Search Results

The SafeT T-Piece Resuscitator is a gas-powered emergency resuscitator intended to provide emergency respiratory support by means of a face mask or a tube inserted into a patient's airway. It is intended for use with neonates and infants weighing less than 10 kg (22 lb).

The SafeT T-Piece Resuscitator is a single-use, non-sterile, manually operated, gas-powered resuscitator for use with patients less than 10 kg (22 lb). It is a simple T-Piece resuscitator with a manometer and the ability to adjust Peak Inspiratory Pressure (PIP) and Positive End-Expiratory Pressure (PEEP). It incorporates a pressure relief valve to protect against excessive pressure. The T-Piece resuscitator can be connected to the patient via a face mask or tube. The subject device consists of several components: T-Piece patient valve with variable PEEP dial and integrated manometer, Adjustable inspiratory pressure controller, 40 cm H2O pressure relief valve, 7' Oxygen tubing with a red universal (Fits-all) connector, 20" x 10 mm circuit tubing, Face mask.

The provided text is a 510(k) summary for the Ventlab, LLC SafeT T-Piece Resuscitator, a medical device for emergency respiratory support. It details the device's characteristics and its comparison to predicate devices to establish substantial equivalence for FDA clearance.

Crucially, this document does not describe a study involving an AI/Machine Learning device or a Multi-Reader Multi-Case (MRMC) study. It focuses on the substantial equivalence of a physical medical device (a resuscitator) based on its technological characteristics and performance testing against established standards, not on an algorithm's performance in interpreting medical images or data.

Therefore, many of the requested points, such as acceptance criteria for AI performance, sample sizes for test/training sets, expert adjudication, MRMC studies, standalone algorithm performance, and ground truth establishment for AI, are not applicable to the context of this document.

However, I can extract the closest analogous information regarding "acceptance criteria" and "study proving the device meets the acceptance criteria" in the context of this physical medical device:

1. A table of acceptance criteria and the reported device performance:

The document doesn't present a specific "acceptance criteria" table in the way one might for an AI device (e.g., sensitivity, specificity thresholds). Instead, "acceptance criteria" are implied by the compliance with applicable performance standards and the comparative data to predicate devices. The "reported device performance" is demonstrated by stating that the subject device "met all quantitative and qualitative requirements" of these standards and by the head-to-head comparison data presented in the "Substantial Equivalence Comparison Table" and subsequent discussion.

The most direct representation of "acceptance criteria" and "reported device performance" is found within the "Substantial Equivalence Comparison Table" itself, where critical performance metrics of the SafeT T-Piece Resuscitator are compared against those of the predicate device (Mercury Medical Neo-Tee) and a reference device (Fisher & Paykel NeoPuff).

Here is a partial table, extracted and reformatted from the "Substantial Equivalence Comparison Table" that acts as the primary evidence of meeting "acceptance criteria" (i.e., being substantially equivalent to the predicate device):

Ask a specific question about this device

The INOblender provides user set concentrations of inhaled Nitric Oxide (NO), in a balance of nitrogen, mixed into a user settable constant flow of oxygen gas that is being delivered to a patient. The intended use for the INOblender is as a back up to a primary nitric oxide delivery system or for short term attended use when a primary delivery device cannot practicably be used. This intended use includes applications within a medical facility and transport outside of a medical facility. The INOblender is not intended for use as a primary NO delivery system for long-term use.

The INOblender provides user set concentrations of inhaled Nitric Oxide (NO), in a balance of nitrogen, mixed into a user settable constant flow of oxygen gas that is being delivered to a patient. The INOblender is designed to take constant oxygen (02) gas flow (5 to 14 L/min) from the integrated O2 flowmeter and blend in NO at the setting on the NO blender's concentration control dial (5 to 80 ppm). The NO blender is calibrated for cylinder concentrations of 800 ppm NO in a balance of nitrogen (N2).

The provided document does not describe the acceptance criteria or a study proving the device meets acceptance criteria in the way typically associated with AI/ML diagnostic devices. This document is a 510(k) summary for a medical device called INOblender®, which is a Nitric Oxide administration apparatus.

Here's why the requested information cannot be fully provided from this document:

• Device Type: The INOblender® is a physical medical device (blender for gas delivery), not an AI/ML diagnostic software or algorithm. Therefore, the concepts of "algorithm only performance," "human readers improve with AI," "ground truth establishment" of a "test set" and "training set" in the context of diagnostic accuracy are not applicable.
• Study Type: The submission is primarily focused on demonstrating substantial equivalence to a predicate device, not on proving diagnostic accuracy or effectiveness through clinical trials in the sense of AI/ML performance.

