Lung-CAD
K230085Imagen Technologies, Inc · cleared 2023-10-03 · product code MYN · Radiology
Premarket evidence — what FDA accepted
source quote (p.5)
“Lung-CAD is computer-assisted detection (CADe) software designed to increase the accurate detection of lung hyperinflation. The subject device is a software-only device.”
source quote (p.5)
“Lung-CAD uses modern deep learning and computer vision techniques to analyze chest radiographs. The device uses a deep learning algorithm to identify regions of interest (ROIs) with lung hyperinflation and produces boxes around the ROIs. Supervised Deep Learning”
Validation studies (2)
Standalone
n=5,000 cases
endpoints: sensitivity; specificity; Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve; Free-Response ROC (FROC) curve
Retrospective clinical
n=244 cases
endpoints: accuracy of readers aided by Lung-CAD (“Aided”) was superior to the accuracy of readers when unaided by Lung-CAD (“Unaided”) as determined by the case-level Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve; Reader AUC estimates significantly improved (p-value < 0.001); Reader AUC improvement for lung hyperinflation
Reported performance (6 observations)
source quote (p.8)
“The results of the standalone testing demonstrated that Lung-CAD detects ROIs with high sensitivity (0.898; 95% Wilson's Confidence Interval: 0.856, 0.929)”
source quote (p.8)
“high specificity (0.894; 95% Wilson's Confidence Interval: 0.885, 0.902)”
source quote (p.8)
“high Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve (0.964, 95% Bootstrap Confidence Interval: 0.956, 0.972)”
source quote (p.9)
“Positive Predictive Value 0.322 (0.289, 0.357)”
source quote (p.9)
“Negative Predictive Value 0.994 (0.991, 0.996)”
source quote (p.10)
“Reader AUC improvement for lung hyperinflation was 0.0632 (95% Confidence Interval: 0.0632, 0.0633).”
Each value carries its own analysis unit and task — never compare or pool across devices. Source: 510(k) summary PDF.
Predicate network
Postmarket — what happened after clearance
Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device). Signals are descriptive observables with sources — never a judgment that the device is unsafe or drifting. Snapshot 2026-07-08.
Reimbursement — how devices like this got paid
Not yet tracked — no payment pathway indexed for this clearance (the reimbursement corpus is a growing seed set).