InferOperate Suite

InferOperate Suite is medical imaging software that provides trained medical professionals with tools for reading, interpreting, reporting, and treatment planning of patients. It accepts DICOM-compliant medical images and uses machine learning and image processing techniques to segment anatomical structures, perform 3D reconstruction and visualization, and support surgical planning. The software is designed for preoperative planning and intraoperative image display for CT chest, abdominal, and pelvic scans.

InferOperate Suite provides 2D/3D viewing, volume rendering, orthogonal multi-planar reconstructions (MPR), surface rendering, measurements, and surgical planning tools. It includes machine learning-based algorithms for segmentation of anatomical structures and interactive segmentation tools. For non-CT modalities, pediatric patients, or unknown ages, it provides non-ML software functions like STL viewer. The system does not modify the original DICOM data and can be deployed on dedicated on-premise or cloud servers.

Software verification and validation testing per IEC 62304 (Medical device software – Software life cycle processes). Testing included software verification/validation tests, performance tests characterized by Dice coefficient and 95% Hausdorff Distance metrics, and cybersecurity testing per FDA guidance on device software functions and cybersecurity considerations.

InferOperate Suite is substantially equivalent to the Visible Patient Suite predicate device because both are Class II medical imaging software systems with identical intended uses (reading, interpreting, reporting, treatment planning), same DICOM input format, and identical tool functionality (2D/3D viewing, MPR, surface rendering, measurements, surgical planning, segmentation). The addition of machine learning-based segmentation algorithms in InferOperate Suite represents a minor technical difference that does not affect safety or effectiveness, as demonstrated by performance validation data showing Dice coefficients meeting or exceeding predetermined targets across multiple anatomical structures.