(268 days)
The Lantos 3D Ear Scanner is intended for use by a trained hearing professional on patients 18 years of age and older presenting for inspection of the external ear canal.
The Lantos 3D Ear scanner:
- . provides a magnified visual image for inspection of the external ear canal and tympanic membrane, and provides illumination of the ear canal for inspection and;
- includes an expanding membrane that can be inserted and conforms to the external ear . canal when filled with an absorbing medium which enables the 3D scan and;
- records and presents to the hearing professional in an image file a topology measurement . of the external ear canal
The device consists of a hand held scanner, a single patient use disposable with a silicone conforming membrane, which when in use is filled with an absorbing medium, and a laptop computer. The hand held scanner, when not in use, is placed in a docking stand.
The hand held scanner is tethered to the docking stand with a custom cable. The laptop computer is mounted on the docking stand. The hand held scanner contains LED illumination, optics, a camera, a motor to move the probe tip, and a motor-driven reservoir.
User Interface (UI) software screens are displayed on the laptop and the user interacts with the device via the laptop keyboard and touchpad, and control buttons on the hand held scanner, as guided by the UI screens. Image data from the patient is stored in the laptop, and may be transmitted by the user via a wireless connection. No patient identifiable data is stored in the Lantos device.
A single use disposable, which includes the conforming membrane, is placed over the hand held scanner probe tip, in a manner similar to placing an ear speculum over the viewing tip of a conventional otoscope. The disposable includes a clear window over the probe tip. The probe can be used to inspect the ear canal, when inserted into the ear canal no closer than 4-5 mm from the tympanic membrane.
Additionally, the disposable can be filled with an absorbing medium so the membrane conforms to the ear, to enable the 3D scanning functionality of the device. After scanning is complete, the absorbing medium drains from the disposable so that the probe and disposable can be removed from the ear canal. The topographical measurement (scan) of the ear may be transmitted and may be used in the same manner as a desktop scan of a silicone earmold impression.
The provided text describes the Lantos 3D Ear Scanner and its regulatory submission (K121326). However, the document does not contain explicit acceptance criteria or a dedicated study section detailing performance against such criteria for AI/algorithm-driven features.
Instead, the performance evaluation focuses on the accuracy and safety of the 3D scanning functionality and general device characteristics.
Here's an attempt to extract relevant information and note what is missing based on your request:
Acceptance Criteria and Study for the Lantos 3D Ear Scanner
The Lantos 3D Ear Scanner is primarily described as an imaging and 3D measurement device for the ear canal. The provided document focuses on its technological characteristics, safety, and substantial equivalence to a predicate otoscope. It does not present an AI algorithm performance against specific diagnostic acceptance criteria.
The closest information related to performance evaluation concerns the accuracy of its 3D measurement capability.
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria Category | Acceptance Criteria (If Stated) | Reported Device Performance |
|---|---|---|
| 3D Topographical Measurement Accuracy | "required to have a mean difference between the model and the scan consistent with manufacturer's requirements for fabrication of devices inserted in the ear." (Quantitative values for "manufacturer's requirements" are not provided.) | "All scans were required to have a mean difference between the model and the scan consistent with manufacturer's requirements for fabrication of devices inserted in the ear." |
The method involved validating the output "using two known models (cylindrical and anatomical) with given values and obtaining multiple scans on each using multiple Scanner units." (Specific numerical results of the mean difference are not provided in this document.) |
| Photobiological Safety | Within ACGIH Threshold Limit Values for retinal thermal exposure, blue light exposure, actinic ultraviolet radiation, and near UV exposure. Risk Group 1 classification for blue-violet LED. | The Scanner falls within ACGIH Threshold Limit Values for retinal thermal exposure (within 10 seconds), blue light exposure (within 10,000 seconds), actinic ultraviolet radiation (within 8 hours), and near UV exposure.
Evaluated according to ANSI/IESNA RP-27.3-2007, resulting in a "Risk Group 1" classification for the blue-violet LED. Exposure time limit for Risk Group 1 was 15 minutes of direct, continuous exposure without membrane. |
| Biocompatibility | Conformance to ISO 10993 Parts 1-13 as applicable. | The device "meets the following Recognized Performance Standards: ISO 10993 Parts 1-13 as applicable." |
| Electrical Safety & EMC | Conformance to IEC 60601-1, IEC 60601-1-2, IEC 60601-2-18. | The device "meets the following Recognized Performance Standards: IEC 60601-1, IEC 60601-1-2, IEC 60601-2-18." |
| Quality/Risk Management System | Conformance to ISO 14971. | The device "meets the following Recognized Performance Standards: ISO 14971." |
2. Sample size used for the test set and the data provenance
- Test Set (for 3D Measurement Accuracy): The document states validation was performed "using two known models (cylindrical and anatomical)" and "obtaining multiple scans on each using multiple Scanner units."
- Sample Size:
- Number of models: 2 (cylindrical and anatomical)
- Number of scanners: "multiple Scanner units" (specific number not given)
- Number of scans per model/scanner: "multiple scans on each" (specific number not given)
- Data Provenance: Not applicable as it uses physical models rather than patient data. No country of origin is specified for these models or the testing activities. It is an internal validation rather than a retrospective or prospective patient study.
- Sample Size:
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- For the 3D measurement accuracy, the ground truth was established by the "known models (cylindrical and anatomical) with given values." This implies a physical, engineered ground truth rather than expert assessment.
- No experts were mentioned for establishing ground truth for the 3D measurement accuracy test.
4. Adjudication method for the test set
- Not applicable as the ground truth was based on measurements from known physical models, not on human interpretation or consensus.
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
- No, an MRMC comparative effectiveness study was not done. The document describes a device for producing visual images and 3D topographical measurements, not an AI-assisted diagnostic tool that would improve human reader performance. Its core function is to provide data, not to interpret it in an AI-driven manner for a human.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- The document describes the performance of the "Lantos 3D Ear Scanner" system, which includes hardware (scanner, disposable, laptop) and software. The "3D scan" is a direct output of the device's optical and mechanical system.
- While an algorithm processes the raw scan data to create the "topology measurement," this is presented as an inherent function of the device itself, not as an AI-driven standalone diagnostic algorithm whose performance needs to be assessed independently against human interpretation in a clinical context. The "algorithm" here is for data processing to generate the 3D model, not for diagnostic interpretation.
- Therefore, a standalone AI diagnostic algorithm performance study, as typically understood for AI medical devices, was not performed, nor is it applicable to the description of this device's reported functionality.
7. The type of ground truth used
- For the 3D topographical measurement, the ground truth was based on measurements from known physical models (cylindrical and anatomical models with given values).
- For safety aspects (photobiological, electrical, biocompatibility), ground truth was established by recognized industry standards (e.g., ACGIH, ANSI/IESNA, ISO, IEC).
8. The sample size for the training set
- The document does not mention a training set. This device, as described, generates a 3D digital model from optical scanning. It's not explicitly presented as an AI/machine learning model that learns from a dataset. The "ground truth" for its measurement accuracy is physical models, not a dataset it was trained on.
9. How the ground truth for the training set was established
- Not applicable, as no training set for an AI/ML algorithm is mentioned.
§ 874.4770 Otoscope.
(a)
Identification. An otoscope is a device intended to allow inspection of the external ear canal and tympanic membrane under magnification. The device provides illumination of the ear canal for observation by using an AC- or battery-powered light source and an optical magnifying system.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 874.9 only when used in the external ear canal.