ClariPulmo is a software tool designed to analyze lung CT images. It helps doctors by automatically segmenting the lungs and quantifying areas of low and high attenuation in the lung tissue. It includes AI-based features like deep learning segmentation, denoising, and kernel normalization to improve image quality and quantification, especially for low-dose CT scans. This software supports clinical assessment of lung abnormalities by providing color-coded images and quantitative reports.
ClariPulmo is a non-invasive image analysis software for use with CT images which is intended to support the quantification of lung CT images, supporting physicians in diagnosis and documentation of pulmonary tissue images from CT thoracic datasets.
ClariPulmo uses a pre-trained deep learning model for automatic lung segmentation. It includes two main analyses: LAA (low attenuation areas) and HAA (high attenuation areas) quantified using user-predefined Hounsfield unit thresholds. Optional denoising and kernel normalization functions based on U-Net architecture improve quantification for low-dose or sharp kernel reconstructions. It produces color overlays, tables, and charts for quantification results.
Performance testing demonstrated excellent agreement of HAA and LAA analyses with expert segmentations (PCC > 0.98). AI-based lung segmentation showed high agreement with expert manual segmentations (PCC 0.977–0.992 and DICE coefficients 0.98–0.99) across multiple datasets including normal and diseased patients, different CT scanners, kernels, and dose levels. The device meets multiple FDA-recognized standards and guidance for safety and effectiveness and does not require clinical studies to demonstrate substantial equivalence.
No predicate devices specified
Submission
12/28/2020
FDA Approval
4/6/2022
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.