Studies presented at major radiology conferences demonstrate the efficacy
of artificial intelligence solutions across key value points for radiology
CureMetrix, Inc., a global healthcare technology company that develops AI-driven software for radiology, has presented research and solution overviews highlighting the value of cmTriage™, cmAssist®, cmDensity™, and cmAngio™ for clinical practices. These AI solutions for mammography were the focus of presentations at four major radiology conferences: the European Congress of Radiology (ECR); the Society of Breast Imaging (SBI) Symposium; the National Consortium of Breast Centers Annual Interdisciplinary Breast Center Conference (NCoBC); and the AuntMinnie.com Spring 2021 Virtual Conference.
“We are grateful to have had the opportunity to present our findings this year at some of the most distinguished gatherings of radiology professionals in our field,” said Navid Alipour, chief executive officer of CureMetrix. “From streamlining worklists, to diagnostic support, to reporting and extending the value for women’s health, AI is making a serious impact in radiology today.”
SBI (April 9-11, 2021)
Alyssa Watanabe, MD, FACR, delivered an abstract presentation, entitled “Superior Reliability of an Artificial Intelligence-based Breast Density Classifier based on BI-RADS 5th Edition.” This retrospective study assessed a semi-supervised learning (SSL)-based model for mammography, and how it can improve reliability and consistency in density assessments in radiologist reports. The study used cmDensity to read mammograms, and compared the program’s consistency with the findings of seven board-certified radiologists with deep experience reading mammograms, and using a variety of quantitative density assessment products.
Key takeaways from this clinical abstract presentation:
- cmDensity, which uses an SSL model for its algorithm, does not use human labels for training, therefore eliminating any subjective human bias;
- While the human readers’ classifications were highly variable, cmDensity consistently predicted the correct A-D density categorization 100% (A), 100% (B), 93% (C) and 100% (D) of the time, respectively.
NCoBC (April 16-19, 2021) – Award Winning Poster
Tara Retson, MD, PhD, of the UC San Diego Altman Clinical and Translational Research Institute, teamed up with Dr. Watanabe and Vivian Lim, MD, Diagnostic Radiologist in Breast Imaging at UC San Diego Health to deliver, “High Performance of an FDA-Cleared Platform for Mammography Triage.” The presentation and award-winning poster highlighted the synergistic ways in which radiologists and AI can work together to improve mammography and patient outcomes.
Key takeaways from this poster and presentation:
- AI and radiologists have a synergistic relationship with AI-based solutions outperforming traditional CAD
- By drawing attention to suspicious exams rather than offering a diagnosis, AI-based triage (cmTriage) may provide positive reader bias to improve accuracy and reduce workload
- AI can also increase radiologist efficacy in assessing dense breasts and provide support as a second reader
AuntMinnie.com AuntMinnie Europe Spring 2021 Virtual Conference (May 5-6, 2021) – CureMetrix Platinum Sponsor
At the AuntMinnie Spring 2021 Virtual Conference, a clinical panel of AI experts and practicing physicians delivered a solutions overview presentation entitled “Groundbreaking AI for Radiology: A Dialogue on Value for Doctors.” Physician panelists were:
- Alyssa Watanabe, MD, FACR, Chief Medical Officer of CureMetrix and associate clinical professor of radiology at the USC Keck School of Medicine
- Lina Le, MD, Director of Breast Imaging at Shin Imaging Center and Anaheim Regional Medical Center
- Marie Tartar, MD, VP of Medical Affairs at CureMetrix and radiologist at Scripps Health
- Matt Budoff, MD, FACC, FAHA, Professor of Medicine at UCLA’s Geffen School of Medicine and Director of Cardiac CT at Harbor-UCLA Medical Center.
The panelists discussed the findings of studies to assess the efficacy of cmTriage and cmAssist, and changing patient attitudes toward AI. They also shared case studies and anecdotal insights about the impacts of those products, as well as the investigational AI solutions cmDensity and cmAngio on their clinical practice.
Key takeaways from this presentation:
- AI tools work across several value points in radiology to save time, improve sensitivity, improve specificity, automate reporting and save lives;
- The use of cmTriage could helps radiologists read mammograms 30 percent faster, on average;
- Use of cmAssist can lead to a 27 percent increase in cancer detection, while reducing false positives by 69 percent over traditional CAD.
- AI-driven cmDensity can automate breast density classification A-D based on the BIRADS 5th Edition and reduce inter- and intra-reader variability
- cmAngio can support a doctor in assessing a patient’s risk of coronary heart disease (CHD), from the same mammogram used to screen for breast cancer; this technology provides added value for radiology and aids healthcare systems in combating the number one killer of women – heart disease.
- And from the consumer’s perspective, women surveyed would choose and recommend an imaging center that uses AI as part of their mammography program.
“AI is not only improving radiologists’ ability to quickly and accurately detect breast cancer earlier—it’s also transforming the mammogram exam as an aid for women’s health,” said Dr. Alyssa Watanabe, chief medical officer of CureMetrix. “CureMetrix AI solutions can triage and sort a radiologist’s worklist, act as a second pair of eyes while reading a mammogram, classify breast density more consistently and utilize a mammogram reading to help detect early signs of coronary heart disease. I look forward to continuing the conversations we began at these conferences, and showing more radiologists what AI from CureMetrix can do for their clinical practices.”
cmTriage™, the first FDA-cleared AI-based triage software in the U.S., helps radiologists prioritize the most suspicious mammogram cases first, directly sorting the doctor’s worklist, allowing for case triage and faster turnaround. It uses the power of AI to change the old “first in, first out” protocol of mammography reading. cmTriage works across all breast densities, and works with both masses and calcifications.
cmAssist® an investigational AI-based CAD software that helps radiologists identify, mark and score regions of interest on diagnostic and screening mammograms. Once cmAssist flags and marks possible anomalies, the software also generates a unique, data-driven neuScore™ which provides the radiologist a quantitative measure of suspiciousness of a marked region of interest, ranging from 0 (least suspicious) to 100 (highly suspicious). This score can be tracked over time to evaluate the stability of identified anomalies.
cmDensity™ is an investigational AI-based triage software that helps radiologists classify a patient’s particular type of breast density accurately and consistently. Dense breast tissue can obscure cancers on mammograms, increasing the difficulty of detection, and breast density reporting to patients is now required by federal law. cmDensity helps doctors quickly classify a patient’s breast density using a four-level density scale: A. (Almost entirely fat); B. (Scattered fibroglandular densities) ; C. (Heterogeneously dense); and D. (Extremely dense).
cmAngio™ is an investigational AI-based triage software that helps doctors assess a patient’s risk of coronary heart disease (CHD) from a mammogram. The software produces a CureMetrix Bradley Score™ based on the existence and characteristics of breast arterial calcifications in the mammogram. This proprietary score can then be used to help the care team assess the patient’s risk of CHD directly from their mammogram. The Bradley Score and their associated risk of CHD can be tracked over time to help patients and doctors understand their changing health profile.
CureMetrix is a global leader in artificial intelligence (AI) for medical imaging, committed to the advancement of technology that improves disease detection and increases cancer survival rates worldwide. With research that leverages artificial intelligence (AI) and deep learning to develop the next generation of medical image analysis, CureMetrix delivers technology that radiologists, healthcare systems and patients can confidently rely on. For more information, please visit www.curemetrix.com.