Search Results

SMARTDent is intended for use as a software solution for managing dental diagnostic images and providing tools to support diagnosis.

SMARTDent provides the function to efficiently manage CT, panorama, cephalometric, intraoral sensor images and intraoral camera images acquired using X-ray imaging equipment of Ray Co., Ltd. and performs various image analysis according to the diagnostic purposes.

The provided document does not contain explicit acceptance criteria or details of a study proving the device meets those criteria. The 510(k) summary for SMARTDent mainly focuses on demonstrating substantial equivalence to a predicate device (CS Imaging) based on technological characteristics and states that performance testing was conducted according to FDA guidance. However, it does not provide specific performance metrics, acceptance criteria, or the results of such testing.

Therefore, many of the requested details cannot be extracted from this document.

Here's a breakdown of what can and cannot be answered based on the provided text:

1. A table of acceptance criteria and the reported device performance

This information is not available in the provided document. The 510(k) summary mentions "Performance Testing" was conducted, but it doesn't list specific acceptance criteria or the reported performance metrics of the SMARTDent device.

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

This information is not available in the provided document. While "Performance Testing" is mentioned, no details about the test set (sample size, data origin, retrospective/prospective nature) are given.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g., radiologist with 10 years of experience)

This information is not available in the provided document. The document states "Clinical testing is not a requirement and has not been performed," which implies that expert-established ground truth for a clinical test set was not part of the submission described.

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

This information is not available in the provided document. Due to the lack of details on clinical or performance testing involving expert review, there's no mention of an adjudication method.

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

This information is not available in the provided document. The document explicitly states, "Clinical testing is not a requirement and has not been performed," which means an MRMC study was not conducted for this submission. The device is described as "a software solution for managing dental diagnostic images and providing tools to support diagnosis" but the provided text doesn't detail any AI-assisted diagnostic features or studies on human reader improvement.

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

This information is not available in the provided document. The "Performance Testing" section is vague and doesn't specify if standalone algorithm performance was evaluated or what metrics were used.

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

This information is not available in the provided document. Given the statement "Clinical testing is not a requirement and has not been performed," specific ground truth types for clinical conditions are not discussed. The "Performance Testing" likely refers to software verification and validation, which would involve functional and technical performance against predefined specifications, rather than clinical ground truth derived from pathology or expert consensus on patient outcomes.

8. The sample size for the training set

This information is not available in the provided document. The document describes SMARTDent as a "medical image management and processing system" with "tools to support diagnosis." It does not mention any artificial intelligence or machine learning components that would typically require a training set. If such components exist, details about their training setup are not provided.

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

This information is not available in the provided document. As no training set is mentioned (or implied heavily), how its ground truth was established is also not present.

Ask a specific question about this device

RCT600 is CBCT and panoramic x-ray imaging system with cephalometric. Which is intended to radiographic examination of the dento-maxillofacial, sinus and TMJ structure for adult and pediatric patients. Cephalometric image is also includes wrist to obtain carpus images for growth and maturity assessment for orthodontic treatment. The device is to be operated and used by dentists or other legally qualified heath care professionals.

RCT600 provides 3D computed tomography for scanning hard tissues such as bone and teeth. By rotating the C-arm, which houses a high-voltage generator, an X-ray tube and a detector on each end, CBCT images of dental maxillofacial structures are obtained by recombining data scanned from the same level at different angles. Functionalities include panoramic image scanning for obtaining images of whole teeth, and a cephalometric option for obtaining cephalometric images.

The RCT600 device, for which K243903 is the 510(k) number, is a Computed Tomography X-Ray System used for radiographic examination of dento-maxillofacial, sinus, and TMJ structures. It also includes cephalometric imaging, specifically wrist images for growth and maturity assessment for orthodontic treatment.

The acceptance criteria are implicitly defined by the safety and effectiveness information provided in the 510(k) summary, specifically by demonstrating substantial equivalence to the predicate device, RCT700 (K213226). The testing performed aims to prove that despite some differences (deletion of one-shot Ceph option, different scan Ceph detector, changes in magnification, and scan times), the RCT600 maintains similar safety and effectiveness characteristics.

Here’s a breakdown of the information requested based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are based on demonstrating performance similar to the predicate device (RCT700, K213226) and meeting recognized international standards for medical electrical equipment and X-ray imaging devices.

