Search Results

The IC-WHCD100 (Inspire) is intended to be used as an aid in the detection and diagnosis of dental caries.

The IC-WHCD100 is a toothbrush-sized handpiece used for diagnosis of caries. A USB cable is used to connect the handpiece to a personal computer with a dental imaging software. After a camera cover is placed over the end, the handpiece is positioned over the tooth to be examined. The camera takes images by illuminating the tooth surface with a white LED light for regular tooth image. With fluoresced light, the device can show bacteria on the surface of tooth. With infrared light, the device can show tooth cavity by highlighting enamel. The user can view the images on 510k cleared dental imaging software such as Apteryx vision (K983111), Romexis (K171385), Sidexis (K132773), etc.

The provided text details the FDA 510(k) summary for the IC-WHCD100 (Inspire) device, which is an intraoral camera intended as an aid in the detection and diagnosis of dental caries. However, the document primarily focuses on demonstrating substantial equivalence to predicate devices and provides limited information regarding specific acceptance criteria and detailed study results. Critical information needed to fully answer the request, such as a precise table of acceptance criteria and reported device performance with numerical metrics (e.g., sensitivity, specificity for caries detection), detailed sample size for the test set, number and qualifications of experts for ground truth, adjudication methods, details of comparative effectiveness studies (MRMC), or a comprehensive standalone performance study report are not explicitly present in the provided text.

Based on the available information, here's what can be extracted and what remains unknown:

1. Table of Acceptance Criteria and Reported Device Performance

The document mentions "Performance Test for imaging (Image Sharpness, Image Size, Image Resolution, tooth Caries Detection)" as a non-clinical test. However, it does not provide a specific table of acceptance criteria or the reported device performance metrics (e.g., sensitivity, specificity, accuracy) for caries detection. It only broadly states that "the performance test results of the subject device supports that the transillumination mode works well despite this difference." and "the performance test result supports that the subject device is substantially equivalent to the predicate devices."

Unfortunately, a specific table with numerical acceptance criteria and corresponding performance data for caries detection is not found in the provided text.

2. Sample Size Used for the Test Set and Data Provenance

The document states "performance test results of the subject device supports that the transillumination mode works well despite this difference." and refers to "Performance Tests (Non-clinical)". However, the specific sample size used for the test set (number of teeth, lesions, or patients) and the data provenance (e.g., country of origin of the data, retrospective or prospective nature of data collection) are not disclosed in this 510(k) summary.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

The provided text does not specify the number of experts used to establish ground truth or their qualifications. It only refers to "Performance Tests" related to caries detection, implying some form of ground truth was used, but details are absent.

4. Adjudication Method for the Test Set

The document does not describe any adjudication method (e.g., 2+1, 3+1, none) used for establishing the ground truth for the test set.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study Was Done

The provided text does not mention or describe a Multi Reader Multi Case (MRMC) comparative effectiveness study where human readers' performance with and without AI assistance was evaluated. The focus is on the device's performance relative to predicate devices, not human-AI collaboration.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) Was Done

The document broadly states "Performance Test for imaging... tooth Caries Detection)". While the device has a caries detection aid capability, the summary does not explicitly detail a standalone algorithm-only performance study with specific metrics (e.g., sensitivity, specificity of the algorithm itself). The device is described as an "aid in the detection and diagnosis," implying human involvement.

7. The Type of Ground Truth Used

The document refers to "tooth Caries Detection" as one of the performance tests. However, the specific type of ground truth used (e.g., expert consensus, pathology/histology, clinical outcomes data, or a combination) is not explicitly stated.

8. The Sample Size for the Training Set

The 510(k) summary does not provide any information regarding a training set sample size. This type of document typically focuses on performance testing for regulatory clearance rather than details of model development. Given that the device is an "intraoral camera with Caries Detection Aid" using specific light sources (405nm and 940nm) to highlight bacteria and cavities, it's possible its "detection aid" might be based on optical properties rather than a complex AI model requiring extensive training data in the traditional sense, but this is speculative given the lack of detail.

9. How the Ground Truth for the Training Set Was Established

Since there is no mention of a training set, there is no information on how its ground truth was established.

