Sponsored Content | Case Study | Artificial Intelligence | April 09, 2020

From Surviving to Thriving: How Regional Medical Imaging Enhanced Productivity and Profitability with ProFound AI for Digital Breast Tomosynthesis

Table 1. Compared to 2-D mammography, which yields four images per patient, digital breast tomosynthesis (DBT), or 3-D mammography, produces hundreds of images per patient. While this provides more information for clinicians, the exponential increase in data can result in reader fatigue and burnout, which may ultimately affect patient care.

Table 1. Compared to 2-D mammography, which yields four images per patient, digital breast tomosynthesis (DBT), or 3-D mammography, produces hundreds of images per patient. While this provides more information for clinicians, the exponential increase in data can result in reader fatigue and burnout, which may ultimately affect patient care.

As the largest independent imaging group in Michigan with 10 locations across the state, Regional Medical Imaging (RMI) has been providing superior imaging services for 35 years. Co-owner and CEO, Randy Hicks, M.D., MBA, provides both a clinical and pragmatic approach to running the business, which has led RMI to become one of the leading radiology facilities in the state, with a team of experts in women’s imaging and other subspecialties.

Hicks takes pride in ensuring that RMI features the latest in state-of-the-art technology that empowers his team to provide the best possible patient care. In 2018, RMI adopted DBT systems across all 10 of its locations. But while digital breast tomosynthesis (DBT), also known as 3-D mammography, offers a number of benefits to patients, such as enhanced breast cancer detection and fewer false positives and unnecessary recalls, it produces hundreds of images per patient, compared to 2-D mammography which typically only produces four images per patient. 

“We were starting to feel like we were on the Titanic. The exponential increase of images quickly began to take its toll on the team,” Hicks said. “Mammograms can be anxiety inducing for women, and ultimately they just want the report that says negative. With the increased workload from DBT we were taking longer to get those results to our patients. We needed a solution to manage this workload, and we needed it badly. The minute ProFound AI was FDA-cleared, we knew we had to have it at RMI.”

In December 2018, ProFound AI for DBT became the first and only software for DBT with artificial intelligence (AI) to be FDA-cleared. Designed to be used concurrently by radiologists reading 3-D mammography, the technology was trained with the latest in deep-learning AI to detect malignant soft tissue densities and calcifications with unrivaled accuracy. It is the only software of its kind that rapidly and accurately analyzes each DBT image, or slice, and provides radiologists with crucial information, such as Certainty of Finding lesion and Case Scores, which can assist in clinical decision-making and prioritizing caseloads. 

Positive clinical results from a large reader study were recently published in Radiology: Artificial Intelligence. In a large reader study, ProFound AI for DBT increased radiologists’ sensitivity by 8 percent, minimized the rate of false positives and unnecessary recalls by 7 percent, and reduced reading time for radiologists by 52.7 percent.1

 

The Paradigm Shift

Upon implementation of ProFound AI, the team at RMI immediately began to experience a number of benefits, including improved cancer detection, a reduction in biopsy rates and a reduction in unnecessary recalls. Reading time for radiologists was also substantially reduced — and the team at RMI found ProFound AI to be even more reliable than expected. 

“The paradigm really shifts when you get ProFound AI — it changes the way you read. Once you use it, it’s really quite shocking how impressive it is,” said Hicks. “The technology itself is better than humans. Once we started using ProFound AI, we realized we are finding more cancers,” Hicks stated. “It takes time to trust it — but believe me, you can trust it. ProFound AI definitely finds more cancers than we do.”

Hicks and the radiologists at RMI found the software to be particularly helpful not just in finding cancers and reducing the time to read, but also in prioritizing caseloads.

“If I am behind on mammograms, I can easily sort by which ones need to be prioritized. We can be much more efficient in sorting cases and getting things done. Speed is important and this technology really does speed our reading time up,” he added.

Hicks also found that ProFound AI improved radiologist performance as a whole across all of the radiologists in his practice. 

“The variability in which people read mammograms is real,” said Hicks. “As a large radiology practice reading 31,000 mammograms per year, we don’t want outliers when it comes to our radiologist performance. ProFound AI acts as a friend in the room for radiologists and actually helps them to learn and improve over time.”

 

The Bottom Line

As the owner and CEO of RMI, Hicks is not only concerned about the health of his patients, he is also concerned about the welfare of the business. Although adopting DBT offered better image quality and more information than ever before, the time it took to read cases was starting to affect the company’s bottom line.

“Although DBT offered improved image quality, we found it required a steep learning curve and the increased workload resulted in a case backlog that required additional staffing to combat,” Hicks mentioned. “For example, a radiologist reviewing 100 cases per day with 2-D would typically review 400 images per day, whereas a radiologist reading 100 cases per day with DBT would be required to read almost 30,000 images per day.” (See Table 1.)

