Koios DS

Koios DS is an artificial intelligence and machine learning-based computer-aided diagnosis software designed to assist physicians in interpreting breast and thyroid ultrasound images. It employs automated image analysis to generate cancer risk assessments and lexicon-based descriptors, aiming to enhance diagnostic accuracy and streamline clinical workflow.

Koios Medical, Inc. has developed Koios DS, an artificial intelligence and machine learning-based computer-aided diagnosis (CADx) software. This software is designed to assist trained interpreting physicians during diagnostic ultrasound examinations of lesions or nodules suspicious for breast or thyroid cancer. Following a review, the Koios DS device was determined to be substantially equivalent to legally marketed predicate devices for its specified indications for use, allowing it to be marketed. Koios DS enables users to select regions of interest within ultrasound images for automated analysis, which generates a cancer risk assessment and applicable lexicon-based descriptors to enhance diagnostic accuracy and reduce variability. The software also functions as a multi-modality image viewer with tools for image adjustment, measurement, and structured reporting. Key enhancements include improved performance for the breast classification engine and maintained strong performance for the thyroid engine with an AI Adapter. New features like Thyroid Smart Click, Image Registration and Matching, and Optical Character Recognition further streamline the workflow.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the AI/ML model for Koios DS handle false positives and negatives in real-world clinical settings, especially considering its CADx nature?; What specific metrics were used to assess the "improved performance" of the breast classification engine and the "strong performance" of the thyroid engine with the AI Adapter?; Can Koios DS integrate with existing PACS systems and electronic health records to provide a more seamless diagnostic workflow?; What kind of training or certification is required for physicians to effectively utilize Koios DS, particularly concerning the interpretation of its cancer risk assessments and lexicon descriptors?; How does the "Image Registration and Matching" feature contribute to diagnostic confidence, particularly in longitudinal studies of lesions?