Koios DS is an artificial intelligence software tool that helps trained physicians analyze ultrasound images of breast lesions and thyroid nodules. It processes user-selected regions within the images to provide AI-derived risk assessments for cancer, and generates descriptors following established medical lexicons to improve diagnostic accuracy and reduce variability among physicians. The software also functions as an image viewer and supports image annotation and reporting, helping clinicians make more informed decisions in managing patients with suspicious lesions.
Koios Decision Support (DS) is an artificial intelligence (AI)/machine learning (ML)-based computer-aided diagnosis (CADx) software device intended for use as an adjunct to diagnostic ultrasound examinations of lesions or nodules suspicious for breast or thyroid cancer.
Koios DS is a web-based software application deployed on a Microsoft IIS server, accessed by clinicians through compatible clients. It applies AI and machine learning based diagnostic engines for breast and thyroid ultrasound image analysis, utilizing computer vision to automatically classify lesions and nodules and generate lexicon-based descriptors. Features include Smart Click for ROI selection, Image Registration and Matching for mapping ROIs from screenshots to DICOM images, and OCR to extract measurement data from images.
Extensive performance testing includes bench and clinical studies demonstrating that the software improves physician diagnostic accuracy for breast and thyroid lesions. For the breast engine, testing on 900 cases shows statistically significant improvements in AUC, sensitivity, and specificity. For the thyroid engine, testing on 650 cases shows improved AUC and diagnostic performance including with common clinical lexicons (ACR TI-RADS and ATA). Clinical reader studies showed significant improvements in AUC, sensitivity, and specificity with Koios DS assistance. Additional tests confirm that features like Smart Click and OCR do not degrade system performance.
No predicate devices specified
Submission
7/22/2024
FDA Approval
11/15/2024
Join hundreds of your peers who rely on RadAI Slice. Get the essential weekly briefing that empowers you to navigate the future of radiology.
We respect your privacy. Unsubscribe at any time.