(27 days)
The ALGO® 3 Newborn Hearing Screener is a mobile, noninvasive instrument used to screen infants for hearing loss. The screener uses AABR® technology. The screener is intended for babies between the ages of 34 weeks (gestational age) and 6 months. Babies should be well enough to be ready for discharge from the hospital, and should be asleep or in a quiet state at the time of screening.
Like the predicate device, the ALGO 3 detects the auditory brainstem response (ABR) to a series of stimulus clicks ('clicks'), which screens the entire hearing pathway from the outer ear to the brainstem. The ABR is not affected by the status of the middle ear, and evaluation of middle ear status is not required prior to ABR detection. The ALGO 3 generates soft clicks at 35 dB nHL ('normal hearing level' scale) which are delivered to the infant's ears through an Acoustic Transducer Assembly cable (ATA) to disposable earphones (Flexicouplers™ Disposable Earphones); a 35 dB nHL stimulus will detect a 30 dB nHL hearing loss. Each click evokes a series of identifiable brain waves from the auditory brainstem area of the infant's brain. Sensors (Jelly Tab™ Sensors) applied to the infant's skin pick up the brain wave response and transmit the signals to the screener via the Patient Cable Attachment (PCA) and Preamplifier (Preamp) assemblies. The screener uses advanced signal processing technology to separate the ABR from background noise and from other brain activity. The ALGO 3 automatically compares the baby's ABR to a template of ABRs derived from normal hearing infants (automated auditory brainstem response technology), and prints a "PASS" or "REFER" result. A "PASS" result is printed if the ALGO 3 can detect, with sufficient statistical confidence, that an ABR is present and consistent with its template. A "REFER" result is printed if the ALGO 3 cannot detect an ABR response with sufficient statistical confidence or one that is consistent with the template. A built-in printer prints the result on a self-adhesive label, which can be affixed directly to the infant's medical chart.
The ALGO 3 is also capable of screening using clicks at 40 dB nHL. Some newborn hearing programs (international programs, for example), and some programs prefer to set the screening threshold at 40 dB nHL rather than at 35. The incorporation of a 40 dB screening application in the ALGO 3 device makes the device more useful to more users.
The ALGO 3 device is easy enough for trained volunteers and other trained non-medical personnel to use. Interpretation of screening results is not required, as the screener automatically determines if the response is consistent with a template and automatically generates a "PASS" or "REFER" result, as previously discussed. In addition, training is minimal because of the easy-to-navigate graphical user interface, HELP topics in software, and a user tutorial in the ALGO 3 software.
Screening results can be tracked using the integrated ALGO DataBook® NHS Data Tracking System software. The DataBook utility retrieves and displays patient and screening data stored in ALGO 3 software. Data can be viewed on the ALGO 3's display; can be saved as an ASCII file on diskette for data management and reporting purposes; and can be exported to other databases for data management and reporting purposes.
There are no user-adjustable parameters, so that each screen is performed under exactly the same conditions. Click rate, click intensity, impedance threshold, and noise rejection thresholds are all pre-programmed into the device software, and users cannot change these screening parameters.
The ALGO 3 is composed of the following components, as is the Predicate Device:
- Screening module, with laptop computer
- Printer module, with label printer
- Cable assemblies (ATA, PCA, Preamp)
- Cart for transport and component storage
- Screening and accessory supplies
Accessories available for use with the ALGO 3 include single-use, disposable FlexiCouplers Disposable Earphones, and single-use, disposable Jelly Tab Sensors.
The provided text does not contain detailed information regarding specific acceptance criteria or a dedicated study proving the device meets said criteria with quantitative performance metrics. This document primarily focuses on the device's description, intended use, and substantial equivalence to a predicate device for 510(k) clearance.
However, based on the information provided, we can infer some aspects and highlight what is missing.
