IntelliSpace Perinatal Rev.K.00

• K Number: K173941
• Device Name: IntelliSpace Perinatal Rev.K.00
• Manufacturer: Philips Medizin Systeme Boeblingen GmbH
• Date Cleared: 2018-09-10 ( 258 days)
• Product Code: HGM , DSJ
• Regulation Number: 884.2740
• Type: Traditional
• Panel: OB
• Reference & Predicate Devices: K141015 , K082633 , K100420

Intended Use

The Philips IntelliSpace Perinatal Obsterical Information Management System is indicated for obsterior and after pregnancy, who require monitoring in a healthcare setting.

The Philips IntelliSpace Perinatal system provides:

• Basic and advanced fetal trace alarming for both antepartum and intrapartum patients.
• Central monitoring of maternal alarming.
• Documentation capabilities and data storage.
• Viewing and alarming of patient physiologic data, at remote locations, via the healthcare facility Remote Desktop Session (RDS).
• An interface to launch the Philips IntelliVue XDS Remote viewing and operating of compatible patient monitors.

Device Description

The Philips IntelliSpace Perinatal Rev. K.00 is a patient-oriented, departmental information management system for the obstetrical care environment. It covers OB care in so far as it is relevant for GYN visits, pregnancy, labor, birth and newborn documentation. It combines surveillance and alarming with comprehensive patient documentation and data storage into one system that covers the continuum of obstetrical care across one or more pregnancies, from the first antepartum visit until delivery and discharge.

AI/ML Overview

The provided document is a 510(k) summary for the Philips IntelliSpace Perinatal Revision K.00. This document describes the device, its intended use, and the testing performed to demonstrate its substantial equivalence to a predicate device.

However, it does not contain details about specific acceptance criteria related to a performance study (e.g., accuracy, sensitivity, specificity) for an AI/ML-based device. The document primarily focuses on non-clinical verification and validation (V&V) activities for software modifications and does not describe a clinical performance study with human readers or an algorithm-only standalone performance study.

Therefore, I cannot populate the requested table and answer many of the questions directly from the provided text. The information given is at a higher level, focusing on regulatory compliance for a software system that manages obstetrical information and integrates with other monitoring devices, rather than a diagnostic AI algorithm with quantifiable performance metrics.

Here is what I can extract and infer based on the document:

1. Table of Acceptance Criteria and Reported Device Performance:

• The document states that "All specified pass/fail criteria have been met" for the nonclinical V&V activities. However, it does not specify what those exact performance criteria are in terms of numerical metrics (e.g., sensitivity, specificity, accuracy) for an AI/ML component. The "performance" described is about the functionality and safety of the software modification, not a diagnostic accuracy metric.

  Acceptance Criterion Type	Specific Criterion (If available in document)	Reported Device Performance (If available in document)
  Functionality	Correctly hand over start-up parameters to XDS software	Met: "conducted tests demonstrate that start-up parameters provided by the modified IntelliSpace Perinatal software are correctly handed over to the XDS software"
  	Correct operation of patient monitors via XDS	Met: "compatible IntelliVue patient monitors can be correctly operated via the XDS software."
  Safety	Effectiveness of implemented risk mitigation measures	Met: "Test results confirmed the effectiveness of implemented risk mitigation measures."
  General V&V	All specified pass/fail criteria	Met: "All specified pass/fail criteria have been met."

2. Sample size used for the test set and the data provenance:

• Sample Size: Not explicitly stated as this was a software engineering V&V activity, not a clinical trial or performance study on a specific dataset. The tests were likely performed on a simulated environment or
  test cases designed to validate software functionality.
• Data Provenance: Not applicable in the context of a "test set" for an AI/ML performance study. The testing was of the software's functionality and safety, not its diagnostic accuracy on patient data.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience):

• Not applicable. The document describes software verification and validation, which typically involves software engineers and quality assurance professionals, not clinical experts establishing ground truth for diagnostic outputs.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

• Not applicable. As no clinical ground truth was established by experts for a test set, no adjudication method was mentioned.

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 MRMC study was performed or described. The document explicitly states: "Therefore, a clinical study was not needed for the changes provided with the Philips IntelliSpace Perinatal Obstetrical Information Management System software revision Rev.K.00." The modifications were software updates related to integration and data management, not an AI assisting human readers in diagnosis.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• No standalone algorithm performance study was performed or described. The device is a perinatal information management system, not a standalone diagnostic AI algorithm. Its function is to provide "Basic and advanced fetal trace alarming," "Central monitoring of maternal alarming," and "Documentation capabilities and data storage." These are system functions, not typically evaluated as a "standalone algorithm" in the same way a diagnostic AI would be. The closest described is functional testing of the software's ability to hand off data and control.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

• Not applicable. The ground truth in this context refers to the expected behavior and output of the software functions as defined by the system requirements and design specifications, not clinical markers. This is a V&V process for software, where "ground truth" would be software specifications and expected functional outcomes.

8. The sample size for the training set:

• Not applicable. The document describes the V&V of a non-AI software system; there is no mention of a training set as would be found with an AI/ML model.

9. How the ground truth for the training set was established:

• Not applicable, as no training set was used or mentioned.

Summary of what the document does convey regarding validation:

• The modifications to the Philips IntelliSpace Perinatal system (Revision K.00) are primarily software-based, focusing on:
  • Launching the Philips IntelliVue XDS Application for remote access to patient monitors.
  • Adapting web TS client resolution for mobile devices.
  • Adding configurable columns to electronic chalkboards.
  • Providing additional options for documentation and reporting.
  • HL7-based interfacing with ADT systems.
  • Updating Windows Server compatibility.
  • Adding more supported HL7-IHE profiles.
  • Introducing roles and permission-based patient data access control.
  • Adjusting alarm sound pressure levels.
  • Adjusting system scalability (fetal monitors and client sessions).
• Nonclinical V&V activities were performed:
  • Safety and Performance testing according to ANSI/AAMI/IEC 62304:2006.
  • Tests required by Hazard Analysis.
  • Functional tests of the modified software, including the new XDS launch feature, based on FDA guidance for software in medical devices.
  • The software was categorized as having a "MAJOR" level of concern.
• Conclusion from V&V: The tests demonstrated that start-up parameters are correctly handed over to XDS, and patient monitors can be operated via XDS. All pass/fail criteria were met.
• Clinical Data: "a clinical study was not needed" because the "similarities and differences... were determined not to have a significant impact on the device's performance, the clinical performance, and the actual use scenarios."

In essence, this document is a regulatory submission for a software update to an existing information management system, not a submission for a novel AI/ML diagnostic tool requiring a performance study against a clinical ground truth.

Regulation Number and Section

§ 884.2740 Perinatal monitoring system and accessories.

(a)
Identification. A perinatal monitoring system is a device used to show graphically the relationship between maternal labor and the fetal heart rate by means of combining and coordinating uterine contraction and fetal heart monitors with appropriate displays of the well-being of the fetus during pregnancy, labor, and delivery. This generic type of device may include any of the devices subject to §§ 884.2600, 884.2640, 884.2660, 884.2675, 884.2700, and 884.2720. This generic type of device may include the following accessories: Central monitoring system and remote repeaters, signal analysis and display equipment, patient and equipment supports, and component parts.(b)
Classification. Class II (performance standards).