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

• Sample Size for Test Set: The document mentions "Sample clinical images" were submitted for each imaging dental modality and that features were "clinically tested." However, it does not specify the number of cases or patients used in the clinical imaging test set.
• Data Provenance: The document does not explicitly state the country of origin of the data. It mentions "Clinical images were gathered from all detectors of the RCT600 system." The applicant is Ray Co., Ltd., based in SOUTH KOREA. Given the context, it is reasonable to infer the data either originated from or was managed by the manufacturer. The study is prospective, as it involved gathering clinical images with the new device.

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

• Number of Experts: "A licensed practitioner" (singular) reviewed the sample clinical images.
• Qualifications of Experts: The expert was an "independent, licensed practitioner/clinician DDS/ABOMR."
  • DDS stands for Doctor of Dental Surgery, indicating a dental professional.
  • ABOMR stands for American Board of Oral and Maxillofacial Radiology, indicating specialization in dental and maxillofacial radiology. No years of experience are specified.

4. Adjudication Method for the Test Set

The document states, "A licensed practitioner reviewed the sample clinical images and deemed them to be of acceptable quality for the device intended use." This indicates a single-reader review without an explicit adjudication process defined for multiple readers.

5. If a Multi-reader Multi-case (MRMC) Comparative Effectiveness Study was done

• No, an MRMC comparative effectiveness study was not done. The document describes a "clinical testing" where sample images were reviewed by a single licensed practitioner to confirm diagnostic quality. There is no mention of human readers improving with or without AI assistance, as this is an imaging device, not an AI diagnostic algorithm.

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

• This question is not directly applicable to the RCT600, as it is an X-ray imaging system, not an AI diagnostic algorithm. The "performance" being evaluated implicitly includes the algorithm within the imaging device that processes and reconstructs the images. The focus is on the imaging system's ability to produce diagnostically acceptable images for human interpretation. The clinical evaluation verifies the diagnostic quality of the images produced by the device, which is an assessment of its standalone imaging performance.

7. The Type of Ground Truth Used

The ground truth for the clinical image review was based on expert consensus (from the singular licensed practitioner) regarding the "acceptable quality for the device intended use" of the sample clinical images.

8. The Sample Size for the Training Set

The document does not specify a sample size for a training set. The device being described is an X-ray imaging system, not a machine learning algorithm that typically requires a separate training set. The "software" component mentioned underwent validation, but no details of training data for the software's inherent functions (e.g., image reconstruction algorithms) are provided, as these are typically validated through engineering and bench testing rather than a separate clinical "training set" in the context of device clearance.

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

As no explicit training set for an AI algorithm is mentioned, the concept of ground truth establishment for a training set is not applicable in the context of this device's submission description. The software validation relies on risk analysis and compliance with FDA guidance documents rather than a specific clinical training dataset and ground truth for learning.

Ask a specific question about this device

RAYDENT SW is a software designed to assist dental professionals in planning patient treatment devices. The software performs simulations based on patient images, allowing reference to treatment plans, and is used as a tool to design treatment devices based on 3D mesh data. Treatment devices include prosthetic devices (Veneer, Crown, Bridge, In/Onlay) and orthodontic devices (Clear Aligner).

To use RAYDENT SW, users must have the necessary education and domain knowledge in orthodontic practice and receive dedicated training in the use of the software.

RAYDENT is a software that provides tools to simulate treatment plans based on patient images generated by compatible scanners and design treatment devices based on appropriate three-dimensional images. It allows dental offices to acquire patient data in conjunction with software on compatible imaging equipment and utilize the acquired images to create treatment plans and devices for skilled dentists and oral and maxillofacial specialists.

The document K233625 is a 510(k) Premarket Notification for the device "RAYDENT SW," a software designed to assist dental professionals in planning patient treatment devices. As such, the document provides information on the device's intended use, comparison to predicate devices, and a summary of performance testing. However, it does NOT include detailed information about acceptance criteria or a specific study proving the device meets those criteria, particularly not in the context of an AI/ML-enabled medical device performance study.

The document states that RAYDENT SW includes "Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices: YES" in its comparison table (Page 7). However, the "Performance Testing" section (Page 9) does not describe an AI/ML-specific performance study with acceptance criteria, sample sizes, ground truth establishment, or human-in-the-loop evaluation. It merely states that "Software, hardware, and integration and validation testing was performed in accordance with the FDA Guidance Document 'Guidance for the Content of Premarket Submissions for Device Software Functions' and 'Guidance for the Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions'." It then concludes that "All test results have been reviewed and approved, showing the RAYDENT SW to be substantially equivalent to the predicate devices."

