globe. In recent months, Derry Imaging has ranked as high as third globally for overall image quality, and second globally for positioning performance. “Our technologists’ overall Analytics scores have continued to improve year over year since 2021, further demonstrating the impact of our ongoing investment in technology, training, and mammography quality”, says Marisa.
In addition to image quality, Derry Imaging also uses Volpara Scorecard to standardize breast density assessment. Because breast density can influence both cancer detection and recommendations for supplemental screening, consistent density assessment is an important component of high-quality breast imaging.
Volpara Scorecard provides objective volumetric breast density measurements and standardized density categorization, helping reduce inter-reader variability while supporting risk assessment and density notification requirements.
Combined with high-quality image acquisition, Volpara Scorecard provides radiologists with standardized breast density information that supports more consistent interpretation and confident clinical decision-making.
By embedding these capabilities into the clinical workflow, Derry Imaging provides radiologists with objective data that complements their expertise and supports greater diagnostic confidence.
While Lunit AI operates behind the scenes, its impact is directly experienced by patients. Improved image quality reduces the likelihood of repeat exams and unnecessary callbacks, streamlining the diagnostic process.
Derry Imaging’s investment in AI-driven quality helps reduce variability across the imaging process—from image acquisition to interpretation—creating a more consistent experience for both patients and clinicians. It reflects their broader commitment to delivering accurate, efficient, and patient-centered care.
As healthcare continues to emphasize value and accountability, Derry Imaging’s approach demonstrates how imaging providers can leverage advanced AI. Lead radiologist Dr. Randy Tauro said, "As a radiologist, having consistently high-quality images is essential because it allows me to focus on interpretation rather than questioning the technical quality of the exam. Pairing those images with an objective, standardized breast density assessment provides another layer of consistency that supports confident clinical decision-making and more uniform reporting across our practice."
By integrating AI across image acquisition, quality management, and breast density assessment, Derry Imaging has created a scalable, performance-driven program that aligns clinical excellence with measurable outcome. This is an example of the future of breast imaging.
Disclaimer: This customer story reflects the experience and opinion of the clinicians and may not be representative of all users. The manufacturer has not independently verified the efficiency results described.
Copyright © Lunit International Limited 2026. All rights reserved.
For full disclaimers, click here and refer to section 1. - and link the click here to: https://microsite.lunit.io/lunit-insight-us-disclaimers
Lunit International Limited, Level 4, 1 Victoria Street Wellington Central, Wellington, 6011 New Zealand
