eCARTv5 Clinical Deterioration Suite (eCART)

K233253

AgileMD, Inc. · cleared 2024-06-21 · product code QNL · Cardiovascular

Premarket evidence — what FDA accepted

Device typesamd

source quote (p.5)

“The AgileMD eCARTv5 Clinical Deterioration Suite (“eCART") is a cloud-based software device that is integrated into the electronic health record (“EHR”) in order to anticipate clinical deterioration in adult ward patients, which is signified as either of the following two predicted outcomes: (1) death or (2) ICU transfer.”

AlgorithmGradient Boosted Machine (“GBM”) learning algorithm

source quote (p.5)

“The primary functions of the system are imparted by the Gradient Boosted Machine (“GBM”) learning algorithm that takes input directly from the EHR, in real time, to provide an assessment of patients and displays its outputs in an intuitive user interface which drives providers to follow standardized clinical workflows (established by their institutions) for elevated-risk patients.”

Adaptive (vs locked)No

PCCPNo

Cybersecurity addressedNo

Validation studies (2)

Retrospective clinical

n=934,454 patients · 3 site(s)

endpoints: Deterioration; Mortality

Prospective clinical

n=151,233 patients · 3 site(s)

endpoints: Deterioration; Mortality

Reported performance (34 observations)

sensitivity0.488CI 48.7%, 49.0%

source quote (p.7)

“48.8% (48.7%, 49.0%)”

specificity0.933CI 93.3%, 93.3%

source quote (p.7)

“93.3% (93.3%, 93.3%)”

aurocas written: “auc”0.828CI 0.827, 0.829

source quote (p.7)

“0.828 (0.827, 0.829)”

aurocas written: “AUROC (Retrospective, Deterioration)”0.835CI 0.834, 0.835

source quote (p.7)

“0.835 (0.834, 0.835)”

aurocas written: “AUROC (Retrospective, Mortality)”0.923CI 0.923, 0.924

source quote (p.7)

“0.923 (0.923, 0.924)”

aurocas written: “AUROC (Prospective, Mortality)”0.913CI 0.911, 0.914

source quote (p.7)

“0.913 (0.911, 0.914)”

ppvas written: “PPV (Retrospective, Deterioration, eCART ≥93)”0.09CI 9.0%, 9.1%

source quote (p.7)

“9.0% (9.0%, 9.1%)”

npvas written: “NPV (Retrospective, Deterioration, eCART ≥93)”0.993CI 99.3%, 99.3%

source quote (p.7)

“99.3% (99.3%, 99.3%)”

ppvas written: “PPV (Prospective, Deterioration, eCART ≥93)”0.089CI 8.8%, 8.9%

source quote (p.7)

“8.9% (8.8%, 8.9%)”

npvas written: “NPV (Prospective, Deterioration, eCART ≥93)”0.993CI 99.3%, 99.3%

source quote (p.7)

“99.3% (99.3%, 99.3%)”

sensitivityas written: “Sensitivity (Retrospective, Deterioration, eCART ≥97)”0.386CI 38.5%, 38.7%

source quote (p.7)

“38.6% (38.5%, 38.7%)”

specificityas written: “Specificity (Retrospective, Deterioration, eCART ≥97)”0.969CI 96.9%, 96.9%

source quote (p.7)

“96.9% (96.9%, 96.9%)”

ppvas written: “PPV (Retrospective, Deterioration, eCART ≥97)”0.142CI 14.1%, 14.2%

source quote (p.7)

“14.2% (14.1%, 14.2%)”

npvas written: “NPV (Retrospective, Deterioration, eCART ≥97)”0.992CI 99.2%, 99.2%

source quote (p.7)

“99.2% (99.2%, 99.2%)”

sensitivityas written: “Sensitivity (Prospective, Deterioration, eCART ≥97)”0.337CI 33.6%, 33.9%

source quote (p.7)

“33.7% (33.6%, 33.9%)”

specificityas written: “Specificity (Prospective, Deterioration, eCART ≥97)”0.973CI 97.3%, 97.3%

source quote (p.7)

