DrAid™ for Liver Segmentation

DrAid™ for Liver Segmentation is a web-based software application for visualization, evaluation, and reporting of liver and physician-identified lesions using multiphase CT images. It is designed for use by trained healthcare professionals (physicians and technicians) in hospital settings to assess liver volume, lesion volume, and maximum lesion diameter for evaluation and treatment planning. The software does not generate diagnoses; clinical interpretation remains the responsibility of qualified healthcare professionals.

The subject device provides semi-automated liver segmentation using an AI algorithm with editable tools, whereas the predicate device (IQQA-LIVER) provides manual segmentation. Both are software applications compatible with DICOM image data and designed for multiphase liver image analysis in healthcare environments. The subject device operates as a web-based platform on standard hospital workstations, while differences in deployment platforms, operating systems, and some features exist between the two devices.

Not stated in this summary.

Both devices share the same intended use (liver and lesion segmentation for trained professionals), are compatible with DICOM multiphase imaging, provide essential image viewing and reporting tools, and are intended for hospital healthcare environments. Although the subject device uses AI-assisted segmentation while the predicate uses manual segmentation, the applicant argues these differing techniques do not raise different safety or effectiveness questions because advanced algorithms enhance efficiency and accuracy without compromising clinical performance. The device demonstrated robust performance (Dice score 0.9649, volume error 2.73%) across diverse patient pathologies and multiple CT scanner manufacturers, supporting equivalence in performance outcomes regardless of the segmentation method used.