However, I can extract information related to the device's non-clinical testing which serves as the "study" demonstrating its capabilities.

Here's a breakdown of the available information based on your request, adapted to the context of this physical device:

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly list "acceptance criteria" in a quantitative table with specific targets and achieved results. Instead, it lists four requirements that were tested for compatibility with a new respiratory care device. The "reported device performance" is the conclusion that these requirements were met.

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size for Test Set: Not applicable in the context of diagnostic data. The "test set" here refers to the device itself and its interaction with a specific new respiratory care device. The testing was conducted by setting up the INOblender® and the new respiratory care device. The number of such setups or repetitions is not specified, but it's not data in the sense of patient cases.
• Data Provenance: Not applicable in the context of diagnostic data. The "testing" refers to equipment performance validation, likely conducted in a laboratory setting by the manufacturer. Details like country of origin for such engineering tests are not provided and typically not relevant for this type of submission. The tests were "non-clinical."

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

• Number of Experts: Not applicable. Ground truth, in the sense of expert consensus on diagnostic interpretations, is not relevant for this type of device. The "ground truth" for these tests would be the known operational parameters and specifications of the devices themselves and the physical/chemical measurements taken during the compatibility and performance tests.
• Qualifications of Experts: Not applicable. The "experts" involved would be engineers and technicians performing the physical and chemical tests, adhering to established protocols. Their specific qualifications are not detailed in this summary.

4. Adjudication Method for the Test Set

• Adjudication Method: Not applicable. Adjudication methods like 2+1 or 3+1 are used for resolving discrepancies in expert interpretations of diagnostic data. For physical device performance testing, the results are typically objectively measured and compared against predefined engineering specifications.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size

• MRMC Study: No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This type of study is relevant for evaluating the impact of AI on human diagnostic performance, which is not the purpose of the INOblender®.
• Effect Size: Not applicable.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

• Standalone Performance: No, a standalone performance study in the context of an algorithm's diagnostic accuracy was not done. The INOblender® is a physical device that functions to deliver gas, not an algorithm. Its "standalone" performance would relate to its ability to accurately blend and deliver gases according to its specifications. The document states "INOblender® NO dose delivery accuracy" was tested, which can be considered a standalone performance aspect for the device's primary function.

7. The Type of Ground Truth Used

• Type of Ground Truth: The "ground truth" for the non-clinical tests was established by known engineering specifications, physical/chemical measurement standards, and the manufacturer's recommendations for calibration and operation. For example, "NO dose delivery accuracy" would be compared against a known, precise NO concentration generated by the device itself or measured by a calibrated external sensor.

8. The Sample Size for the Training Set

• Sample Size for Training Set: Not applicable. There is no AI/ML algorithm involved, so there is no "training set" in the context of machine learning. The device is hardware-based, relying on established physical and chemical principles.

9. How the Ground Truth for the Training Set Was Established

• Ground Truth for Training Set: Not applicable for the reasons mentioned above.

Summary of Device-Specific Information:

• Device Name: INOblender®
• Intended Use: To provide user-set concentrations of inhaled Nitric Oxide (NO) mixed into a constant flow of oxygen gas. Intended as a backup to a primary NO delivery system or for short-term attended use when a primary device is impractical.
• Technology: Component technology (blender, regulator, NO gas tank).
• Predicate Device: K052663 (another INOblender® with slightly different labeling for compatibility with resuscitators).
• Non-Clinical Tests:
  • O2 dilution
  • Effect on respiratory care device (compatibility with Fisher & Paykel NeoPuff)
  • INOblender® NO dose delivery accuracy
  • NO2 generation
• Clinical Tests: "The subject of this premarket submission... did not require clinical studies to support substantial equivalence." (Meaning, no clinical tests were performed for this specific submission, as it relied on non-clinical data and substantial equivalence to a predicate).

Ask a specific question about this device

The T-piece Resuscitation System provides the basic equipment required for pulmonary resuscitation of infants. Pulmonary resuscitation includes practices necessary to establish a clear airway and provide oxygen or air/oxygen mixtures and/or manual ventilation to the infant. These are clinical practices that represent the established standard of care. * Resuscitation may be required whenever an infant fails to establish effective, adequate breathing patterns necessary to meet tissue oxygen demands and/or to rid the body of carbon dioxide. For professional use only, by trained clinicians. * As stated in collaborative guidelines written by the American Heart Association (AHA) and the American Academy of Pediatrics (AAP) in the Textbook of Neonatal Resuscitation, 5th Edition.