“97.3% (97.3%, 97.3%)”

ppvas written: “PPV (Prospective, Deterioration, eCART ≥97)”0.142CI 14.1%, 14.3%

source quote (p.7)

“14.2% (14.1%, 14.3%)”

npvas written: “NPV (Prospective, Deterioration, eCART ≥97)”0.991CI 99.1%, 99.1%

source quote (p.7)

“99.1% (99.1%, 99.1%)”

sensitivityas written: “Sensitivity (Retrospective, Mortality, eCART ≥93)”0.758CI 75.6%, 75.9%

source quote (p.7)

“75.8% (75.6%, 75.9%)”

specificityas written: “Specificity (Retrospective, Mortality, eCART ≥93)”0.926CI 92.6%, 92.6%

source quote (p.7)

“92.6% (92.6%, 92.6%)”

ppvas written: “PPV (Retrospective, Mortality, eCART ≥93)”0.019CI 1.9%, 1.9%

source quote (p.7)

“1.9% (1.9%, 1.9%)”

npvas written: “NPV (Retrospective, Mortality, eCART ≥93)”1CI 100.0%, 100.0%

source quote (p.7)

“100.0% (100.0%, 100.0%)”

sensitivityas written: “Sensitivity (Prospective, Mortality, eCART ≥93)”0.717CI 71.3%, 72.1%

source quote (p.7)

“71.7% (71.3%, 72.1%)”

specificityas written: “Specificity (Prospective, Mortality, eCART ≥93)”0.929CI 92.8%, 92.9%

source quote (p.7)

“92.9% (92.8%, 92.9%)”

ppvas written: “PPV (Prospective, Mortality, eCART ≥93)”0.018CI 1.8%, 1.8%

source quote (p.7)

“1.8% (1.8%, 1.8%)”

npvas written: “NPV (Prospective, Mortality, eCART ≥93)”0.999CI 99.9%, 99.9%

source quote (p.7)

“99.9% (99.9%, 99.9%)”

sensitivityas written: “Sensitivity (Retrospective, Mortality, eCART ≥97)”0.647CI 64.5%, 64.9%

source quote (p.7)

“64.7% (64.5%, 64.9%)”

specificityas written: “Specificity (Retrospective, Mortality, eCART ≥97)”0.965CI 96.5%, 96.5%

source quote (p.7)

“96.5% (96.5%, 96.5%)”

ppvas written: “PPV (Retrospective, Mortality, eCART ≥97)”0.034CI 3.4%, 3.4%

source quote (p.7)

“3.4% (3.4%, 3.4%)”

npvas written: “NPV (Retrospective, Mortality, eCART ≥97)”0.992CI 99.2%, 99.2%

source quote (p.7)

“99.2% (99.2%, 99.2%)”

sensitivityas written: “Sensitivity (Prospective, Mortality, eCART ≥97)”0.581CI 57.6%, 58.6%

source quote (p.7)

“58.1% (57.6%, 58.6%)”

specificityas written: “Specificity (Prospective, Mortality, eCART ≥97)”0.97CI 97.0%, 97.0%

source quote (p.7)

“97.0% (97.0%, 97.0%)”

ppvas written: “PPV (Prospective, Mortality, eCART ≥97)”0.034CI 3.4%, 3.5%

source quote (p.7)

“3.4% (3.4%, 3.5%)”

npvas written: “NPV (Prospective, Mortality, eCART ≥97)”0.999CI 99.9%, 99.9%

source quote (p.7)

“99.9% (99.9%, 99.9%)”

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

recalls in product code, 24mo

MAUDE reports in code, 12mo

—

vs code's own 3-yr baseline

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 Cardiovascular panel. A curated reference index, not legal or regulatory advice — each item states its own status, and a draft is never binding.

