LungQ Software

LungQ is a command-line software accessory for CT systems, developed to provide quantitative analysis of pulmonary tissues from thoracic CT scans by employing 3D segmentation, volumetric measurements, density evaluations, and fissure analysis to support diagnosis and follow-up. The software's analytical capabilities were validated through an equivalence study against a predicate device using COPD patient data and through performance testing that demonstrated superior fissure analysis accuracy.

LungQ is a standalone, command-line software designed to assist physicians in the interpretation of Computed Tomography (CT) scans of the thorax, particularly for identifying and documenting pulmonary abnormalities. Its core methodology involves processing DICOM CT thoracic datasets to provide quantitative support for diagnosis and follow-up examinations. The software offers several analytical tools, including three-dimensional segmentation and isolation of pulmonary sub-compartments (lungs and lobes), volumetric analysis of these structures, density evaluations (including low attenuation areas and 15th percentile density), and fissure evaluation. It generates a report of these quantitative CT values but does not feature a graphical user interface and cannot manually edit segmentations. The software underwent non-clinical testing, including software verification and validation, which it successfully passed. An equivalence study was conducted comparing LungQ 1.1.0 to the predicate device, VIDA PW2, using 250 CT scans from the COPDGene study, encompassing subjects with varying stages of COPD and control subjects. This study demonstrated equivalence in lung and lobar volume measurements (difference ≤ 10%) and density scores (LAA-950HU agreement limits -1% to 1%; LAA-910HU agreement limits -10% to 10%; 15th Percentile agreement limits -10 HU to 10 HU). Additionally, a separate performance test on 55 scans compared LungQ 1.1.0's fissure analysis to a previously published version, showing that LungQ 1.1.0 significantly outperformed the older version with an average Area Under the Curve (Az) of 0.95 versus 0.76, using manual reference standards for voxel-wise fissure completeness.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the command-line interface of LungQ affect its integration into diverse clinical imaging workflows and its ease of use for clinicians accustomed to graphical user interfaces?; What is the demonstrated clinical impact of LungQ's improved fissure analysis accuracy (Az 0.95) on patient diagnosis, staging, or treatment planning for specific pulmonary conditions?; Given LungQ does not allow manual editing of segmentations, what are the known limitations or failure modes of its automated segmentation algorithms, particularly in atypical anatomies or severe disease states?; How robust are LungQ's quantitative outputs (volume, density, fissure completeness) across a broader range of CT scanner models, acquisition protocols, and reconstruction kernels beyond those tested in the equivalence and fissure analysis studies?; What is the performance of LungQ in quantifying pulmonary abnormalities other than those typically found in COPD, such as interstitial lung diseases or malignancies, given the validation data was primarily from a COPD cohort?