Ask a specific question about this device

The x-ray system creates data for digital exposures in the maxillofacial area and in subareas for dentistic dentistry, for hard-tissue diagnostics within ENT medicine, and carpus exposures.

The proposed Axeos is a dental cone-beam CT system (CBCT), which comprises sensor units for 2D cephalometric exposures. 2D panoramic radiograph and 3D volume exposure. The combination of sensors varies depending on the installed device configuration. In the proposed device, the 2D sensor, as well as the 2D cephalometric sensor, are identical to the corresponding sensors utilized in the predicate device (K150217). However, with respect to the 3D CBCT exposures, the proposed Axeos device utilizes a new version of the flat panel sensor with a larger field of view compared to the flat panel version utilized within the predicate Orthophos SL (K150217).

The proposed Axeos device uses an x-ray beam that rotates around the patient's head. Detectors acquire two-dimensional x-ray images at varying radiographic angles. An enhanced software algorithm generates the 2D panoramic and 2D cephalometric images and the reconstruction algorithm reconstructs the 3D volumetric image from the raw image data.

The exposed area can be adapted to a specific region of interest to keep the radiation dose as low as possible for the patient. This is achieved by collimating the x-ray beam and the adjustment of starting and ending points of the sensor/x-ray source movement. Furthermore, the radiation dose can be adapted by various parameters such as program types and exposure technique factors. These functions are available for CBCT, cephalometric and panoramic exposures.

A Class I laser beam is utilized to define reference lines for the correct patient position. The patient is stabilized through different bite blocks and the motor-driven forehead and temple supports.

The obtained digital image data are processed to provide a reconstructed image to the operator/user. The reconstructed images are transferred to the currently marketed SIDEXIS 4 image processing software (K132773) and stored in the SIDEXIS 4 software database. The proposed Axeos device uses the same Sidexis 4 image processing software (K132773) as does the predicate device (K150217) and there are no changes to the SIDEXIS 4 software (K132773) introduced in this premarket notification.

The proposed Axeos device includes metal artifact reduction software feature which automatically reduces image artifacts caused by radiopaque objects. This identical software feature is also included in the predicate Orthophos SL (K150217) device and remains unchanged in the proposed device.

A user control panel allows user actions as: height adjustment, selection of programs, and exposure parameters and delivers information about the unit status.

For 3D imaging, the proposed Axeos allows the user to select volume sizes within which multiple fields of view, anatomic positions, and collimation may be selected. This is intended to allow the clinician to select the field of view based on the diagnostic need and to minimize the dose exposed to the patient.

The main components of the proposed Axeos device are:

• . X-ray source
• . X-ray detector (flat panel and/or PAN sensor)
• Operator panel
• . Laser locator
• Cephalometric arm with detector (Ceph sensor)
• Remote control (only by wire).
• . Test phantoms: Exposure phantom, Constancy test phantom, Contrast element, and Ceph test phantom.

The provided text describes the Dentsply Sirona Axeos, a dental cone-beam CT system, and its substantial equivalence to a predicate device (Orthophos SL, K150217). The document focuses on regulatory approval rather than clinical studies to establish diagnostic performance against specific acceptance criteria. Therefore, most of the requested information regarding acceptance criteria, specific study designs, sample sizes, expert involvement, and ground truth establishment cannot be extracted directly from this document.

However, I can provide information on what is present in the document that relates to performance and equivalence.

1. Table of Acceptance Criteria and Reported Device Performance

The document does not specify quantitative acceptance criteria for diagnostic performance outcomes (e.g., sensitivity, specificity, or image quality metrics for specific pathologies). Instead, it relies on demonstrating substantial equivalence to a predicate device through technological comparison and non-clinical performance testing.

The "performance" reported primarily relates to technical specifications and the conclusion that the device's design outputs meet design input requirements and conform to intended user needs.

2. Sample Size Used for the Test Set and Data Provenance

This document does not describe a clinical test set with a specific sample size. The performance evaluation was primarily non-clinical, focusing on technological characteristics and conformity to standards.

Consequently, there is no information on:

• Sample size used for a test set.
• Country of origin of data.
• Whether data was retrospective or prospective.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

Not applicable, as no clinical test set requiring ground truth established by experts is described in this document.