Therefore, based on the provided text, I cannot extract the information required to answer your prompt about the acceptance criteria and a study proving the device meets those criteria in the context of AI/ML performance.

To answer your specific points:

1. A table of acceptance criteria and the reported device performance: Not found in the provided document. The document mentions general validation testing but no specific performance metrics or acceptance criteria for an AI component.
2. Sample sized used for the test set and the data provenance: Not found.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not found.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not found.
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 found. The document explicitly states "Clinical testing is not a requirement and has not been performed" (Page 9), implying no such MRMC study was conducted.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not explicitly detailed for an AI component. The general performance testing is mentioned, but without specifics for the AI.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc): Not found.
8. The sample size for the training set: Not found.
9. How the ground truth for the training set was established: Not found.

The document focuses on substantiating equivalence primarily through comparison of indications for use, technological characteristics, and general software/hardware validation, rather than an in-depth AI/ML performance study.

Ask a specific question about this device

The RAYSCAN a- P, SC, OCL, OCS panoramic X-ray imaging system with Cephalostat is an extra-oral source X-ray system, intended for dental radiographic examination of the teeth, jaw, and oral structures, to include panoramic examinations and implantology and for TMJ studies and cephalometry. Images are obtained using the standard narrow beam technique.

RAYSCAN α-Expert (RAYSCAN α-P, SC, OCL, OCS) provides panoramic for scanning teeth, jaw and oral structures. By rotating the C-arm, which houses a high-voltage generator, an all-in-one Xray tube and a detector on each end, panoramic images of oral and maxillofacial structures are obtained byrecombining data scanned from different angles. Functionalities include panoramic image scanning for obtaining images of whole teeth, and a Cephalometric scanning option for obtaining Cephalic images.

The provided text describes a 510(k) premarket notification for the "RAYSAN α-Expert" dental X-ray system. The submission affirms its substantial equivalence to a predicate device, K142058. While it outlines several tests conducted to support this claim, it does not provide explicit acceptance criteria in a table format nor does it detail a specific study with quantitative performance metrics for a direct comparison against such criteria.

Here's a breakdown of the information that can be extracted, and where there are gaps regarding the requested specifics:

1. A table of acceptance criteria and the reported device performance

The document does not provide a table of acceptance criteria with corresponding device performance metrics. Instead, it states that "All test results were satisfactory" for performance (imaging performance) testing conducted according to IEC 61223-3-4. It also mentions that "a licensed practitioner reviewed the sample clinical images and deemed them to be of acceptable quality for the intended use." This indicates a subjective assessment of image quality rather than quantitative performance against defined acceptance criteria.

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Test Set Sample Size: The document mentions that "images were gathered from all detectors of RAYSCAN α-Expert using protocols with random patient age, gender, and size" and that "Clinical imaging samples were collected from new detectors on the proposed device at the two offices where the predicate device was installed for the clinical test images." However, it does not specify the exact number of images or patients in the clinical test set.
• Data Provenance: The images were collected "at the two offices where the predicate device was installed for the clinical test images." The manufacturer is Ray Co., Ltd. located in South Korea. It's implied these are prospective clinical images gathered for the purpose of the submission.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

• Number of Experts: "The clinical performance of RAYSCAN α-Expert were clinically tested and approved by two licensed practitioners/clinicians."
• Qualifications of Experts: They are described as "licensed practitioners/clinicians." No specific details such as years of experience, specialization (e.g., radiologist, dentist), or board certification are provided.

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

The document states, "A licensed practitioner reviewed the sample clinical images and deemed them to be of acceptable quality for the intended use." It implies individual review, but does not specify any formal adjudication method (e.g., whether the two practitioners independently reviewed images and consensus was reached, or if there was a third adjudicator in case of disagreement).

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

• MRMC Study: No MRMC comparative effectiveness study is mentioned. This device is an X-ray imaging system, not an AI-assisted diagnostic tool for humans, so this type of study would not be applicable. The comparison is between the new device's image quality and the image quality of the predicate device.
• Effect Size: Not applicable.

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

This refers to an X-ray imaging device, not an algorithm. Therefore, "standalone (algorithm only)" performance is not applicable. The device's primary function is image acquisition.

