Intended Use

TAIMedImg DeepMets is a software device intended to assist trained medical professionals by providing initial object contours on axial T1-weighted contrast-enhanced (T1WI+C) brain magnetic resonance (MR) images to accelerate workflow for radiation therapy treatment planning. It is intended only for patients with known brain metastases with lesions ≥10 mm diameter. It offers automated Gross Tumor Volume (GTV) contours of brain metastases and is adjunctive to manual contouring, requiring review by trained professionals.

Technology

The device is a software system comprising an AI inference module utilizing deep learning neural networks for preprocessing, segmentation (contouring) of brain metastases on axial T1-weighted contrast-enhanced MRI, and postprocessing to generate contours. It also includes a DICOM Radiotherapy Structure Set (RTSS) converter module that outputs standard DICOM-RTSTRUCT files compatible with radiotherapy treatment planning systems.

Performance

Performance testing included software verification and validation according to FDA guidance and standards IEC 62304 and ISO 14971. The AI model was trained on a large retrospective dataset and validated on an independent US dataset of 158 patients with 289 lesions. Key metrics such as lesion-wise sensitivity (mean 89.97%), false-positive rate, Dice similarity coefficient, Hausdorff distance, and centroid distance met or exceeded acceptance criteria. Subgroup analyses showed consistent performance across demographic and imaging variables, supporting safety and effectiveness.

Predicate Devices

No predicate devices specified

Device Timeline

  • 1

    Submission

    2/14/2025

    3 months
  • 2

    FDA Approval

    5/28/2025

Ready to Sharpen Your Edge?

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.