iCAD Highlights Promising Clinical Data Demonstrating Ability to Reduce Breast Interval Cancer Rates and Decrease Recall Rates at ECR 2024; Showcases Additional Platform Offerings - Seite 2

Sunday, March 3, 9:30 – 11:00 am, RPS2305 – Retrospective Evaluation of Interval Breast Cancer: Can the Number of Interval Carcinomas be Reduced Utilizing AI Diagnostic Software? In this study conducted by Jonas Subelack et al., researchers from the University of St. Gallen and Krebsliga Ostschweiz in Switzerland and Radiologie am Theater in Germany assessed whether iCAD’s ProFound AI could help reduce interval carcinomas (ICs). Analyzing data from 151,245 screening mammograms between 2010 and 2019, the study identified 264 ICs where cancer was detected within 24 months after screening that was considered normal. Expected results anticipate improved IC detection rates with the software compared to initial screening, with detection accuracy influenced by set thresholds. The study expects the software to outperform radiologists in IC detection, suggesting potential integration into mammography screening to reduce ICs.

Sunday, March 3, 11:30 am – 12:30 pm - RPS 2405 – Adding Artificial Intelligence (AI) Case Malignancy Scoring in a Breast Screening Program to Overcome Delay in Most Probably True Positive Cases: A Retrospective Study. The retrospective study at the Reggio Emilia Breast Screening Program in Italy led by A. Nitrosi et al. aimed to assess the efficacy of implementing iCAD’s ProFound AI case scoring strategy in a breast screening program to expedite the reading process for potentially true positive cases, ensuring compliance with local regulations requiring readings within two weeks. Analyzing data from 32,012 2D mammography screening exams, the study utilized the ProFound AI 2D system. The AI-generated Case Scores, representing the algorithm's confidence in malignancy, were used to prioritize readings. Results showed that reading screening exams above certain Case Score thresholds identified 61% to 89% of screen-detected cancers while requiring only 5.4% to 20% of screening exams to be read. Notably, prioritizing readings of exams with Case Scores above 40% allowed for the recall of the majority (85%) of true positive cases within a short time frame.

ECR 2024 Awarded - Certificate of Merit
E-Poster Presentation – Breast Arterial Calcification (BAC): A Proxy for Medium and Large Vessel Atherosclerotic Calcium. In this retrospective study by C. Parghi et al., researchers investigated the correlation between breast arterial calcification (BAC) as detected by ProFound AI Heart Health on screening mammograms from Solis Mammography and the extent of atherosclerotic disease observed on CT imaging within 12 months of the mammogram. Analyzing data from 1,449 women, they found that a BAC score of 3 or more was associated with a significantly higher rate of clinically significant atherosclerotic disease in large and medium vessels compared to a BAC score of less than 3. Specifically, the rate increased from 11.0% to 31.0% (p < 0.001). The sensitivity, specificity, and accuracy for detecting clinically significant atherosclerotic disease when the BAC score was 3 or more were 21.8%, 92.9%, and 83.8% respectively. These findings suggest that quantitative BAC assessment could offer insight and potentially identify those who need further cardiovascular screening.