The T-piece Resuscitation System incorporates the following features for the practice of infant resuscitation: a suction device for clearance of the trachea and nasal passages; two medical gas flowmeters to deliver oxygen or air/oxygen mixtures to the infant requiring such therapy; and an airway pressure manometer. An optional air/oxygen blender including high-pressure yokes may be included with the system which allows the clinician to adjust FiO2 % from 21-100%. The airway pressure manometer allows a trained clinician to see pressure throughout the respiratory cycle. Peak Inspiratory Pressure (PIP) is adjusted using the PIP knob located on the front panel of the resuscitation system that allows the clinician to set the maximum pressure being delivered to the infant in order to facilitate adequate pressurization of the lungs. Positive End Expiratory Pressure (PEEP) can be set using the adjustable PEEP valve located on the T-piece patient circuit. The T-piece resuscitation system is intended for use only with GE Healthcare Tpiece resuscitation circuits.

The provided text describes the Giraffe and Panda T-piece Resuscitation System, but it does not contain information about acceptance criteria or a study proving the device meets specific performance criteria through clinical data.

Instead, the document states:

"Pulmonary resuscitation of infants includes well established clinical practices; animal or clinical testing to support safety and effectiveness is not necessary. The conformance of the Giraffe and Panda T-piece Resuscitation System to performance specifications and to multiple recognized performance standards is being established through bench testing."

Therefore, I cannot fulfill your request for the following information based on the provided text:

1. A table of acceptance criteria and the reported device performance: The document mentions "performance specifications" and "recognized performance standards" but does not detail them or present a table of results.
2. Sample size used for the test set and the data provenance: No test set or clinical data is mentioned, as clinical testing was deemed (by the submitter) unnecessary.
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 or ground truth is described.
4. Adjudication method: Not applicable.
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: Not applicable. This device is a resuscitation system, not an AI-assisted diagnostic tool.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not applicable.
7. The type of ground truth used: Not applicable.
8. The sample size for the training set: Not applicable as this is not an AI/machine learning device.
9. How the ground truth for the training set was established: Not applicable.

The document indicates that the device's conformance is established through bench testing against performance specifications and recognized standards. However, the details of these tests and results are not provided in this 510(k) summary.

Ask a specific question about this device

The Fisher & Paykel Healthcare IW960 Servo-Control Wall-Mount CosyCot Infant Warmer and IW970 Manual-Control Wall-Mount CosyCot Infant Warmer are infant radiant warmers (as per 80 FMT, CFR §880.5130) containing an infrared heating element mounted above a built-in pediatric bed assembly, in order to maintain an infant's body temperature by means of controlled radiant heat.

The IW960 and IW970 are designed to provide warmth to babies in the first few weeks of life, when an infant's self-thermoregulation capacity may be reduced, or if external thermal support is required or desirable. This may include new-born babies in delivery room applications, including premature / low birth-weight infants, and support of critically ill babies in neonatal intensive care units (NICU's) or special care baby units (SCBU's). Situations which necessitate unobstructed access to an infant, including during resuscitation or surgical procedures, may indicate the need for a radiant heat source instead of equivalent support devices such as infant incubators.

