AISAP Cardio V1.0

K234141

Aisap · cleared 2024-08-01 · product code POK · Radiology

Premarket evidence — what FDA accepted

Device typesamd
source quote (p.7)
“Software as a medical device (SaMD)”
Algorithmmachine learning NN (neural network) models
source quote (p.5)
“AISAP CARDIO V1.0 uses machine learning NN (neural network) models trained to recognize patterns and make decisions. AISAP CARDIO V1.0 contains classification models which identify categories within data, regression models which predict numerical values, and instance segmentation models that detect and segment objects within images.”
Adaptive (vs locked)FDA source did not state this
PCCPFDA source did not state this
Cybersecurity addressedYes
source quote (p.8)
“Risk assessment, performance, and cybersecurity of AISAP CARDIO V1.0 have been evaluated and verified in accordance with pre-defined software specifications and applicable performance standards through software verification testing.”

Validation studies (4)

Standalone

n=200 cases

endpoints: Left Ventricular Ejection Fraction (LVEF); Inferior Vena Cava (IVC) Maximal Diameter; Left atrial area; Right atrial area; LV end diastolic diameter; Aortic root diameter; Right ventricle (RV) fractional area change (FAC)

standards: American Society of Echocardiography (ASE) guidelines

Standalone

n=329 cases

endpoints: AUC for MR, AS, AR, TR; Sensitivity for MR, AS, AR, TR; Specificity for MR, AS, AR, TR; Kappa for MR, AS, AR, TR

Reader study (MRMC)

n=260 cases

endpoints: AUCaided; AUCunaided; Kappa agreement (aided vs unaided); Accuracy agreement (aided vs unaided)

Bench

n=500 other

endpoints: Accuracy of view classification (PLAX, PSAX, A4C, SC IVC)

Reported performance (24 observations)

aurocas written: “MR AUC”0.975CI 0.960,0.987
source quote (p.10)
“MR 310 0.975 (0.960,0.987)”
sensitivityas written: “MR Sensitivity”95.3CI 90.5,98.9
source quote (p.10)
“95.3% (90.5,98.9)”
specificityas written: “MR Specificity”90.2CI 86.3,93.9
source quote (p.10)
“90.2% (86.3,93.9)”
agreement_kappaas written: “MR Kappa”0.879CI 0.852,0.906
source quote (p.10)
“0.879 (0.852,0.906)”
aurocas written: “AS AUC”0.969CI 0.950,0.984
source quote (p.10)
“AS 272 0.969 (0.950,0.984)”
sensitivityas written: “AS Sensitivity”86.5CI 76.2,95.7
source quote (p.10)
“86.5% (76.2,95.7)”
specificityas written: “AS Specificity”94.5CI 91.3,97.3
source quote (p.10)
“94.5% (91.3,97.3)”
agreement_kappaas written: “AS Kappa”0.865CI 0.825,0.901
source quote (p.10)
“0.865 (0.825,0.901)”
aurocas written: “AR AUC”0.993CI 0.986,0.999
source quote (p.10)
“AR 323 0.993 (0.986,0.999)”
sensitivityas written: “AR Sensitivity”96.5CI 91.3,100.0
source quote (p.10)
“96.5% (91.3,100.0)”
specificityas written: “AR Specificity”97CI 94.9,98.9
source quote (p.10)
“97.0% (94.9,98.9)”
agreement_kappaas written: “AR Kappa”0.913CI 0.892,0.932
source quote (p.10)
“0.913 (0.892,0.932)”
aurocas written: “TR AUC”0.973CI 0.955,0.987
source quote (p.10)
“TR 295 0.973 (0.955,0.987)”
sensitivityas written: “TR Sensitivity”93.5CI 86.9,98.5
source quote (p.10)
“93.5% (86.9,98.5)”
specificityas written: “TR Specificity”89.3CI 85.2,93.1
source quote (p.10)
“89.3% (85.2,93.1)”
agreement_kappaas written: “TR Kappa”0.879CI 0.854,0.905
source quote (p.10)
“0.879 (0.854,0.905)”
agreement_kappaas written: “MR Kappa (aided)”0.881CI 0.872,0.890
source quote (p.11)
“MR Kappa 0.881 (0.872,0.890)”
accuracyas written: “MR Accuracy (aided)”73.6CI 71.9%,75.2%
source quote (p.11)
“Accuracy 73.6% (71.9%,75.2%)”
agreement_kappaas written: “TR Kappa (aided)”0.881CI 0.871,0.892
source quote (p.11)
“TR Kappa 0.881 (0.871,0.892)”
accuracyas written: “TR Accuracy (aided)”75.3CI 73.6%,77.0%
source quote (p.11)
“Accuracy 75.3% (73.6%,77.0%)”
agreement_kappaas written: “AR Kappa (aided)”0.913CI 0.905,0.921
source quote (p.11)
“AR Kappa 0.913 (0.905,0.921)”
accuracyas written: “AR Accuracy (aided)”80.6CI 79.1%,82.1%
source quote (p.11)
“Accuracy 80.6% (79.1%,82.1%)”
agreement_kappaas written: “AS Kappa (aided)”0.85CI 0.834,0.864
source quote (p.11)
“AS Kappa 0.850 (0.834,0.864)”
accuracyas written: “AS Accuracy (aided)”74.7CI 73.0%,76.3%
source quote (p.11)
“Accuracy 74.7% (73.0%,76.3%)”

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
0
drift signals on this device

Recall and MAUDE counts are product-code-level (reports aren't reliably attributable to one device); a recall is shown as device-attributed only when the recall record itself lists this clearance number. 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).

Applicable FDA guidance — what the submission is measured against

FDA guidance documents and guiding principles applicable to 510(k) AI/ML devices in the Radiology panel. A curated reference index, not legal or regulatory advice — each item states its own status, and a draft is never binding.

Applicability is derived from the device's FDA advisory panel and pathway — cross-cutting guidances apply to every AI/ML device; panel-specific ones are flagged. Titles, dates, and links verified against fda.gov as of July 2026.

Constat Precedent · public FDA/CMS data · descriptive decision-support, not regulatory or reimbursement advice. Share this page: constat.dev/precedent/device/K234141