Intended Use

icobrain is intended for automatic labeling, visualization and volumetric quantification of segmentable brain structures from a set of MR or NCCT images. It automates the manual process of identifying and quantifying brain structures. It has two pipelines: icobrain cross for single time point images and icobrain long for longitudinal volume changes.

Technology

Software package operates on off-the-shelf hardware, DICOM compatible. Uses classical machine learning and deep learning including supervised voxel classification with Convolutional Neural Networks. Inputs T1-weighted and FLAIR MR images and NCCT images. Outputs volumetric brain structure reports and annotated DICOM images.

Performance

Performance testing validated accuracy and reproducibility on MR and CT images, including TBI and dementia patients. Lesions, basal cisterns, lateral ventricles, and midline shift were compared with manually labeled ground truth or cleared icobrain MR segmentation. Testing included 544 datasets with Pearson correlation coefficient of 0.95 and intraclass correlation coefficient of 0.94, meeting acceptance criteria.

Predicate Devices

No predicate devices specified

Device Timeline

  • 1

    Submission

    7/19/2018

    3 months
  • 2

    FDA Approval

    11/6/2018

Other devices from icometrix NV

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.