(199 days)
The INVOS™ Patient Monitor, model PM7100, is a noninvasive cerebral/somatic oximetry system intended for use as an adjunct monitor of regional hemoglobin oxygen saturation of blood in the brain or in other tissue beneath the sensor. It is intended for use on individuals > 2.5 kg at risk for reduced-flow or no-flow ischemic states. For patients ≤ 2.5kg, the INVOS™ Patient Monitor, model PM7100 is only intended for adjunct trend monitoring of regional hemoglobin oxygen saturation of blood tissue beneath the sensor.
It is also intended for use as an adjunct trend monitor of regional hemoglobin oxygen saturation of blood tissue beneath the sensor in any individual.
The INVOS™ Pediatric rSO2 sensor is indicated for single patient use when cerebral/somatic monitoring of site-specific regional oxygen saturation (rSO2) is required in patients weighing greater than 4 kg and less than 40 kilograms. This sensor is only intended to be used with INVOS™ PM7100 system with INVOS™ Near Infrared Spectroscopy (NIRS) technology. For additional information regarding setup and use of the INVOS™ PM7100 System including indications for use, contraindications, warnings and cautions, consult the Monitoring System Operator's Manual.
The INVOS™ OxyAlert NIRSensor disposable sensor Model IS is indicated for single patient use when cerebrallsomatic monitoring of site-specific regional oxygen saturation (rSO2) is required in patients weighing > 2.5 kilograms and
The INVOS™ PM7100 Patient Monitor is a cerebral/somatic tissue oximeter intended for use as an adjunct trend monitor of regional hemoglobin and oxygen saturation monitoring. The monitor utilizes a near infrared diffuse reflectance spectroscopy system employing near infrared light at four wavelengths for the adult and pediatric system configurations. One pair of wavelengths is used to estimate the percentage of hemoqlobin saturated with oxygen in tissue beneath the sensor; another pair of wavelengths is used for the sensor on/off detection algorithm. The infant system configuration currently only employs the two wavelengths needed to estimate regional oxygenated hemoglobin.
The subject device is non-sterile and consists of a multi-channel touch screen display, preamplifier, cables, and three single use sensor types for use in the adult (PMSENS71-A; cleared in K182868), pediatric (PMSENS71-P; new to subject device), and infant (IS; new to subject device) populations.
The subject device utilizes up to four detachable sensors to collect signals, and up to two preamplifiers receive signals from the sensors, digitize the signals, process the data and then periodically estimate the rSQ2 at each sensor site. The preamplifiers then transmit the measured and calculated parameter data to the monitor where the information is displayed. The oximeter is powered primarily by AC power at 100 VAC to 240 VAC ±10% and is equipped with an internal rechargeable lithium-ion battery for intra hospital transport and back-up purposes. The INVOS™ PM7100 Patient Monitor is intended for use in hospitals, and is not intended for home use or out-of-hospital transport.
The PM7100 Monitor configures the PMPAMP71 preamplifier modules for monitoring and allows the user to configure sensor placements on a patient's body as well as establish baseline rSQ2 values. Device features include a user configurable rSO2 baseline, alarms, signal strength indicator and area under the curve thresholds. The monitor measures and displays an rSO2 trend line in a graph for the estimated regional oxygen saturation value unique to the specific area under each sensor. the baseline rSQ2 value, the current estimated rSQztrend accuracy value and percent change from patient rSO2 baseline. The PM7100 Monitor also displays alarm information and indicates connected sensor type. The Monitor is equipped with technical (system status) and physiological (patient status) alarms. Alarm conditions are detected via the sensor, the physiological and technical information are then processed in the preamplifier/processor, which then communicates this information to the monitor then provides a visual and audio alarm notification. The device permits the user to silence alarms, mark events, and manage case history data.
The PMPAMP71 preamplifier interfaces with the PM7100 monitor via a cable for communications and power, and with one or two sensors via reusable sensor cables to receive optical signals. The optical signal flows from the sensor into the PMPAMP71 which in turn generates saturation (rSO2) for tissue under the sensor that is communicated to the PM7100 monitor.
The INVOS™ Adult SpO2Sensor, PMSENS71-A, is a non-sterile, non-invasive, disposable sensor intended for application on cerebral and somatic sites. The PMSENS71-A was designed for use with the INVOS™ PM7100 and 5100C cerebral/somatic monitoring systems for monitoring of site-specific regional oxygen saturation (rSO2) in adult patients weighing >40 kilograms. There have been no significant changes to this sensor since clearance under K182868.
The INVOS™ Pediatric rSO2 Sensor, PMSENS71-P, is a new non-sterile, non-invasive, disposable sensor intended for application on cerebral and somatic sites in pediatric patients greater than 4 kg and less than 40 kilograms. The PMSENS71-P was designed to support the existing two wavelength rSO2algorithm along with a new two wavelength sensor on/off detection algorithm. Accordingly, the PMSENS71-P sensor is designed to emit and collect sensor data with a total of four wavelengths.
The INVOS™ Infant Regional Saturation Sensor, model IS, is a non-sterile, non-invasive, disposable sensor intended for application on cerebral and somatic sites in the infant and neonate patient population. The IS sensor configuration currently only employs the two wavelength m. Accordingly, the IS sensor is designed to emit and collect sensor data with a total of two wavelengths. The INVOS™ Reusable Infant Sensor Adapter Cable (PMAC71RIC) is a new non-sterile, non-invasive reusable cable intended for adapting the infant sensor connection to the PM7100 preamplifier.
The INVOS™ Docking Station connects to the INVOS™ PM7100 Patient Monitor and is designed with a mounting surface on the back of the docking station which can connect to the Patient Monitor Stand or any other compatible mounting service. The docking station is equipped with inputs for AC power, USB, serial port and VGA port. The docking station also facilitates recharging of the back-up battery. There have been no significant changes to the docking station since clearance in K182868.