The Fisher & Paykel Healthcare IW960 Servo-Control and IW970 Manual-Control Wall-Mount CosyCot Infant Warmers consist of a heater assembly, controller unit, support column, built-in bed assembly, and wall mounting hardware components. The heater assembly includes a single rod infrared healing clement housed inside a parabolic reflector. An observation lamp is mounted at the back of the heater unit. The thermoplastic enclosure is of similar cross-sectional shape to the reflector and is approx. 125mm high × 197mm wide × 625mm deep. It can be rotated to either side of the warmer, clear of the infant bed. A metal grill on the underside of the heater assembly prevents contact with the element. The heater assembly is mounted on top of the support column, which consists of a single aluminum extrusion section with thermoplastic front panel sections attached. Instruction / warning labeling is located on the back at the top of the support column, with a duplicate copy of label instructions provided. Identification / specification labeling is located on the front panel at the base of the support column, with the power inlet and auxiliary power outlet sockets. The dimensions of the support column are approx. 1448mm high × 193mm wide × 90mm deep (104mm deep at controller section top). The column section is supported by dual wall-mounting brackets, attached to the top and bottom of the column extrusion. The controller unit is built into the top of the support column. The transformer and Power PCB are mounted to the aluminum extrusion back section. The Control PCB is mounted on the inside of the thermoplastic front panel. This panel contains the control buttons, displays, main power switch and temperature sensor socket. Controls consist of buttons to select operating modes, timer functions and lamp operation. A control knob selects temperature or power level. LED displays include indicators for operating mode, alarms, timer, lamp and heater power, and 3-digit displays for temperature and timer readings. The bassinet assembly is supported by a four-bar link system attached to the support column extrusion. It consists of a stretched fabric surface over an aluminum extrusion frame. Removable barriers on each side may be folded down for access to an infant, and latch into clips in the bassinet frame. The bassinet may be tilted continuously through +10° to -10° to achieve Trendelenburg and Fowler positions. A spring-loaded cable brake system is used to change the tilt position. An x-ray tray module can be mounted underneath the bassinet for placement of an x-ray cassette. The bassinet assembly is approx. 650mm square × 86mm high, with side barrier panels 127mm above the bassinet surface. A variety of accessory mounting and storage options are available. Infant resuscitator and oxygen flowmeter modules may be mounted in column front panels. A gas supply module may be mounted to the back of the support column extrusion. The column extrusion features a channel mounting system in either side to support mounting blocks for further accessories, which may be fixed at the required height. These include short and long mounting poles, side shelves, gas supply and venturi suction unit mounting blocks and accessory hooks. Storage trays and bins may also be mounted under the bassinet area using this mounting system. In Baby mode, the IW960 provides stable control of the baby's skin temperature by automatically adjusting the heater power to compensate for varying metabolic and environmental conditions. In Manual mode, both the IW960 and IW970 provide useradiustable heater power. In Prewarm mode, the IW960 and IW970 maintain power at a constant level of 25% ready for use. A double thermistor sensor probe measures the baby's skin temperature, and audible and visual alarms alert the user to high or low temperature situations, equipment fault, power failure and periodic reminders to reassess the baby's clinical condition, depending on the control mode being used. Various independent safety features are included to control maximum output and avoid thermal injury to the infant.

The provided document describes the Fisher & Paykel Healthcare IW960 and IW970 Wall-Mount CosyCot Infant Warmers. It is a 510(k) Summary of Safety and Effectiveness Information for a medical device cleared in 1997. Due to the nature and age of the document, it does not contain the detailed, quantitative efficacy study data typically found in modern AI/ML device submissions. The information provided focuses on demonstrating substantial equivalence to a predicate device and meeting relevant performance standards for infant radiant warmers.

Here's an analysis based on the available information:

1. A table of acceptance criteria and the reported device performance

The document mentions that the proposed devices meet specific aspects of performance required by the standard for Infant Radiant Warmers, IEC 601-2-21. While it lists categories of performance, it does not provide specific numerical acceptance criteria or reported numerical performance values in a table format.

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

The document does not explicitly state the sample size for any test set in terms of patient data. The "Clinical verification studies" mentioned seem to refer to human-factors or usability testing in a clinical environment, rather than a clinical trial with a defined patient cohort for performance evaluation, as would be expected for AI/ML devices.

• Test Set Sample Size: Not specified.
• Data Provenance: Not specified, but "Clinical verification studies" imply a clinical setting, presumably in New Zealand, where the company is located. The studies would be considered prospective in nature for device validation.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

The document does not mention the use of experts to establish a "ground truth" for a test set in the context of diagnostic or interpretive outcomes, as would be relevant for AI/ML. The device is an infant warmer, and its performance is measured against physical parameters of temperature, irradiance, and stability, as well as its ability to generally warm infants.

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

Not applicable. There is no information about an adjudication process for this type of device and performance testing.

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

Not applicable. This device is an infant warmer, not an AI/ML diagnostic or assistive tool for human readers.

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

This refers to a physical device performance, not an algorithm. The "standalone" performance would be the device's ability to control temperature and irradiance under various conditions, which was tested. The "algorithm" for this device relates to the control circuitry for temperature regulation.

• Standalone Performance: Yes, the device's physical performance (temperature control, irradiance) was tested independently. The document states: "Performance testing for the IW960 and IW970 has been carried out in the areas of functional verification, temperature control, irradiance distribution patterns and clinical verifications."

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

The "ground truth" for this device's performance would be:

• Physical Measurements: Accurate and stable temperature readings from calibrated sensors.
• Engineering Standards: Compliance with the requirements of IEC 601-2-21 for various physical parameters (e.g., maximum irradiance, temperature distribution).
• Clinical Efficacy (Qualitative): The observation that the warmers successfully warm babies to a stable desired temperature, implying the "ground truth" is a clinically recognized stable infant temperature.

