Will this artificial intelligence technology improve diagnostic accuracy and reduce false positive biopsy orders?
Koios DS, an artificial intelligence (AI)-based software platform that reportedly improves the accuracy and efficiency of diagnosing breast and thyroid cancer, recently garnered clearance from the Food and Drug Administration (FDA), according to the software manufacturer Koios Medical.
A recent multicenter retrospective review involving the assessment of 900 breast lesions showed that the addition of Koios DS Breast to ultrasound imaging improved correct BI-RADS classification of sonographic breast lesions by 14 of the 15 physicians participating in the study.
For thyroid cancer, Koios Medical reported a 14 percent increase in detection rates with its Thyroid DS platform and more than a 35 percent reduction in false positive biopsy orders. The company also noted a greater than 50 percent reduction in interpretation variability.
Built with ultrasound data from a worldwide, 48-site network, the AI-generated findings with Koios DS are directly aligned with the BI-RADS and TI-RADS rating systems from the American College of Radiology, according to Koios Medical. The company added that the software is also aligned with the American Thyroid Association’s system for tissue classification and scoring.
“ … This novel software demonstrates that (by) using AI for decision support, physicians can make clinically meaningful shifts in performance, improving interpretation efficacy and diagnostic performance, improving sensitivity and reducing false positives,” noted Lev Barinov, VP of clinical excellence and product management at Koios Medical.
On the reimbursement front, Koios Medical pointed to new CPT Category 3 codes from the American Medical Association that clinicians may employ to bill for use of the Koios DS software in interpreting, classifying and reporting findings from traditional ultrasound examinations.
AI-Initiated Recalls After Screening Mammography Demonstrate Higher PPV for Breast Cancer
March 18th 2025While recalls initiated by one of two reviewing radiologists after screening mammography were nearly 10 percent higher than recalls initiated by an AI software, the AI-initiated recalls had an 85 percent higher positive predictive value for breast cancer, according to a new study.
The Reading Room: Artificial Intelligence: What RSNA 2020 Offered, and What 2021 Could Bring
December 5th 2020Nina Kottler, M.D., chief medical officer of AI at Radiology Partners, discusses, during RSNA 2020, what new developments the annual meeting provided about these technologies, sessions to access, and what to expect in the coming year.
Can Ultrasound-Based Radiomics Enhance Differentiation of HER2 Breast Cancer?
March 11th 2025Multicenter research revealed that a combined model of clinical factors and ultrasound-based radiomics exhibited greater than a 23 percent higher per patient-level accuracy rate for identifying HER2 breast cancer than a clinical model.