Rapid NCCT Stroke
K222884iSchemaView, Inc. · cleared 2023-03-02 · product code QAS · Radiology
Premarket evidence — what FDA accepted
Device typesamd
source quote (p.5)
“Rapid NCCT Stroke (RNS) is a radiological computer-assisted triage and notification software device. RNS is a non-enhanced CT (NCCT) processing module which operates within the integrated Rapid Platform to provide triage and notification prioritization of suspected intracranial Hemorrhage (ICH) and NCCT Large Vessel Occlusion (LVO) of the ICA and MCA-M1. The RNS is an AI/ML SaMD.”
AlgorithmAI/ML SaMD. ICH analysis uses the ICH Algorithm; NCCT LVO suspicion uses combined analysis of ASPECTS and Hyperdense Vessel Sign (HVS) algorithms. The technology is described as an AI/ML/Neural Network.
source quote (p.5)
“The RNS is an AI/ML SaMD. The output of the module is a priority notification to clinicians indicating the suspicion of ICH or NCCT LVO. ICH analysis uses the ICH Algorithm to identify findings within the ICH algorithm; and the NCCT LVO suspicion uses the combined analysis of the ASPECTS and Hyperdense Vessel Sign (HVS) algorithms. Technology AI/ML/Neural Network”
Adaptive (vs locked)FDA source did not state this
PCCPFDA source did not state this
Cybersecurity addressedFDA source did not state this
Validation studies (1)
Retrospective clinical
n=254 cases
endpoints: software's performance in identifying NCCT head images containing intracranial hemorrhage (ICH); software's performance in identifying NCCT head images containing large vessel occlusion (LVO) findings
standards: EN ISO 14971:2012, IEC 62304:2015, NEMA PS 3.1 - 3.20
Reported performance (26 observations)
sensitivity0.635CI 0.544 - 0.717
source quote (p.9)
“LVO was observed at Se:0.635 (95% CI: 0.544 - 0.717)”
specificity0.951CI 0.891 – 0.979
source quote (p.9)
“LVO was observed at ... Sp: 0.951 (95% CI: 0.891 – 0.979).”
sensitivityas written: “ICH Sensitivity”0.962
source quote (p.9)
“ICH performance was observed at Se:0.962”
specificityas written: “ICH Specificity”0.974
source quote (p.9)
“Sp:0.974”
sensitivityas written: “Gender - Female Sensitivity”0.579CI 0.450 - 0.698
source quote (p.10)
“Sensitivity 57 0.579 0.450 0.698”
specificityas written: “Gender - Female Specificity”0.981CI 0.902 - 0.997
source quote (p.10)
“Specificity 54 0.981 0.902 0.997”
sensitivityas written: “Gender - Male Sensitivity”0.69CI 0.562 - 0.794
source quote (p.10)
“Sensitivity 58 0.690 0.562 0.794”
specificityas written: “Gender - Male Specificity”0.933CI 0.821 - 0.977
source quote (p.10)
“Specificity 45 0.933 0.821 0.977”
sensitivityas written: “Age < 50 Sensitivity”0.583CI 0.320 - 0.807
source quote (p.10)
“Sensitivity 12 0.583 0.320 0.807”
specificityas written: “Age < 50 Specificity”0.905CI 0.711 - 0.973
source quote (p.10)
“Specificity 21 0.905 0.711 0.973”
sensitivityas written: “Age 50-70 Sensitivity”0.61CI 0.457 - 0.743
source quote (p.10)
“Sensitivity 41 0.610 0.457 0.743”
specificityas written: “Age 50-70 Specificity”0.976CI 0.877 - 0.996
source quote (p.10)
“Specificity 42 0.976 0.877 0.996”
sensitivityas written: “Age >70 Sensitivity”0.661CI 0.537 - 0.767
source quote (p.10)
“Sensitivity 62 0.661 0.537 0.767”
specificityas written: “Age >70 Specificity”0.971CI 0.855 - 0.995
source quote (p.10)
“Specificity 35 0.971 0.855 0.995”
sensitivityas written: “Slice Thickness < 4.0 Sensitivity”0.724CI 0.543 - 0.853
source quote (p.10)
“Sensitivity 29 0.724 0.543 0.853”
specificityas written: “Slice Thickness < 4.0 Specificity”0.964CI 0.877 - 0.990
source quote (p.10)
“Specificity 55 0.964 0.877 0.990”
sensitivityas written: “Slice Thickness ">= 4.0 Sensitivity”0.605CI 0.499 - 0.701
source quote (p.11)
“Sensitivity 86 0.605 0.499 0.701”
specificityas written: “Slice Thickness ">= 4.0 Specificity”0.938CI 0.832 - 0.979
source quote (p.11)
“Specificity 48 0.938 0.832 0.979”
sensitivityas written: “Manufacturer GE Sensitivity”0.73CI 0.570 - 0.846
source quote (p.11)
“Sensitivity 37 0.730 0.570 0.846”
specificityas written: “Manufacturer GE Specificity”1CI 0.741 - 1.000
source quote (p.11)
“Specificity 11 1.000 0.741 1.000”
sensitivityas written: “Manufacturer PHILIPS Sensitivity”0.545CI 0.280 - 0.787
source quote (p.11)
“Sensitivity 11 0.545 0.280 0.787”
specificityas written: “Manufacturer PHILIPS Specificity”1CI 0.851 - 1.000
source quote (p.11)
“Specificity 22 1.000 0.851 1.000”
sensitivityas written: “Manufacturer SIEMENS Sensitivity”0.5CI 0.336 - 0.664
source quote (p.11)
“Sensitivity 32 0.500 0.336 0.664”
specificityas written: “Manufacturer SIEMENS Specificity”0.915CI 0.816 - 0.963
source quote (p.11)
“Specificity 59 0.915 0.816 0.963”
sensitivityas written: “Manufacturer TOSHIBA Sensitivity”0.686CI 0.520 - 0.814
source quote (p.11)
“Sensitivity 35 0.686 0.520 0.814”
specificityas written: “Manufacturer TOSHIBA Specificity”1CI 0.741 - 1.000
source quote (p.11)
“Specificity 11 1.000 0.741 1.000”
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) K233512 (decision 2024-01-16) from iSchemaView, Inc. for a matching device line ("Rapid (6.0)") — a new clearance for the same line is a change event.
first seen 2026-07-08 · k_number:K233512
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).