8. The sample size for the training set

Not applicable. This is not an AI/ML device that requires a training set of data. The device's control system is based on classical control theory and pre-programmed parameters, not machine learning.

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

Not applicable, as there is no training set for an AI/ML model. The "ground truth" for the device's control logic would be established through engineering design, physiological principles of thermoregulation, and compliance with safety standards for medical electrical equipment.

Ask a specific question about this device

The Fisher & Paykel Healthcare IW930 Servo-Control CosyCot Infant Warmer and IW950 Manual-Control CosyCot Infant Warmer are infant radiant warmers (as per 80 FMT, CFR §880.5130) containing an infrared heating element mounted above a builtin pediatric bed assembly, in order to maintain an infant's body temperature by means of controlled radiant heat.

The IW930 and IW950 are designed to provide warmth to babies in the first few weeks of life, when an infant's self-thermoregulation capacity may be reduced, or if external thermal support is required or desirable. This may include new-born babies in delivery room applications, including premature / low birth-weight infants, and support of critically ill babies in neonatal intensive care units (NICU's) or special care baby units (SCBU's). Situations which necessitate unobstructed access to an infant, including during resuscitation or surgical procedures, may indicate the need for a radiant heat source instead of equivalent support devices such as infant incubators.

The Fisher & Paykel Healthcare IW930 Servo-Control and IW950 Manual-Control CosyCot Infant Warmers consist of a heater assembly, controller unit, support column, built-in bed assembly, and base section.

The heater assembly includes a single rod infrared heating element housed inside a parabolic reflector. An observation lamp is mounted at the back of the heater unit. The thermoplastic enclosure is of similar cross-sectional shape to the reflector and is approx. 125mm high × 197mm wide × 625mm deep. It can be rotated to either side of the warmer, clear of the infant bed. A metal grill on the underside of the heater assembly prevents contact with the element.

The heater assembly is mounted on top of the support column, which extends down to the base of the unit, and consists of a single aluminum extrusion section with thermoplastic front panel sections attached. Instruction / warning labeling is located on the back at the top of the support column. Identification labeling is located on the front panel at the base of the support column, with the power inlet and auxiliary power outlet sockets. The dimensions of the support column are approx. 1448mm high × 193mm wide × 90mm deep (104mm deep at controller section top).

The controller unit is built into the top of the support column. The transformer and Power PCB are mounted to the aluminum extrusion back section. The Control PCB is mounted on the inside of the thermoplastic front panel. This panel contains the control buttons, displays, main power switch and temperature sensor socket.

Controls consist of buttons to select operating modes, timer functions and lamp operation. A control knob selects temperature or power level. LED displays include indicators for operating mode, alarms, timer, lamp and heater power, and 3-digit displays for temperature and timer readings.

The bassinet assembly is supported by a four-bar link system attached to the support column extrusion. It consists of a stretched fabric surface over an aluminum extrusion frame. Removable barriers on each side may be folded down for access to an infant, and latch into clips in the bassinet frame. The bassinet may be tilted continuously through +10° to -10° to achieve Trendelenburg and Fowler positions. A spring-loaded cable brake system is used to change the tilt position. An x-ray tray module can be mounted underneath the bassinet for placement of an x-ray cassette. The bassinet assembly is approx. 650mm square x 86mm high, with side barrier panels 127mm above the bassinet surface.

The base assembly consists of a steel main chassis with thermoplastic leg sections. The leg assemblies pivot at their connection points at each end of the main chassis to allow for height adjustment. The height may be mechanically fixed at one of three positions, or an optional electrical height adjustment system allows bassinet heights between 93 and 108 cm. It is activated by foot-controlled buttons on the right front leg of the unit. An electrical actuator system moving rods connected to the leg assemblies changes their angle with respect to the chassis and hence alters the warmer height. Four casters provide for positioning of the unit, two are lockable to prevent inadvertent movement. The main chassis of the base assembly is approx. 87mm high × 207mm wide × 582mm deep. The base has a rectangular 'footprint' area 770mm wide and a depth variable from 120mm to 110mm depending on height adjustment position.

