Search Results

The HemoSphere Advanced Monitoring Platform is intended to be used by qualified personnel or trained clinicians in a critical care environment in a hospital setting. The Viewfinder Remote mobile application can be used for supplemental near real-time remote display of monitored hemodynamic parameter data as well as Faults, Alerts and Notifications generated by the HemoSphere Advanced Monitoring Platform.

The HemoSphere Advanced Monitoring Platform is intended for use with compatible Edwards Swan-Ganz and Oximetry Catheters, FloTrac sensors, Acumen IQ sensors, TruWave DPT sensors, ForeSight sensors, and ClearSight/Acumen IQ finger cuffs.

The HemoSphere Advanced Monitoring platform was designed to simplify the customer experience by providing one platform with modular solutions for their hemodynamic monitoring needs. The user can choose from the available optional sub-system modules or use multiple sub-system modules at the same time. This modular approach provides the customer with the choice of purchasing and/or using specific monitoring applications based on their needs. Users are not required to have all of the modules installed at the same time for the platform to function.

HemoSphere Advanced Monitoring Platform, subject of this submission, consists of the HemoSphere Advanced Monitor that provides a means to interact with and visualize hemodynamic and volumetric data on the monitor screen and its five (5) optional external modules: the HemoSphere Swan-Ganz Module (K163381 Cleared, April 14, 2017), the HemoSphere Oximetry Cable (K163381 Cleared, April 14, 2017), HemoSphere Pressure Cable (K180881 Cleared, November 16, 2018), HemoSphere Technology Module (previously referred to as "Tissue Oximetry Module"; K190305 cleared, August 29, 2019), HemoSphere ForeSight Module (K180003, May 10, 2018), and the HemoSphere ClearSight Module (K203687 cleared, May 28, 2021). The HemoSphere Advanced Monitor also has wired and wireless capabilities which was originally used only for connecting to a Hospital Information System (HIS) for data charting purposes. This capability is now used to allow it to stream continuously monitored data to the Viewfinder Remote, a mobile device-based application, for remote viewing the information (K211465 cleared July 8, 2021). The remotely transmitted data from the patient monitoring sessions include all hemodynamic parameter data and the associated physiological alarm notifications, historical trend data, and parameter waveform data.

The provided text describes several features and modifications to the HemoSphere Advanced Monitoring Platform, but it does not contain a specific table of acceptance criteria and reported device performance, nor does it detail a standalone AI algorithm study or an MRMC comparative effectiveness study for the machine learning features (like HPI or AFM).

The document primarily focuses on demonstrating substantial equivalence to predicate devices, and the performance data mentioned is generally high-level "All tests passed" rather than specific quantitative results against defined acceptance criteria for AI model performance.

Therefore, many of the requested details about the study that proves the device meets the acceptance criteria (especially for the AI/ML features) are not available in this document. The information that can be extracted relates more to the overall device functionality and compliance than to a detailed AI performance validation.

Based on the provided text, here's what can be extracted and what information is missing:

Information NOT available in the document regarding acceptance criteria and AI study specifics:

• 1. A table of acceptance criteria and the reported device performance for AI features: This level of detail is not provided. The document states that "Completion of all performance verification and validation activities demonstrated that the subject devices meet their predetermined design and performance specifications," and "Measured and derived parameters were tested using a bench simulation. Additionally, individual modules were tested at a system level to verify the safety of these modules." However, it does not enumerate specific acceptance criteria (e.g., accuracy, sensitivity, specificity thresholds) or quantitative performance results for the AI algorithms (HPI, AFM).
• 2. Sample size used for the test set and the data provenance, whether retrospective or prospective: This information is not provided. The document mentions "bench simulation" and "system level" testing but does not quantify the dataset size or its nature.
• 3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not specified.
• 4. Adjudication method (e.g., 2+1, 3+1, none) for the test set: Not specified.
• 5. If a multi reader multi case (MRMC) comparative effectiveness study was done, and the effect size of how much human readers improve with AI vs without AI assistance: Not mentioned. The document states, "No new clinical testing was performed in support of the subject 510(k)," indicating that specific MRMC studies for this submission were not conducted.
• 6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done: The document states that the "Acumen Hypotension Prediction Index feature is considered to be additional quantitative information regarding the patient's physiological condition for reference only and no therapeutic decisions should be made based solely on the Acumen Hypotension Prediction Index (HPI) parameter," and similarly for AFM, "No therapeutic decisions should be made based solely on the Assisted Fluid Management suggestions." This implies the AI features are intended for human-in-the-loop use, but it doesn't explicitly refer to a "standalone" performance study in the context of a typical AI performance evaluation. It does mention "Measured and derived parameters were tested using a bench simulation," which could include algorithm-only testing, but specific metrics are absent.
• 7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): Not specified.
• 8. The sample size for the training set: Not specified.
• 9. How the ground truth for the training set was established: Not specified.

Information that CAN be inferred or extracted (though limited for AI-specific performance):

• Regarding "Acumen Hypotension Prediction Index Feature" and "Acumen Assisted Fluid Management Feature":
  • The document explicitly states that for the Acumen HPI feature, "No modifications have been made to the previously granted AFM algorithm" (referring to the core algorithm, K203687), and for HPI Smart Alerts/Trends modifications, "There are no changes to the core HPI algorithm, the behavior of the HPI parameter display, nor the indications for use and intended use of the HPI parameter due to this Smart Trends/ Smart Alerts modification." This suggests that the core algorithms for HPI and AFM were previously cleared (DEN190029 for AFM and K203687 for HPI), and this submission focuses on incorporating and updating the display and connectivity of these features rather than re-validating the core algorithms themselves. This means the detailed AI performance validation, if done, would have been part of the previous submissions (DEN190029 and K203687).
  • The HPI feature defines hypotensive events as "mean arterial pressure

Ask a specific question about this device