Final guidance2026-01
Clinical Decision Support Software
Clinical decision support · SaMD (general)
New final guidance issued Jan 2026, superseding the Sept 2022 version; narrows the device-CDS scope. Applies to software that informs clinical management.
Final guidance2026-01
General Wellness: Policy for Low Risk Devices
SaMD (general) · Clinical decision support
Revised final (Jan 2026); now addresses noninvasive products estimating physiologic parameters (SpO2, BP, glucose). Reshapes the device / non-device line for AI wellness features.
Final guidance2025-09
Computer Software Assurance for Production and Quality Management System Software
SaMD (general) · Postmarket
Final (Sept 2025). Covers software used in production/QMS (incl. ML development-pipeline tooling), superseding Section 6 of the 2002 GPSV — not device software functions themselves.
Final guidance2025-06
Cybersecurity in Medical Devices: Quality Management System Considerations and Content of Premarket Submissions
Cybersecurity · Software premarket content
Reissued June 2025 (retitled 'Quality Management System', was Sept 2023 'Quality System'); adds coverage of FD&C Act §524B cyber devices.
Final guidance2024-12
Marketing Submission Recommendations for a Predetermined Change Control Plan for Artificial Intelligence-Enabled Device Software Functions
Predetermined Change Control Plan · AI/ML lifecycle · Software premarket content
Final (Dec 2024). Supersedes the April 2023 AI/ML PCCP draft.
Final guidance2023-10
Electronic Submission Template for Medical Device 510(k) Submissions
Software premarket content
eSTAR has been mandatory for 510(k)s since Oct 2023 — operationally unavoidable, though not AI-specific.
Final guidance2023-08
Off-The-Shelf Software Use in Medical Devices
Software premarket content · SaMD (general)
Final (Aug 2023). Applies when a device incorporates off-the-shelf software components (common in ML stacks).
Final guidance2023-06
Content of Premarket Submissions for Device Software Functions
Software premarket content · SaMD (general)
Final (June 2023); replaced the May 2005 'Software Contained in Medical Devices' guidance. Documentation level drives the software content of the submission.
Final guidance2022-09
Policy for Device Software Functions and Mobile Medical Applications
SaMD (general) · Clinical decision support
Current version Sept 2022. Frames which software functions FDA regulates as devices.
Final guidance2021-10
De Novo Classification Process (Evaluation of Automatic Class III Designation)
De Novo pathway
Final (Oct 2021), issued with the De Novo final rule. Most relevant to first-of-a-kind devices without a predicate (DEN-numbered clearances).
Final guidance2016-12
Postmarket Management of Cybersecurity in Medical Devices
Cybersecurity · Postmarket
Final guidance2002-01
General Principles of Software Validation
SaMD (general) · Software premarket content
Still active except Section 6 (superseded Sept 2025 by the Computer Software Assurance final guidance).
Draft guidance2025-01
Artificial Intelligence-Enabled Device Software Functions: Lifecycle Management and Marketing Submission Recommendations
AI/ML lifecycle · Software premarket content · Transparency
Draft as of July 2026 (published Jan 2025); finalization is on CDRH's FY2026 agenda but not yet published. Treat as FDA's stated direction, not a binding expectation.
Draft guidance2024-08
Predetermined Change Control Plans for Medical Devices
Predetermined Change Control Plan · Postmarket
Draft (Aug 2024) extending PCCPs beyond AI to all devices under FD&C §515C; not final as of July 2026.
Guiding principles2024-06
Transparency for Machine Learning-Enabled Medical Devices: Guiding Principles
Transparency · AI/ML lifecycle
Guiding principles2023-10
Predetermined Change Control Plans for Machine Learning-Enabled Medical Devices: Guiding Principles
Predetermined Change Control Plan · AI/ML lifecycle
FDA/Health Canada/MHRA joint principles (Oct 2023); companion to the GMLP and Transparency principles.
Guiding principles2021-10
Good Machine Learning Practice for Medical Device Development: Guiding Principles
AI/ML lifecycle · SaMD (general)
FDA/Health Canada/MHRA joint principles (Oct 2021). Foundational, not a binding guidance; IMDRF issued a related GMLP document Jan 2025.

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.