A variety of accessory mounting and storage options are available. Infant resuscitator and oxygen flowmeter modules may be mounted in column front panels. A gas supply module may be mounted to the back of the support column. The column extrusion features a channel mounting system in either side to support mounting blocks for further accessories, which may be fixed at the required height. These include short and long mounting poles, side shelves, venturi suction unit mounting blocks and accessory hooks. Storage trays and bins may be mounted under the bassinet area also using this mounting system. Labeling is provided on the support column stating the maximum accessory loads which may be used depending on the height of the load, in order to ensure stability of the unit is maintained.

In Baby mode, the IW930 provides stable control of the baby's skin temperature by automatically adjusting the heater power to compensate for varying metabolic and environmental conditions. In Manual mode, both the IW930 and IW950 provide useradjustable heater power. In Prewarm mode, the IW930 and IW950 maintain power at a constant level of 25% ready for use.

A double thermistor sensor probe measures the baby's skin temperature, and audible and visual alarms alert the user to high or low temperature situations, equipment fault, power failure and periodic reminders to reassess the baby's clinical condition, depending on the control mode being used. Various independent safety features are included to control maximum output and avoid thermal injury to the infant.

Here's a breakdown of the acceptance criteria and the study information for the Fisher & Paykel IW930 and IW950 CosyCot Infant Warmers, based on the provided 510(k) summary:

The provided document is a 510(k) summary for a medical device and describes performance testing rather than a comparative study. Therefore, some of the requested information (like multi-reader multi-case studies, effect sizes, specific training set details, and expert adjudication for test sets) would not typically be found in this type of submission. This document focuses on demonstrating that the device meets safety and performance standards.

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size (Test Set): Not explicitly stated in terms of patient numbers or specific measurement counts. The document refers to "Performance testing" and "Clinical verification studies" as being carried out, implying a test set was used, but the size is not quantified.
• Data Provenance: The studies were likely conducted internally by Fisher & Paykel Healthcare, based in Auckland, New Zealand, given the company's origin and the submission of the 510(k) from there. It's almost certainly retrospective for the 510(k) submission, meaning the testing was completed before the submission date.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

This information is not provided in the document. The "clinical verification studies" would have involved medical professionals, but the role of "experts" in establishing a ground truth for a test set (as might be relevant for AI or diagnostic devices) is not detailed. For a medical device like an infant warmer, ground truth might refer to actual measured temperatures against a gold standard thermometer, which doesn't directly involve human expert consensus in the same way.

4. Adjudication Method for the Test Set

This information is not applicable/provided. Adjudication methods (like 2+1, 3+1) are typically used in studies where human readers are interpreting images or data, and their disagreements need to be resolved to establish a ground truth. For performance testing of a physical medical device like an infant warmer, direct measurements and adherence to engineering standards are the primary assessment methods.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, an MRMC comparative effectiveness study was not done. This type of study is relevant for diagnostic devices that involve human interpretation of results, often comparing human performance with and without AI assistance. This document describes the performance testing of a physical infant warmer, not a diagnostic or AI-assisted interpretation device. Therefore, there is no mention of "effect size of how much human readers improve with AI vs without AI assistance."

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, in essence. The performance testing described (functional verification, temperature control, irradiance distribution, clinical verifications) assesses the device's inherent capabilities and adherence to standards when operated as intended. While humans operate the device, the "performance" described refers to the device's output (e.g., stable temperature, accurate measurement) in achieving its therapeutic goal. The device itself (the IW930 with its servo-control) is an automated system whose "algorithm" (control logic) operates primarily standalone to maintain temperature.

7. The Type of Ground Truth Used

The ground truth used for performance assessment would primarily be based on:

• Direct Physical Measurements/Engineering Standards: Comparison of measured temperatures, irradiance levels, and control accuracy against established specifications and the international standard IEC 601-2-21.
• Clinical Outcomes/Observations: In the "clinical verification studies," the "ability of the warmers to warm up babies to a stable desired set temperature level accurately in a short period of time, and the ability to control the set temperature very accurately for a stable situation" would be observed against clinical goals for infant thermoregulation. This could involve comparing device readings to a gold standard thermometer for body temperature.

8. The Sample Size for the Training Set

This information is not provided and is largely not applicable in the context of this device and document. Infant warmers are not typically "trained" in the sense that AI algorithms are with large datasets. The device's control system is designed and calibrated based on engineering principles and regulatory standards, not through machine learning training.

9. How the Ground Truth for the Training Set Was Established

This information is not provided and is not applicable for the reasons stated above (the device is not "trained" with a ground truth dataset in the AI sense). The device's calibration and control parameters would be established through engineering design, testing, and adherence to performance specifications.

Ask a specific question about this device