FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The BTNeo System is indicated for measurement of a neonate's brain temperature. The brain temperature may be displayed with a compatible vital sign monitor system and is intended to be displayed along with core temperature.

The BTNeo System is indicated for use by qualified healthcare professionals that care for neonates (from birth through the first 28 days of life) in intensive care units, operating rooms, and recovery rooms.

Device Description

The BrainTemp Neonate (BTNeo) System is an external brain temperature measurement system that measures temperature at a depth of approximately 2.0 cm below the forehead sensor. The device system consists of the following components:

. Sensor: Used to measure temperature of the brain beneath the sensor. Single-use and affixed to the patient's forehead via medical grade adhesive.
0 Sensor Connector: Connects the system to the disposable sensor.
Cable with In-line Radiometer with Switch Module: Converts sensor readings to brain . temperature measurements via a proprietary algorithm.
. Display Interface: Connected to the Radiometer and displays brain temperature data and warning lights. There is an optional connection to a compatible vital sign monitor for display of brain temperature data in comparison to core body temperature.

AI/ML Overview

The provided text describes the BTNeo System, a brain temperature measurement device for neonates, and the studies conducted to support its De Novo classification. Here's a breakdown of the acceptance criteria and the study that proves the device meets them:

Acceptance Criteria and Reported Device Performance

The acceptance criteria for the BTNeo System are primarily outlined under the "Special Controls" section, which details the performance requirements for such a device. The "Performance Testing - Bench" and "Performance Testing - Animal" sections describe the studies conducted to meet these criteria.

Acceptance Criteria Category	Specific Criteria	Reported Device Performance and Study Description
In vivo performance	Device performs as intended for anticipated conditions of use and can accurately and reliably measure brain temperature compared to a ground truth measurement.	A Good Laboratory Practices (GLP) animal study using a weanling porcine model (n=3 piglets) was conducted. The device's performance was compared to an invasive brain temperature probe inserted approximately 2 cm below the skin surface in a cerebral hemisphere. The study involved warming and cooling procedures to demonstrate accurate measurement of brain temperature changes. The correlation between the invasive brain probe and the device was 0.97.
Non-clinical performance	Device can accurately measure changes in brain temperature under simulated conditions of use. Testing must assess repeatability within pre-specified, clinically relevant parameters. Technical specifications of hardware and software must be fully characterized.	Testing was conducted in accordance with IEC 80601-2-56: 2017. This involved evaluation across the operating temperature range (low, middle, high) and over a 24-hour use period using a phantom model (adult human scapula covered with synthetic pediatric skin) and an antenna simulator. The running average of measurements met the criteria of accuracy within ± 0.3 °C per Clause 201.101.2 of the standard for each test run.
Electrical Safety	Electrical safety, thermal safety, mechanical safety, and electromagnetic compatibility (EMC) testing must be performed.	The system was tested and passed according to ANSI/AAMI ES 60601-1-1:2005/(R)2012, IEC 60601-1-2:2014, IEC 61000-4-39:2017, and IEC 60601-1-8/AMD1:2012.
Software Documentation	Software documentation must include a detailed technical description of the algorithm(s) and be accompanied by verification and validation testing to ensure device and algorithm functionality, as informed by software requirements and hazard analysis.	Software verification and validation testing and documentation were provided according to FDA guidance. The device was determined to have a moderate level of concern. Documentation confirmed proper operation, and hazard analysis addressed potential risks. Cybersecurity documentation was also provided and included hazard analysis and mitigation.
Biocompatibility	The tissue-contacting device components must be demonstrated to be biocompatible.	Cytotoxicity testing (ISO 10993-5) and sensitization and irritation testing (ISO 10993-10) were conducted for the sensor component, which contacts intact skin for a prolonged duration.
Usability Evaluation	Usability evaluation must demonstrate that the intended user(s) can safely and correctly use the device, based solely on reading the directions for use.	Testing and evaluation conformed to IEC 60601-1-6: 2010+A1:2013, covering key use characteristics like sensor placement and re-application. All defined success criteria were met, supporting basic safety and essential performance. Peer-reviewed literature also supported the use of silicone adhesive on neonates.
Labeling	Labeling must include instructions for use, warnings (e.g., not for core temp, conditions impacting accuracy, skin injury risk, limitations for diagnosis/therapy), and summaries of in vivo testing (description of device outputs, study population/environment, data collection methods, observed adverse events/complications).	The labeling was deemed sufficient, satisfying 21 CFR 801.109. It includes device description, indications, contraindications, warnings (including those specified in the acceptance criteria, such as not for core temp, not for diagnosis/therapy, and skin integrity), precautions, and instructions for use. Summaries of the animal study findings are also included.

Study Details for Device Performance

2. Sample Size and Data Provenance

Test Set Sample Size: 3 weanling piglets (animal study).
Data Provenance: The animal study was a prospective GLP (Good Laboratory Practices) study conducted in an animal model (weanling porcine). The country of origin for the study is not explicitly stated in the provided text, but it's an animal study rather than human clinical data.

3. Number of Experts and Qualifications for Ground Truth

The ground truth for the animal study (invasive brain temperature probe) was established by direct physical measurement using a thermocouple inserted approximately 2 cm below the skin surface into a cerebral hemisphere. This method inherently serves as the expert-established ground truth for brain temperature within that model.
The text does not mention the need for or involvement of "experts" in the traditional sense (e.g., radiologists) to interpret images or reach a consensus for ground truth determination, as this was a direct physical measurement study. The expertise would lie in the veterinary and research staff conducting the GLP animal study. Their specific qualifications are not detailed beyond the study being GLP-compliant.

4. Adjudication Method for the Test Set

No adjudication method (e.g., 2+1, 3+1) was applicable or mentioned, as the ground truth was established by direct physical measurement (invasive thermocouple) in the animal model, not through expert review or consensus of subjective data.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC comparative effectiveness study involving human readers comparing AI-assisted vs. non-AI-assisted performance was done or mentioned. The device is a direct measurement system, not an AI-driven image analysis or diagnostic aid that would typically involve human reader studies.

6. Standalone (Algorithm Only) Performance

Yes, a standalone performance evaluation of the device's accuracy was a primary objective of both the bench testing and the animal study.
- Bench testing: The device's running average met the criteria of ± 0.3 °C deviation from the reference temperature in the phantom model.
- Animal study: The device's output (YSI Out) was directly compared to the invasive thermocouple, demonstrating a correlation of 0.97. This shows the algorithm's ability to measure brain temperature independently.

7. Type of Ground Truth Used

Bench Study: The ground truth was a precisely controlled and known temperature established by the testing apparatus in the phantom model, adhering to an international standard (IEC 80601-2-56: 2017).
Animal Study: The ground truth was invasive direct measurement using a thermocouple placed approximately 2 cm below the skin surface into a cerebral hemisphere in a live animal model. This serves as the "gold standard" or "outcome data" for brain temperature in that experimental setup.

8. Sample Size for the Training Set

The document does not explicitly mention a "training set" in the context of machine learning model development. This device likely relies on a physics-based or empirically derived algorithm rather than a machine learning model that would require a separate training dataset. The "training" of the device's performance would have occurred through the calibration and development process, which is not detailed as a separate data set in this
document.

9. How Ground Truth for Training Set Was Established

As indicated in point 8, a distinct "training set" with ground truth in the context of machine learning is not described. The device's algorithm would have been developed and refined through engineering and calibration processes, likely based on scientific principles and empirical data from laboratory experiments (similar to the described bench studies) and potentially early animal model work, but this is not presented as a specific "training set" with corresponding ground truth establishment details.

Summary

Regulation Number and Section

N/A