(210 days)
The NTrainer System reinforces non-nutritive suck (NNS) in newborns and infants born prematurely.
A Soothie™ pacifier is inflated by a low pressure air pump that stimulates the baby's lips, tongue and jaw thus allowing the NTrainer System to reinforce non-nutritive suck (NNS) in newborns and infants. This is accomplished by providing the newborns or infants with a synthetic patterned oral somatosensory input. The NTrainer consists of the motor assembly, air cylinder, control electronics and computer system including a display. The only interface and contact with the baby is the disposable Soothie Pacifier. TM The application software has two modes of operation, NeoSuck, the assessment mode which defines the baby's state of non-nutritive suck development and, NTrainer, the therapy mode which reinforce the development of non-nutritive suck.
Here's a breakdown of the acceptance criteria and the study details for the NTrainer System, based on the provided 510(k) summary:
1. Table of Acceptance Criteria and Reported Device Performance
The provided document does not explicitly state pre-defined quantitative acceptance criteria (e.g., "device must achieve X% increase in NNS burst cycles"). Instead, it describes various performance factors that were measured to quantify improvement. The conclusion states that the device "performs better than the Sondrex PAL System in entraining NNS in newborns and infants."
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Reinforce non-nutritive suck (NNS) | Demonstrated improvements in: |
- Increase in NNS burst cycles / minute
- Increase in mouthing event / cycles / minute
- Increase in NNS bursts / minute
- Increase in NNS cycles / burst
- Increase in NNS cycles (% of mouthing)
- Reduction in non-NNS events / minute |
| Safety | "no adverse events during the study" |
| Effectiveness (compared to predicate) | "as effective and performs better than the Sondrex PAL System in entraining NNS" |
2. Sample Size and Data Provenance
- Test Set Sample Size: 30 newborns with moderate-severe Respiratory Distress Syndrome (RDS).
- 20 newborns in the NTrainer therapy group (RDS Test group)
- 10 newborns in the control group (RDS Control group)
- Data Provenance: Not explicitly stated (e.g., country of origin, specific institution). However, the software was developed by programmers at the University of Kansas, suggesting a US-based study is likely. The study is prospective as it involved providing "NTrainer therapy" and observing effects.
3. Number of Experts and Qualifications for Ground Truth
The document does not mention the use of experts to establish a "ground truth" in the traditional sense of diagnostic accuracy for the clinical study. The performance factors measured (e.g., NNS burst cycles/minute) appear to be objective physiological measurements, likely recorded directly by the device or observed and quantified by study personnel. Therefore, there's no mention of a ground truth established by independent experts for the clinical outcomes.
For the bench tests, the "specifications" were used as the standard, but the origin or authorship of these specifications (i.e., if they were validated by independent experts) is not detailed.
4. Adjudication Method for the Test Set
Not applicable. This study focused on measuring changes in physiological parameters in response to therapy rather than classifying data or making diagnostic decisions that would require adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a multi-reader multi-case (MRMC) comparative effectiveness study was not performed. This type of study typically involves multiple human readers evaluating cases with and without AI assistance to measure improvement in diagnostic performance. The NTrainer is a therapeutic device, and the study evaluated its therapeutic effect, not its diagnostic accuracy.
6. Standalone (Algorithm Only) Performance Study
The clinical study evaluated the NTrainer System (device and software) in use with human infants. Thus, it was not a "standalone" algorithm-only performance study in the sense of a diagnostic AI without human interaction. The device itself is designed to provide therapy, not an assessment that can be interpreted by humans. The "NeoSuck, the assessment mode" defines the baby's state, but this assessment is part of the device's function, not a separate algorithm evaluated in isolation from the hardware.
7. Type of Ground Truth Used
The "ground truth" in the clinical study was based on objective physiological measurements and recorded "performance factors" related to non-nutritive suck. These measurements quantifiably describe the baby's NNS development and activity before and after therapy. It is not an expert consensus, pathology, or outcomes data in the usual sense for diagnostic devices.
8. Sample Size for the Training Set
The document does not provide information on a specific "training set" sample size for the NTrainer software. It mentions that the "NTrainer and NeoSuck application software were originally developed by programmers at the University of Kansas (KU)." This suggests the software likely evolved over time, possibly with iterative development and testing, rather than a single distinct "training set" as might be seen with a modern machine learning model.
9. How Ground Truth for the Training Set Was Established
Given the information provided, there's no explicit description of a "ground truth" establishment process for a training set. The software development likely involved:
- Engineering specifications for the device's intended function (e.g., how the pacifier should be inflated, how pressure should be measured).
- Clinical understanding of NNS patterns and desired therapeutic responses in infants.
- "Review and determined to be software having a 'Minor Level of Concern' in accordance with FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices," which implies adherence to software development and validation best practices, but not necessarily a "ground truth" in the diagnostic sense.
§ 882.5050 Biofeedback device.
(a)
Identification. A biofeedback device is an instrument that provides a visual or auditory signal corresponding to the status of one or more of a patient's physiological parameters (e.g., brain alpha wave activity, muscle activity, skin temperature, etc.) so that the patient can control voluntarily these physiological parameters.(b)
Classification. Class II (special controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter when it is a prescription battery powered device that is indicated for relaxation training and muscle reeducation and prescription use, subject to § 882.9.