We’ll be attending the SBI/ACR Breast Imaging Symposium 2021 that will take place virtually April 9 through 11. We would love to meet with you online in our virtual booth to introduce you MammoScreen®, our award-winning AI-guided decision-making support for breast cancer screening. Chat with us live, view our new animated video or schedule a demo request all from the comfort of your computer.
Free learning lab opportunity
In addition to our virtual booth at SBI, we are excited to be hosting a free learning lab for attendees and non-attendees alike. Even if you can’t make it to SBI’s Breast Imaging Symposium, you can still join us for this free webinar:
The AI Elephant in the Reading Room
Saturday April 10, 3:30 pm – 4:30 pm (ET)
Artificial intelligence (AI) for mammography can’t be ignored. It won’t take your job, but it raises a number of questions: What is it like? How easy is it to use/understand? What is the best way to evaluate and use it in practice? Will it become the standard of care? Participate in our learning lab, where you will discover:
● The differences between AI and CAD
● What to evaluate about AI
● How AI can benefit you and your patients
In this learning lab, hear from your peers and experts sharing their experience using AI. This will be an interactive session with questions and answers throughout.
Interested? Register to attend our Learning Lab: https://app.livestorm.co/therapixel-inc/the-ai-elephant-in-the-reading-room?type=detailed
Don’t miss our oral presentation
We are proud to share that our scientific abstract was among 18 that were chosen by the Society of Breast Imaging to be orally presented at the conference. Join us to listen to our presentation:
Potential Benefits of an AI System in the Early Detection of Breast Cancer
Friday, April 9, 2021, 3:15 PM – 4:15 PM (ET)
The purpose of this study was to investigate MammoScreen®’s ability to detect breast cancer earlier. To do this, 16,004 patients who presented for screening at one medical center in the US were included in the study. For each included patient, 3 consecutive screening mammograms were collected (each separated by one screening interval). Sensitivity and recall rate for all screening intervals were computed for the AI system and the initial radiologists who interpreted the mammogram. Results have shown that the addition of AI positive assessments to radiologist interpretations of screening mammograms has the potential to detect breast cancers 1 or 2 screening cycles earlier than without AI.
Authors: S. Pacilè, P. Fillard, J. Lopez
Presenter: S. Pacilè, PhD, Clinical Research Manager
View our e-posters online during the event
Discover our e-posters at the poster gallery on https://www.eventscribe.net/2021/SBI-ACR/ during the event:
Exploring the ability of an AI system to differentiate breast microcalcifications
Authors: S. Pacilè, M. Bereby-Kahane C. Balleyguier, A. Tardivon, P. Fillard
This study aims at investigating MammoScreen’s ability to improve both positive and negative predicted value of mammography thus reducing the rate of unnecessary biopsies. A dataset of 2D Full-Field digital mammograms (FFDM) was retrospectively collected. Only women who were biopsied because of the presence of a suspect cluster of calcifications were selected to be included in the study. Predictions of the AI system were computed and compared to human performance in terms of positive and negative predicted value (NPV). The biopsy reduction rate was also computed. Both the PPV and NPV of the AI system were found to be statistically significantly superior with respect to the original decision. The rate of unnecessary biopsies was reduced by 25%. Results show that the use of this AI system for interpretations of screening mammograms holds the potential to improve the differentiation of breast microcalcifications while keeping the highest level of sensitivity.
Advantages of using an artificial intelligence tool as concurrent reader for breast cancer
screening
S. Pacilè, J. Lopez, P. Chone, T. Bertinotti, P. Fillard
Multi-reader multi-case study to demonstrate the benefits brought by MammoScreen in the breast cancer detection process. Results have shown that the average AUC across readers was 0.769 (95% confidence interval [CI]: 0.724, 0.814) without AI and 0.797 (95% CI: 0.754, 0.840) with AI. The average difference in AUC was 0.028 (95% CI: 0.002, 0.055, P = .035). Average sensitivity was increased by 0.033 when using AI support (P = .021). Reading time changed dependently to the AI-tool score. For low likelihood of malignancy (< 2.5%) the time was about the same in the first reading session and slightly decreased in the second reading session. For higher likelihood of malignancy, the reading time was on average increased with the use of AI. We concluded that the concurrent use of this AI tool improves the diagnostic performance of radiologists in the detection of breast cancer without prolonging their workflow.
We hope to see you at our virtual booth and free learning lab in April!