(28 days)
X Sensor is intended to collect dental x-ray photons and convert them into electronic impulses that may be stored, viewed and manipulated for diagnostic use by dentists.
X Sensor is a digital intraoral sensor which acquires digital intra-oral images. X Sensor acquires intra oral images with a sensor that is connected to a computer to produce an image almost instantaneously following exposure. The primary advantage of direct sensor systems is the speed with which images are acquired. X Sensor includes the firmware for the sensor and the previously cleared imaging software "ExDent-I".
The provided text is a 510(k) summary for the X Sensor, a digital dental intraoral sensor. It describes the device's technical characteristics, intended use, and its comparison to a predicate device (EzSensor XHD). However, this document does not contain the detailed information necessary to fully answer your request regarding acceptance criteria and the study proving the device meets these criteria in the context of AI/ML performance.
Specifically, the document focuses on regulatory clearance for a medical device (an X-ray sensor) based on hardware and image quality performance relative to a predicate device, as well as electrical, mechanical, and software safety. It does not mention any AI/ML components or studies evaluating AI/ML performance.
Therefore, many parts of your request, such as those related to AI/ML specific acceptance criteria, sample sizes for AI/ML test and training sets, expert adjudication, MRMC studies, or standalone algorithm performance, cannot be answered from the provided text.
However, I can extract information related to the device's general performance testing and comparison to the predicate device, which serves as a form of "acceptance" for medical device clearance.
Here's what can be extracted and how it relates to your request, with a clear indication of what information is not present:
Acceptance Criteria and Device Performance (General Device Performance)
Based on the document, the "acceptance criteria" appear to be meeting or exceeding the performance of the predicate device (EzSensor XHD) in key technical metrics and demonstrating adequate image quality for diagnostic use.
| Acceptance Criterion (Implicit) | Reported Device Performance (X Sensor) |
|---|---|
| Image Quality (General) | "The performance test result indicates that the X Sensor intra oral sensor performed equally to the EzSensor XHD, the predicated device, as both sensors have the same pixel pitch, thereby providing the same maximum line-pair resolution." |
| "The clinical images obtained from the X Sensor and EzSensor XHD were reviewed and rated comparatively." | |
| "The image quality in terms of contrast and resolution are overall similar for the X sensor, the proposed new device and EzSensor XHD, the predicate device." | |
| "There are no observable radiographic findings and no quality issues with intra oral diagnostic images provided by both sensors." | |
| "The proposed device produces overall better definition and grayscale of bony and soft tissue images." | |
| "In conclusion, both the proposed new device and the predicate device produced radiographic images with adequate quality for intra oral diagnosis in terms of resolution and anatomic details." | |
| Detective Quantum Efficiency (DQE) (6 lp/mm) | 0.258 (Better than predicate's 0.204) |
| Modulation Transfer Function (MTF) (3 lp/mm) | 0.889 (Better than predicate's 0.685) |
| Noise Power Spectrum (NPS) | Demonstrated better performance outcome than predicate. (Specific value not provided) |
| Maximum Resolution (lp/mm) | 33.8 (Same as predicate due to same pixel pitch) |
| Electrical Safety (IEC 60601-1 Series) | Compliance demonstrated. |
| Electromagnetic Compatibility (IEC 60601-1-2) | Compliance demonstrated. |
| Software Function (FDA Guidance) | Development followed "Content of Premarket Submissions for Device Software Functions." Provides "basic level of documentation for the firmware." |
| Cybersecurity (FDA Guidance) | Development, documentation, and testing followed "Cybersecurity in Medical Devices..." guidance. |
| Pediatric Information (FDA Guidance) | Development followed "Pediatric Information for X-ray Imaging Device Premarket Notifications" guidance. |
| Mechanical Durability (Drop & Vibration, etc.) | Performed, risks analyzed and mitigated (e.g., stainless-steel frame, soft silicon exterior for USB connector). |
Study Information (Based on Available Text)
-
Sample size used for the test set and the data provenance:
- The document states "The clinical images obtained from the X Sensor and EzSensor XHD were reviewed and rated comparatively."
- Specific sample size for the clinical images reviewed is NOT provided.
- Data provenance (country of origin, retrospective/prospective) is NOT provided. Given it's a 510(k) for a device like an X-ray sensor, the "test set" likely refers to physical images generated during bench testing and some limited clinical image capture, rather than a large dataset for AI/ML validation.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The document states "The clinical images obtained from the X Sensor and EzSensor XHD were reviewed and rated comparatively."
- The number of experts and their qualifications are NOT specified. This phrasing suggests a qualitative human review of generated images to ensure diagnostic utility.
-
Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Adjudication method is NOT specified. The review appeared to be a comparative assessment of image quality.
-
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 MRMC study was performed or discussed. This device is an X-ray sensor, not an AI-assisted diagnostic tool in the sense of an algorithm interpreting images for a human. It provides the images.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- This is NOT applicable. The "device" is the sensor itself, which captures images. There is no mention of an algorithm that performs standalone diagnostic interpretations.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- For the image quality assessment, the "ground truth" seems to be clinical utility/diagnostic adequacy as judged by human review of images generated by both the new device and the predicate device. Quantitative metrics (DQE, MTF, NPS) also served as objective performance measures.
-
The sample size for the training set:
- NOT applicable/NOT provided. This document describes a medical device, an X-ray sensor, not an AI/ML algorithm that requires a "training set" in the context of machine learning. The "training" for such a device would be its engineering and design optimization.
-
How the ground truth for the training set was established:
- NOT applicable/NOT provided. As above, there's no mention of a machine learning training set or associated ground truth.
Summary of Limitations:
The provided document is a regulatory submission for an X-ray sensor. It focuses on demonstrating substantial equivalence to a predicate device based on technical specifications, image quality, and regulatory compliance (electrical safety, EMC, software documentation, cybersecurity). It does not describe an AI/ML diagnostic algorithm or any studies related to its performance, and therefore cannot answer the specific questions posed about AI/ML acceptance criteria and validation.
§ 872.1800 Extraoral source x-ray system.
(a)
Identification. An extraoral source x-ray system is an AC-powered device that produces x-rays and is intended for dental radiographic examination and diagnosis of diseases of the teeth, jaw, and oral structures. The x-ray source (a tube) is located outside the mouth. This generic type of device may include patient and equipment supports and component parts.(b)
Classification. Class II.