(266 days)
The CARESCAPE B650 is a multi-parameter patient monitor intended for use in multiple areas and intra hospital transport within a professional healthcare facility.
The CARESCAPE B650 is intended for use on adult, pediatric, and neonatal patients and on one patient at a time. The CARESCAPE B650 is indicated for monitoring of:
• hemodynamic (including ECG, ST segment, arrhythmia detection, ECG diagnostic analysis and measurement, invasive pressure, non-invasive blood pressure, pulse oximetry, cardiac output (thermodilution and pulse contour), temperature, mixed venous oxygen saturation, and central venous oxygen saturation),
• respiratory (impedance respiration, airway gases (CO2, O2, N2O, and anesthetic agents), spirometry, gas exchange), and
• neurophysiological status (including electroencephalography, Entropy, Bispectral Index (BIS), and neuromuscular transmission).
The CARESCAPE B650 can be a stand-alone monitor or interfaced to other devices. It can also be connected to other monitors for remote viewing and to data management software devices via a network.
The CARESCAPE B650 is able to detect and generate alarms for ECG arrhythmias: atrial fibrillation, accelerated ventricular rhythm, asystole, bigeminy, bradycardia, ventricular couplet, irregular, missing beat, multifocal premature ventricular contractions (PVCs), pause, R on T, supra ventricular tachycardia, tachycardia, trigeminy, ventricular bradycardia, ventricular fibrillation/ventricular tachycardia, ventricular tachycardia, and VT>2. The CARESCAPE B650 also shows alarms from other ECG sources.
The CARESCAPE B650 also provides other alarms, trends, snapshots and events, and calculations, and can be connected to displays, printers and recording devices.
The CARESCAPE B650 is intended for use under the direct supervision of a licensed healthcare practitioner, or by personnel trained in proper use of the equipment in a professional healthcare facility.
CARESCAPE B650 V3 is a new version of a portable multi-parameter patient monitoring system. The CARESCAPE B650 V3 includes the monitor with built-in CPU, power unit, a 15 inch touch display, the CARESCAPE Software version 3 and the battery. CARESCAPE B650 V3 is equipped with so called ePort interface that supports use of PDM or CARESCAPE ONE patient data acquisition modules for patient monitoring. CARESCAPE B650 V3 is also equipped with two module slots where patient data acquisition modules (E-Modules), can be connected to perform patient monitoring. The CARESCAPE B650 V3 includes features and subsystems that are optional or configurable.
Here's an analysis of the provided text regarding the acceptance criteria and study proving device meets those criteria, specifically concerning the CARESCAPE B650 medical monitor.
It's important to note that this 510(k) summary focuses on demonstrating substantial equivalence to a predicate device, rather than proving the performance against specific pre-defined acceptance criteria for a novel AI algorithm. The device here is a multi-parameter patient monitor with an updated arrhythmia detection algorithm (EK-Pro V14) as one of its changes. The document states that no clinical studies were required to support substantial equivalence, meaning the proof of meeting acceptance criteria for its functions comes from bench testing and comparison to the predicate, rather than a prospective clinical trial.
Therefore, the answers will reflect this context, particularly the lack of a traditional "AI algorithm" study with specific statistical performance metrics against a defined ground truth derived from expert consensus or pathology, as one might expect for a new AI diagnostic tool.
1. A table of acceptance criteria and the reported device performance
The document does not present a formal table of acceptance criteria with corresponding performance metrics in the way one might expect for a novel AI component. Instead, it demonstrates equivalence through:
- Identical specifications: Many features (patient type, use environments, size & weight, module housing, display/screen, waveforms/parameter windows, available parameters, printing, mounting options) are stated as "Identical" to the predicate. This implies that the performance for these aspects meets the original predicate's established performance, which implicitly acts as the "acceptance criteria."
- Equivalent specifications with justification: For changed components like the CPU, modules, EK-Pro arrhythmia algorithm, graphical user interface, alarms, remote alarm device, and networking capability, the document justifies the "equivalence" or improvement. The acceptance criteria here are implicitly that the new component performs at least as well as, or better than, the predicate, without introducing new safety or effectiveness concerns.
| Feature (Related to Acceptance Criteria) | Reported Device Performance / Status (CARESCAPE B650 V3) | Nature of "Acceptance" / Justification |
|---|---|---|
| Arrhythmia Detection Algorithm | Uses EK-Pro V14 (Predicate used EK-Pro V13) | Equivalent: This is a key change. Although not explicitly detailing sensitivity/specificity percentages, the clearance implies that the performance of V14 is considered equivalent or improved, and has passed internal verification and validation testing to ensure it meets the necessary standards for arrhythmia detection. |
| Overall Performance | Demonstrated design meets specifications. | Bench testing: Software, hardware, and performance testing, including applicable consensus standards. This is the primary method of "proving" it meets acceptance criteria without clinical data. |
| New Features (e.g., Alarms, Modules, Networking) | New functionalities and hardware, such as enhanced alarm flexibility, new acquisition modules (E-EEGX, CARESCAPE ONE, E-sCAIOVX, E-sCOVX), and faster wireless data transfer (IEEE 802.11n, 802.11r). | Equivalent/Improved: These changes are described as providing more flexibility, covering the same parameters, simplifying installation, enhancing security, and improving functionality, all while maintaining the fundamental function and safety of the device. Default alarm settings comply with IEC 60601-1-8. |
| Safety, Security, Privacy Risks | Addressed in design and development (Security Risk Assessment, Threat model, integral controls). Withstands network storm. | Compliance: Meets FDA Guidance for Content of Premarket Submissions for Management of Cybersecurity in Medical Devices. Demonstrated resilience in specific network conditions. |
2. Sample sized used for the test set and the data provenance
The document does not specify a "test set" in the context of a dataset for validating an AI algorithm's performance on clinical data. Since no clinical studies were required or conducted for this 510(k) submission, there is no mention of sample sizes of patient data for a test set.