“ProFound AI improved our reading efficiency and eliminated the need to hire more staff,” he added. “The cost of ProFound AI pays for itself based on improved productivity alone.”

For more information: www.icadmed.com

Reference

1. Conant, E. et al. (2019). Improving Accuracy and Efficiency with Concurrent Use of Artificial Intelligence for Digital Breast Tomosynthesis. Radiology: Artificial Intelligence. 1 (4). Accessed via https://pubs.rsna.org/doi/10.1148/ryai.2019180096

Related Content

AI has the potential to help radiologists improve the efficiency and effectiveness of breast cancer imaging

Getty Images

Feature | Breast Imaging | May 28, 2020 | By January Lopez, M.D.
Headlines around the world the past several months declared that...
An example of DiA'a automated ejection fraction AI software on the GE vScan POCUS system at RSNA 2019.

An example of DiA'a automated ejection fraction AI software on the GE vScan POCUS system at RSNA 2019. Photo by Dave Fornell.

News | Ultrasound Imaging | May 26, 2020
May 12, 2020 — DiA Imaging Analysis, a provider of AI based ultrasound analysis solutions, said it received a governm
a Schematic of the system. The entire solid tumour is illuminated from four sides by a four-arm fibre bundle. A cylindrically focused linear array is designed to detect optoacoustic signals from the tumour. In vivo imaging is performed in conical scanning geometry by controlling the rotation and translation stages. The sensing part of the transducer array and the tumour are submerged in water to provide acoustic coupling. b Maximum intensity projections of the optoacoustic reconstruction of a phantom of pol

a Schematic of the system. The entire solid tumour is illuminated from four sides by a four-arm fibre bundle. A cylindrically focused linear array is designed to detect optoacoustic signals from the tumour. In vivo imaging is performed in conical scanning geometry by controlling the rotation and translation stages. The sensing part of the transducer array and the tumour are submerged in water to provide acoustic coupling. b Maximum intensity projections of the optoacoustic reconstruction of a phantom of polyethylene microspheres (diameter, 20 μm) dispersed in agar. The inset shows a zoomed-in view of the region boxed with a yellow dashed line. In addition, the yellow boxes are signal profiles along the xy and z axes across the microsphere centre, as well as the corresponding full width at half-maximum values. c Normalized absorption spectra of Hb, HbO2 and gold nanoparticles (AuNPs). The spectrum for the AuNPs was obtained using a USB4000 spectrometer (Ocean Optics, Dunedin, FL, USA), while the spectra for Hb and HbO2 were taken from http://omlc.org/spectra/haemoglobin/index.html. The vertical dashed lines indicate the five wavelengths used to stimulate the three absorbers: 710, 750, 780, 810 and 850 nm. Optoacoustic signals were filtered into a low-frequency band (red) and high-frequency band (green), which were used to reconstruct separate images.

News | Breast Imaging | May 26, 2020
May 26, 2020 — Breast cancer is the most common cancer in women.
 Recently the versatility of mixed and augmented reality products has come to the forefront of the news, with an Imperial led project at the Imperial College Healthcare NHS Trust. Doctors have been wearing the Microsoft Hololens headsets whilst working on the front lines of the COVID pandemic, to aid them in their care for their patients. IDTechEx have previously researched this market area in its report “Augmented, Mixed and Virtual Reality 2020-2030: Forecasts, Markets and Technologies”, which predicts th

Doctors wearing the Hololens Device. Source: Imperial.ac.uk

News | Artificial Intelligence | May 22, 2020
May 22, 2020 — Recently the versatility of
Phone call and linkage-to-care-based intervention increases mammography uptake among primary care patients at an urban safety-net hospital

Getty Images

News | Mammography | May 22, 2020
May 22, 2020 — Telephone outreach coupled with scheduling assistance significantly increased...
In response to the significant healthcare delivery changes brought on by COVID-19, Varian has launched new capabilities for its Noona software application, a powerful tool designed to engage cancer patients in their care for continuous reporting and symptom monitoring.
News | Radiation Oncology | May 21, 2020
May 21, 2020 — In response to the significant healthcare delivery changes brought on by...
NucleusHealth, a provider of cloud-based medical image management technology and teleradiology services, announced today that it has received Conformité Européene (CE) Mark approval for Nucleus.io.
News | Teleradiology | May 21, 2020
May 21, 2020 — NucleusHealth, a provider of cloud-based medical image management technology and teleradiology service
The Breast Imaging and Reporting System (BI-RADS) was established by the American College of Radiology to help classify findings on mammography. Findings are classified based on the risk of breast cancer, with a BI-RADS 2 lesion being benign, or not cancerous, and BI-RADS 6 representing a lesion that is biopsy-proven to be malignant.

Getty Images

News | Breast Imaging | May 19, 2020
May 19, 2020 — Women with mammographically detected breast lesions that are probably benign should have follow-up sur