4. Adjudication Method for the Test Set

Not applicable, as no clinical test set requiring adjudication is described.

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

Not applicable. This document is for a medical imaging device (CT system), not an AI diagnostic algorithm, and thus no MRMC study or AI assistance is mentioned.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) was done

Not applicable, as this is a CT imaging system, not a standalone AI algorithm. It produces images for interpretation by a human.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

Not applicable. For the non-clinical performance, "ground truth" would be engineering specifications and measurements against physical phantoms (e.g., for resolution, dose, etc.), or compliance with regulatory standards. The document mentions "Test phantoms: Exposure phantom, Constancy test phantom, Contrast element, and Ceph test phantom" (Page 5) which are used for technical assessments.

8. The Sample Size for the Training Set

Not applicable. This document describes a medical imaging device, not a machine learning model that requires a training set of data.

9. How the Ground Truth for the Training Set was Established

Not applicable, as there is no training set for a machine learning model described here.

Ask a specific question about this device

The X-ray system creates data for digital exposures in the maxillofacial area and in subareas for dentistry and pediatric dentistry, for hard-tissue diagnostics within ENT medicine, and carpus exposures.

The proposed Orthophos S is a dental cone-beam CT system (CBCT), which comprises sensor units for 2D cephalometric exposures, 2D panoramic radiograph and 3D volume exposure. The combination of sensors varies depending on the installed device configuration. In the proposed device, the 3D sensor, as well as the 2D cephalometric sensor, are identical to the corresponding sensors utilized in the predicate device (K150217). However, with respect to the 2D panoramic exposures, the proposed Orthophos S device utilizes a CMOS (scintillation) sensor whereas the predicate Orthophos SL (K150217) utilizes a CMOS (direct conversion) sensor.

Orthophos S uses an x-ray beam that rotates around the patient's head. Detectors acquire twodimensional x-ray images at varying radiographic angles. An enhanced software algorithm generates the 2D panoramic and 2D cephalometric images and the reconstruction algorithm reconstructs the 3D volumetric image from the raw image data.

The exposed area can be adapted to a specific region of interest to keep the radiation dose as low as possible for the patient. This is achieved by collimating the x-ray beam and the adjustment of starting and ending points of the sensor/x-ray source movement. Furthermore, the radiation dose can be adapted by various parameters such as program types and exposure technique factors. These functions are available for CBCT, cephalometric and panoramic exposures.

A Class I laser beam is utilized to define reference lines for the correct patient position. The patient is stabilized through different bite blocks and the motoric driven forehead and temple supports.

The obtained digital image data is processed to provide a reconstructed image to the operator/user. The reconstructed images are transferred to the currently marketed SIDEXIS 4 imaging processing software (K132773) and stored in the SIDEXIS 4 software database. The proposed Orthophos S device uses the same SIDEXIS 4 image processing software (K132773) as does the predicate device (K150217) and there are no changes to the SIDEXIS 4 software (K132773) introduced in this premarket notification.

The proposed Orthophos S device includes metal artifact reduction software feature which automatically reduces image artifacts caused by radiopaque objects. This identical software feature is also included in the predicate Orthophos SL (K150217) device and remains unchanged in the proposed device.

A user control panel allows user actions as e.g. height adjustment, selection of programs and exposure parameters and delivers information about the unit status.

For 3D imaging, the proposed Orthophos S allows the user to select volume sizes, within which multiple fields of view, anatomic positions, and collimation may be selected. This is intended to allow the clinician to select the field of view based on the diagnostic need and to minimize the dose exposed to the patient.

The provided document is a 510(k) summary for the Orthophos S device. It primarily focuses on demonstrating substantial equivalence to a predicate device (Orthophos SL, K150217) rather than providing detailed acceptance criteria and a study proving performance against those criteria as would be expected for a novel AI/ML device.

However, based on the information available, we can extract the relevant points regarding the device's performance, even if not presented as a formal study with specific acceptance criteria as you've requested for an AI/ML context. The document describes general performance verification without numerical acceptance criteria or detailed study methodologies typical of AI performance validation.

Here's an attempt to structure the information based on your request, acknowledging the limitations of the provided document for answering all specific points relevant to an AI/ML device.

