Radiomics App v1.0

Microsoft Radiomics App v1.0 is a software-only medical device that employs computerized medical image analysis and machine learning algorithms to assist radiation oncology professionals in generating and refining 3D contours of tumors and healthy organs from CT and MR scans. The system aims to improve the efficiency and accuracy of preparing data for radiation treatment planning through tools for assisted and automated segmentation, image fusion, and 3D visualization.

Microsoft Radiomics App v1.0 is a software-only medical device designed for use by trained radiation oncologists, dosimetrists, and physicists. Its primary purpose is to assist these professionals in deriving optimal organ and tumor contours from Computed Tomography (CT) and Magnetic Resonance (MR) images, which serve as critical inputs for radiation treatment planning. The application supports loading, saving, and displaying medical images and contours; creating, transforming, and modifying contours for radiotherapy planning, adaptive therapy, and patient follow-up archiving; localizing and defining solid tumors and healthy anatomical structures; fusion display of compatible images; and three-dimensional rendering of medical images and segmented contours. It is specified that images reviewed using the Radiomics App should not be used for primary image interpretations, and the software is not intended for use with digital mammography. The Radiomics App v1.0 is founded on research in computerized medical image analysis, computer vision, and machine learning, employing algorithms for the assisted delineation of anatomical structures in 3D radiological scans. It provides several segmentation tools: an 'Assisted Contouring' module for manual, user-guided segmentation; a 'Machine-learning based contouring' module that uses ML algorithms to provide an initial automatic segmentation which users can accept or refine; and a 'Contour refinement' module for editing segmentations. These segmentations are exported for use in radiotherapy treatment planning systems. Software verification and validation testing, including assessments of measurements, volumetric rendering, multi-planar reconstruction, assisted contouring, and automatic contouring, were conducted to ensure the software performs according to its specifications and meets user needs and intended uses.

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 performance of the machine-learning based contouring module vary with different anatomical regions or tumor morphologies in CT versus MR imaging?; What specific machine learning algorithms are implemented, and what was the diversity and size of the datasets used for their training and validation?; Are there published data or internal validation metrics detailing the typical time savings or contour accuracy improvements when using the Radiomics App compared to fully manual contouring?; How does the software handle image artifacts or low-quality scans, and what is the impact on the reliability of both assisted and automated segmentation tools?; What are the specific output formats for the segmented contours, and how seamless is the integration with a range of commercially available radiotherapy treatment planning systems?