Annalise Enterprise CXR Triage Pneumothorax
K213941Annalise-AI · cleared 2022-02-24 · product code QFM · Radiology
Premarket evidence — what FDA accepted
Device typesamd
source quote (p.5)
“Annalise Enterprise CXR Triage Pneumothorax is a software workflow tool that interfaces with RIS and PACS to obtain chest x-ray images to process.”
Algorithmconvolutional neural network trained using deep learning techniques
source quote (p.5)
“The artificial intelligence algorithm within the device uses deep learning techniques to identify the presence of pneumothorax and tension pneumothorax. The AI algorithm used in the device is a convolutional neural network trained using deep learning techniques.”
Adaptive (vs locked)No
PCCPNo
Cybersecurity addressedNo
Validation studies (2)
Retrospective clinical
n=949 cases · 4 site(s)
Bench
n=621 cases
endpoints: average triage turn-around time
Reported performance (15 observations)
sensitivity93.9CI 95% CI: 91.8, 96.1
source quote (p.8)
“The default "balanced sensitivity and specificity” operating point demonstrated sensitivity of 93.9% (95% CI: 91.8, 96.1) and specificity of 92.2% (95% CI: 89.9, 94.4) for pneumothorax”
specificity92.2CI 95% CI: 89.9, 94.4
source quote (p.8)
“The default "balanced sensitivity and specificity” operating point demonstrated sensitivity of 93.9% (95% CI: 91.8, 96.1) and specificity of 92.2% (95% CI: 89.9, 94.4) for pneumothorax”
aurocas written: “auc”0.979CI 95% CI: 0.970-0.986
source quote (p.8)
“The results included an AUC of 0.979 (95% CI: 0.970-0.986) and 0.988 (95% CI: 0.981-0.993) for pneumothorax and tension pneumothorax respectively”
aurocas written: “AUC for tension pneumothorax”0.988CI 95% CI: 0.981-0.993
source quote (p.8)
“The results included an AUC of 0.979 (95% CI: 0.970-0.986) and 0.988 (95% CI: 0.981-0.993) for pneumothorax and tension pneumothorax respectively”
sensitivityas written: “Sensitivity for tension pneumothorax (default operating point)”94.3CI 95% CI: 90.2, 98.4
source quote (p.8)
“and sensitivity of 94.3% (95% CI: 90.2, 98.4) and specificity of 95.8% (95% CI: 94.3, 97.1) for tension pneumothorax.”
specificityas written: “Specificity for tension pneumothorax (default operating point)”95.8CI 95% CI: 94.3, 97.1
source quote (p.8)
“and sensitivity of 94.3% (95% CI: 90.2, 98.4) and specificity of 95.8% (95% CI: 94.3, 97.1) for tension pneumothorax.”
sensitivityas written: “Sensitivity for pneumothorax (optimized for sensitivity operating point)”96.6CI 95% CI: 94.7, 98.3
source quote (p.8)
“The “optimized for sensitivity” operating point demonstrated sensitivity of 96.6% (95% CI: 94.7, 98.3) and specificity of 84.1% (95% CI: 82.1, 87.1) for pneumothorax”
specificityas written: “Specificity for pneumothorax (optimized for sensitivity operating point)”84.1CI 95% CI: 82.1, 87.1
source quote (p.8)
“The “optimized for sensitivity” operating point demonstrated sensitivity of 96.6% (95% CI: 94.7, 98.3) and specificity of 84.1% (95% CI: 82.1, 87.1) for pneumothorax”
sensitivityas written: “Sensitivity for tension pneumothorax (optimized for sensitivity operating point)”95.9CI 95% CI: 91.9, 99.2
source quote (p.8)
“and sensitivity of 95.9% (95% CI: 91.9, 99.2) and specificity of 94.9% (95% CI: 93.3, 96.4) for tension pneumothorax.”
specificityas written: “Specificity for tension pneumothorax (optimized for sensitivity operating point)”94.9CI 95% CI: 93.3, 96.4
source quote (p.8)
“and sensitivity of 95.9% (95% CI: 91.9, 99.2) and specificity of 94.9% (95% CI: 93.3, 96.4) for tension pneumothorax.”
sensitivityas written: “Sensitivity for pneumothorax (optimized for specificity operating point)”89.1CI 95% CI: 86.2, 92.0
source quote (p.8)
“The “optimized for specificity" operating point demonstrated sensitivity of 89.1% (95% CI: 86.2, 92.0) and specificity of 95.7% (95% CI: 94.0, 97.4) for pneumothorax”
specificityas written: “Specificity for pneumothorax (optimized for specificity operating point)”95.7CI 95% CI: 94.0, 97.4
source quote (p.8)
“The “optimized for specificity" operating point demonstrated sensitivity of 89.1% (95% CI: 86.2, 92.0) and specificity of 95.7% (95% CI: 94.0, 97.4) for pneumothorax”
sensitivityas written: “Sensitivity for tension pneumothorax (optimized for specificity operating point)”83.7CI 95% CI: 76.4, 90.2
source quote (p.8)
“and sensitivity of 83.7% (95% CI: 76.4, 90.2) and specificity of 97.8% (95% CI: 96.7, 98.7) for tension pneumothorax.”
specificityas written: “Specificity for tension pneumothorax (optimized for specificity operating point)”97.8CI 95% CI: 96.7, 98.7
source quote (p.8)
“and sensitivity of 83.7% (95% CI: 76.4, 90.2) and specificity of 97.8% (95% CI: 96.7, 98.7) for tension pneumothorax.”
time_to_resultas written: “average triage turn-around time”20.57CI 95% CI: 19.90-21.24
source quote (p.8)
“The results demonstrated an average triage turn-around time of 20.57 seconds, (95% CI: 19.90-21.24)”
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
0
recalls in product code, 24mo
0
MAUDE reports in code, 12mo
—
vs code's own 3-yr baseline
1
drift signals on this device
- re_clearance
The FDA AI/ML device list shows a newer 510(k) K250831 (decision 2025-04-23) from Annalise-AI for a matching device line ("Annalise Enterprise") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K250831
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).