Here's an attempt to structure the answer based on your request, indicating where information is present and where it is not:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Inferred from functionality) | Reported Device Performance (Implied) |
|---|---|
| Accuracy: Ability to correctly identify "PASS" (hearing present) or "REFER" (hearing loss suspected) based on ABR response. | The ALGO 3 "automatically compares the baby's ABR to a template of ABRs derived from normal hearing infants... and prints a 'PASS' or 'REFER' result." "A 'PASS' result is printed if the ALGO 3 determines, with sufficient statistical confidence, that an ABR is present and consistent with its template. A 'REFER' result is printed if the ALGO 3 cannot detect an ABR response with sufficient statistical confidence or one that is consistent with the template." |
| Sensitivity for 30 dB nHL hearing loss: Detects hearing loss at this threshold. | "A 35 dB nHL stimulus will detect a 30 dB nHL hearing loss." |
| User proficiency: Ease of use by trained non-medical personnel. | "The ALGO 3 device is easy enough for trained volunteers and other trained non-medical personnel to use. Interpretation of screening results is not required... training is minimal because of the easy-to-navigate graphical user interface, HELP topics in software, and a user tutorial in the ALGO 3 software." |
| Screening time: Efficiency of the screening process. | "A typical screening process can be completed in 15 minutes or less." |
| Reproducibility/Consistency: Consistent results under the same conditions. | "There are no user-adjustable parameters, so that each screen is performed under exactly the same conditions. Click rate, click intensity, impedance threshold, and noise rejection thresholds are all pre-programmed into the device software, and users cannot change these screening parameters." |
The document does not provide quantitative performance metrics such as sensitivity, specificity, positive predictive value, or negative predictive value from a specific study to directly validate these criteria. The emphasis is on similarity to the predicate device, the ALGO-2 Newborn Hearing Screener, which likely had its own performance data for establishing substantial equivalence.
2. Sample Size Used for the Test Set and Data Provenance
This information is not explicitly stated in the provided text. The document describes the device's functionality and comparison to a predicate but does not detail a specific clinical trial or study undertaken for the ALGO 3's 510(k) submission that would involve a test set.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
This information is not explicitly stated in the provided text. The device uses an "automated auditory brainstem response technology" and "compares the baby's ABR to a template of ABRs derived from normal hearing infants." It does not mention human experts establishing ground truth for individual test cases in the context of a validation study for the ALGO 3's performance itself. The "template of ABRs" would have been established previously, potentially using expert interpretation from a historical dataset.
4. Adjudication Method for the Test Set
This information is not explicitly stated in the provided text, as there is no described specific test set or clinical study for the ALGO 3 featuring human adjudication. The device itself performs an automated "PASS" or "REFER" determination.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
A MRMC comparative effectiveness study was not described or indicated in the provided text. The ALGO 3 is designed to be automated, used by "trained volunteers and other trained non-medical personnel," with "interpretation of screening results is not required." Therefore, a study comparing human readers with and without AI assistance is not relevant to the described operational model of the device.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study
The device itself is a standalone algorithm in its core function. It "automatically determines if the response is consistent with a template and automatically generates a 'PASS' or 'REFER' result." The document implies that the device's performance, given its automation, is its standalone performance. However, no specific study detailing this standalone performance (e.g., sensitivity, specificity vs. a gold standard clinical diagnosis) is provided in this summary. The 510(k) submission relies on substantial equivalence to a predicate device, not necessarily a new standalone clinical trial for the ALGO 3 itself.
7. Type of Ground Truth Used
The ground truth for the device's operation is implicitly "ABRs derived from normal hearing infants." The device's algorithm compares the captured ABR to this established template. The ultimate ground truth relevant to "hearing loss" would be a clinical diagnosis of hearing loss, but the document describes the device comparing to a "template" of normal ABRs, which acts as the reference for its automated decision-making.
8. Sample Size for the Training Set
This information is not explicitly stated in the provided text. The "template of ABRs derived from normal hearing infants" would have been the equivalent of a training or reference set, but its size is not mentioned.
9. How the Ground Truth for the Training Set Was Established
The text states the ALGO 3 "compares the baby's ABR to a template of ABRs derived from normal hearing infants." This implies that the template was established by taking ABR readings from a cohort of infants confirmed to have normal hearing. How this "normal hearing" was confirmed (e.g., through follow-up audiological assessments, other screening methods, or clinical consensus) and the methodology for creating the "template" (e.g., statistical averaging, machine learning model) is not detailed in this document.
§ 882.1900 Evoked response auditory stimulator.
(a)
Identification. An evoked response auditory stimulator is a device that produces a sound stimulus for use in evoked response measurements or electroencephalogram activation.(b)
Classification. Class II (performance standards).