The testing involved:
- Bench testing: "software, hardware and performance including applicable consensus standards was conducted on the CARESCAPE B650, demonstrating the design meets the specifications."
- Usability testing: Formative and summative usability testing.
- Network testing: "Testing was completed on CARESCAPE B650 to show the device can withstand network storm..."
The data provenance for these internal engineering and usability tests would be the manufacturer's own internal testing environment. No information on country of origin or retrospective/prospective nature of data is provided because it's not a clinical study on patient data.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
Not applicable. As no clinical "test set" from patient data was used against which the device's diagnostic performance (e.g., AI algorithm accuracy) was measured by external experts, there is no mention of experts establishing ground truth for such a set. The "ground truth" for demonstrating substantial equivalence primarily lies in adherence to performance specifications validated through engineering bench testing and comparison to the predicate device's established performance.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
Not applicable, for the same reasons as #3. No external "test set" requiring expert adjudication was utilized to demonstrate the device's performance characteristics.
5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
No. The document explicitly states: "The subject of this premarket submission, CARESCAPE B650 did not require clinical studies to support substantial equivalence." Therefore, no MRMC study was performed, and no effect size on human reader improvement with AI assistance is provided. The device integrates an EK-Pro arrhythmia detection algorithm, but the submission focuses on its equivalence to the previous version and the monitor's overall functionality, not on its assistance to human readers via a comparative effectiveness study.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
The EK-Pro V14 arrhythmia detection algorithm is a component of the multi-parameter patient monitor. The submission states that the EK-ProV14 arrhythmia analysis algorithm is a change from the predicate's EK-Pro V13. The "standalone" performance of such internal algorithms is typically validated through internal testing against standardized datasets (e.g., annotated ECG databases for arrhythmia detection), which would fall under "Summary of Non-Clinical Tests: Bench testing related to software...". While the specific details of this internal algorithmic validation are not provided in the summary, the clearance implies it met the necessary performance and safety standards in a standalone manner as an embedded function.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
For the arrhythmia detection algorithm, the ground truth would typically be established by:
- Expert consensus: Review of ECG waveforms by cardiologists or electrophysiologists.
- Annotated databases: Use of publicly available or proprietary ECG databases with expertly adjudicated arrhythmia annotations.
However, since this is a 510(k) for substantial equivalence and not a de novo submission requiring extensive clinical validation, the specific ground truth methodology for the EK-Pro V14 algorithm's development or internal testing is not detailed in this public summary. The "ground truth" implicitly aligns with the performance established by the predicate device and relevant industry standards.
For the overall device, the "ground truth" for demonstrating equivalence relies on:
- Predicate device performance: The previously cleared CARESCAPE Monitor B650 with ESP V2 software (K131223) serves as the benchmark.
- Design specifications: The device is manufactured to meet predefined internal design specifications, which are validated through rigorous bench testing.
- Applicable consensus standards: Compliance with relevant international and national standards for medical devices (e.g., electrical safety, EMC, usability).
8. The sample size for the training set
The document does not mention the sample size for a training set. This is not a submission for a de novo AI device where such details for machine learning model training would be critical. The EK-Pro algorithm is likely a rule-based or traditional signal processing algorithm, or at least one iterated and improved over time, for which a distinct "training set" in the deep learning sense might not be explicitly defined or described in this type of regulatory document.
9. How the ground truth for the training set was established
Not applicable. As no training set is described or required for this type of 510(k) (which primarily relies on substantial equivalence and bench testing), the method for establishing its ground truth is not provided. If the EK-Pro V14 algorithm did indeed involve machine learning, its ground truth for training would have been established internally by GE Healthcare, likely through methods similar to those described in point 7 (expert annotations on ECG data), but these details are not part of the publicly available 510(k) summary.
§ 870.1025 Arrhythmia detector and alarm (including ST-segment measurement and alarm).
(a)
Identification. The arrhythmia detector and alarm device monitors an electrocardiogram and is designed to produce a visible or audible signal or alarm when atrial or ventricular arrhythmia, such as premature contraction or ventricular fibrillation, occurs.(b)
Classification. Class II (special controls). The guidance document entitled “Class II Special Controls Guidance Document: Arrhythmia Detector and Alarm” will serve as the special control. See § 870.1 for the availability of this guidance document.