Device Name: Orthophos S
Device Type: Dental Cone-Beam CT (CBCT) X-ray system

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state numerical acceptance criteria for image quality or specific performance metrics as one would find for an AI/ML model. Instead, it relies on demonstrating that the new device meets the design input requirements and performs equivalently to the predicate device.

Note on "Reported Device Performance": The document primarily states that testing was performed and demonstrated substantial equivalence and conformity to standards, rather than providing specific quantitative results for image quality comparison. This is typical for a 510(k) submission where the emphasis is on equivalence rather than detailed, comparative performance reporting against novel metrics.

Detailed Study Information (Based on the document's content, emphasizing what is not present for an AI/ML context)

The provided document describes the Orthophos S as a dental CT system. While it has an "enhanced software algorithm" for 2D panoramic and cephalometric images and a "reconstruction algorithm" for 3D volumetric images, and includes a "metal artifact reduction software feature," the
510(k) submission is for the imaging device itself, not specifically an AI/ML-driven diagnostic or interpretative software. Therefore, many of the questions structured for an AI/ML study (e.g., sample size for training, number of experts for ground truth, MRMC study, standalone performance) are not directly addressed in this type of submission.

Here's an analysis based on the provided text:

2. Sample sizes used for the test set and the data provenance:

   • The document states that "Functional testing as well as performance testing of the proposed device Orthophos S has been performed to verify the design output met the design input requirements and to validate that the device confirm to the intended user needs and the intended use as defined." and "Verification activities for confirmation of the image quality of the proposed device has been performed."
   • Specific sample sizes (e.g., number of patients/images in a test set, number of image acquisitions) are not provided in this 510(k) summary.
   • Data Provenance: Not specified. This type of performance testing typically involves in-house validation with phantoms and potentially healthy volunteers or cadaveric samples, but the document does not elaborate on the origin (country, retrospective/prospective) of any data used for testing.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Given that this is a 510(k) for an imaging device, not an AI diagnostic software, the concept of "ground truth established by experts" in the clinical sense (e.g., for disease detection) is not explicitly discussed.
   • The "image quality review" mentioned would typically involve internal qualified personnel or engineers comparing images from the new device against the predicate or established quality standards for diagnostic imaging. The number and qualifications of such "experts" are not specified.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

   • This is not applicable or specified. The testing described appears to be a technical verification of image quality parameters and adherence to standards rather than a clinical interpretation study requiring adjudication.

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 MRMC study was performed or reported. This device is an imaging system, and while it has "enhanced software algorithms," it's not a CAD (Computer-Aided Detection/Diagnosis) or interpretative AI tool that would assist human readers in a diagnostic task. The document specifically states: "Given the differences from the predicate device, no clinical data is necessary to support substantial equivalence." This confirms that comparative effectiveness clinical studies, like MRMC, were not conducted.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • Given that this is an imaging device, the concept of "standalone algorithm performance" (like an AI model reading an image) is not directly applicable in the way it would be for an AI diagnostic software. The device's "algorithms" are intrinsic to its function (image reconstruction, artifact reduction) and are part of the overall system performance. The performance tested is the device's ability to produce images of sufficient quality, not the performance of an AI model making a diagnosis.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

   • For an imaging device, "ground truth" for image quality is typically established through:
     • Physical phantoms: Objects with known properties used to measure spatial resolution, contrast, noise, etc.
     • Technical specifications: Comparing output (e.g., kV, mA, exposure time, FOV) against design specifications.
     • Comparison to predicate device images: Qualitative and quantitative comparison to images from a device already on the market (the predicate).
   • The document does not detail the specific "ground truth" methods used, but implies technical evaluation and comparison to the predicate, rather than biological or clinical ground truth.

8. The sample size for the training set:

   • Not applicable / Not provided. This document does not describe the development of a specific AI model that would have a training set. The "enhanced software algorithm" and "reconstruction algorithm" mentioned are likely deterministic or traditional image processing methods, not machine learning models requiring training data in the sense of AI/ML.

9. How the ground truth for the training set was established:

   • Not applicable. As no AI training set is described, no ground truth establishment for it is mentioned.

Ask a specific question about this device