The INVOS™ Patient Monitor Stand is an accessory that connects to the INVOS™ Docking Station. The monitor stand has not changed since clearance in K 182868.
The document provided describes the acceptance criteria and the study that proves the device meets the acceptance criteria for the INVOS™ PM7100 Patient Monitor, INVOS™ Pediatric rSO2 Sensor, and INVOS™ Infant Regional Saturation Sensor.
Here's the breakdown of the information requested:
1. Table of Acceptance Criteria and Reported Device Performance
Device: INVOS™ PM7100 System (with Infant/Neonatal IS sensor)
| Acceptance Criteria | Reported Device Performance |
|---|---|
| rSO2 trend precision ≤ 2.9% standard deviation of the rSO2 trend error | 2.09% rSO2 for cerebral |
| 1.96% rSO2 for somatic |
Device: INVOS™ PM7100 System (with Pediatric rSO2 sensor, PMSENS71-P)
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Mean bias of rSO2 measurements (compared to predicate SPFB sensor with INVOS™ 5100C) does not exceed ±4% rSO2 | Mean bias = ±0.466% rSO2 |
| Successfully detecting sensor removal within 1 minute | 100% |
| Successfully detecting sensor application within 15 seconds | 100% |
| Successfully detecting the sensor is off the subject | 96.43% |
| Successfully detecting the sensor is on the subject | 100% |
2. Sample Size Used for the Test Set and Data Provenance
Infant/Neonatal Sensor Study (Hypoxia Study):
- Sample Size: 24 subjects
- Data Provenance: Not explicitly stated regarding country of origin, but described as a non-invasive hypoxia study conducted on healthy, non-smoking adults and adolescent volunteers. Given the FDA submission context, it's highly likely to be conducted in the US. This was a prospective clinical study.
Pediatric Sensor Feasibility Study (Room Air Study):
- Sample Size: 28 subjects
- Data Provenance: Not explicitly stated regarding country of origin. This was a prospective clinical feasibility study.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
The document does not explicitly mention the use of experts to establish a "ground truth" in the traditional sense of image annotation or disease diagnosis for the clinical studies.
- For the Infant/Neonatal sensor study, the "ground truth" or reference for the rSO2 trend precision appears to be the cleared INVOS™ 5100C system in conjunction with the Infant/Neonatal Sensor (predicate system).
- For the Pediatric sensor feasibility study, the "ground truth" or reference for rSO2 performance was the predicate INVOS™ 5100C system in conjunction with the Pediatric sensor, SPFB. Sensor on/off detection accuracy was evaluated against observed physical states.
Therefore, this aspect of ground truth establishment by human experts, as typically seen in AI/ML performance studies, is not applicable in the context of this device's testing.
4. Adjudication Method for the Test Set
Adjudication methods (e.g., 2+1, 3+1) are typically used for establishing ground truth in studies involving human interpretation (e.g., radiology reads). As the studies described here involve direct physiological measurements and comparisons to a predicate device, or assessment of a sensor's physical state detection, an adjudication method in this sense is not applicable. The "ground truth" was established by the measurements from the predicate device or the direct observation of sensor states.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size of Human Readers Improving with AI vs. Without AI Assistance
No, an MRMC comparative effectiveness study was not done. The device is an oximeter, and the studies focused on its measurement accuracy and trend precision compared to a predicate device, and the functionality of new features like sensor on/off detection. It is not an AI/ML-based diagnostic device that assists human readers.
6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done
Yes, in essence, the "Performance Data Summary" describes standalone performance for the specific algorithms and hardware of the device.
- Non-clinical Performance Testing: This section details bench-top optical phantom testing to verify static rSO2 calculation, confirming the subject device's rSO2 algorithm performs within acceptable limits compared to the predicate system. This is a standalone performance test.
- Clinical Studies: The clinical studies evaluated the direct output of the device (rSO2 measurements, trend precision, sensor on/off detection) against a reference (predicate device or physical observation), without direct human-in-the-loop interpretation impacting the device's output. While a human uses the device, the performance metrics are about the device's intrinsic measurement capabilities.
7. The Type of Ground Truth Used
- Infant/Neonatal Sensor Study: Comparative "ground truth" was established based on measurements from the cleared INVOS™ 5100C system in conjunction with the Infant/Neonatal Sensor (predicate system) during induced hypoxia. This is a form of comparative reference device data.
- Pediatric Sensor Feasibility Study (rSO2 performance): Comparative "ground truth" was established based on measurements from the cleared INVOS™ 5100C system in conjunction with the Pediatric sensor, SPFB (predicate system) at room air. This is also a form of comparative reference device data.
- Pediatric Sensor Feasibility Study (sensor on/off detection): The ground truth was based on direct observation of the sensor's physical state (on subject, off subject, face up/down on sheet, hanging).
8. The Sample Size for the Training Set
The document does not describe the use of a "training set" in the context of machine learning. The studies described are verification and validation studies (test sets) for the device's performance, not for training a model. The algorithms are based on Near Infrared Spectroscopy (NIRS) technology, not on machine learning that requires a specific training phase with a dedicated dataset.
9. How the Ground Truth for the Training Set Was Established
Not applicable, as no machine learning training set is described or utilized in the regulatory submission for this device. The device's operation is based on established NIRS principles and algorithms, rather than data-driven machine learning models requiring large labeled training sets.
§ 870.2700 Oximeter.
(a)
Identification. An oximeter is a device used to transmit radiation at a known wavelength(s) through blood and to measure the blood oxygen saturation based on the amount of reflected or scattered radiation. It may be used alone or in conjunction with a fiberoptic oximeter catheter.(b)
Classification. Class II (performance standards).