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

The ground truth for the clinical image quality assessment appears to be expert opinion/consensus (from two licensed practitioners) regarding whether the images were "of acceptable quality for the intended use." There's no mention of pathology or outcomes data for establishing ground truth.

8. The sample size for the training set

The document mentions software validation, but this X-ray system is not described as an AI/ML device that requires a distinct "training set" in the context of machine learning model development. This question is not directly applicable to the type of device described.

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

As the device is not described as involving an AI/ML model with a training set, this question is not directly applicable. The software mentioned is for saving patient and image data, inquiries, and image generation, and was validated according to FDA guidance for software in medical devices, not specific AI/ML training.

Summary of what is present and what is missing:

• Acceptance Criteria/Performance Table: Not provided in the requested format. General statement of "satisfactory" test results and "acceptable quality."
• Test Set Sample Size & Provenance: Sample size not quantified. Provenance is South Korea, likely prospective.
• Number & Qualification of Experts: Two licensed practitioners/clinicians. No further qualification details.
• Adjudication Method: Not specified.
• MRMC Study: Not applicable.
• Standalone Performance: Not applicable.
• Type of Ground Truth: Expert opinion on image quality.
• Training Set Sample Size: Not applicable (not an AI/ML device in this context).
• Training Set Ground Truth: Not applicable.

Ask a specific question about this device

RAYSCAN a-Expert3D, panoramic x-ray imaging system with cephalostat, is an extra-oral source x-ray system, which is intended for dental radiographic examination of the teeth, jaw, and oral structures, specifically for panoramic examinations and implantology and for TMJ studies and cephalometry, and it has the capability, using the CBCT technique, to generate dentomaxillofacial 3D images.

RAYSCAN α-3D, SM3D, M3DS and M3DL are 3D computed tomography for scanning hard tissues like bone and teeth. By rotating the C-arm, which houses a high-voltage generator, a X-ray tube and a detector on each end, CBCT images of dental maxillofacial structures are obtained by recombining data scanned from the same level at different angles. Functionalities include panoramic image option and cephalometric option.

The provided text describes a 510(k) premarket notification for the RAYSCAN a-Expert3D, a dental X-ray imaging system. The document focuses on demonstrating substantial equivalence to a predicate device, rather than proving that the device meets specific acceptance criteria through a comprehensive clinical study.

Therefore, the requested information regarding detailed acceptance criteria, sample sizes, expert qualifications, and specific study designs (MRMC, standalone performance) is largely not present in the provided text. The document primarily highlights non-clinical bench testing and the provision of clinical image samples for review by licensed practitioners to further support substantial equivalence.

Based on the available information, here's what can be extracted and what is missing:

Overview of Device Performance and Study Information

The submission for the RAYSCAN a-Expert3D is a 510(k) for substantial equivalence to a predicate device (RAYSCAN α-Expert3D K190812 and RCT800 K230753). The performance assessment primarily relies on demonstrating that the modified device (with updated X-ray voltage/current and detector types) maintains similar safety and effectiveness compared to the predicate, as supported by non-clinical and limited clinical data.

1. Table of Acceptance Criteria and Reported Device Performance

The document does not present explicit quantitative acceptance criteria for device performance, such as specific accuracy, sensitivity, or specificity thresholds. Instead, it states that "All test results were satisfactory" for bench testing. The primary "acceptance criterion" implied throughout the 510(k) process is demonstrating substantial equivalence to the predicate device.

2. Sample size(s) used for the test set and the data provenance

• Test Set Sample Size: The document states that "Clinical imaging samples were collected from new detectors on the proposed device at the two offices where the predicate device was installed for the clinical test images." It also mentions "images were gathered from all detectors installed with RAYSCAN a-Expert3D using protocols with random patient age, gender, and size." However, no specific numerical sample size (e.g., number of patients or images) for the clinical test set is provided.
• Data Provenance:
  • Country of Origin: Not explicitly stated for the clinical data. The manufacturer is in South Korea.
  • Retrospective or Prospective: Not explicitly stated. The phrasing "Clinical imaging samples were collected from new detectors on the proposed device" could suggest prospective collection for the purpose of this submission, but it's not definitive.

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

• Number of Experts: "two licensed practitioners/clinicians."
• Qualifications of Experts: "A licensed practitioner reviewed the sample clinical images and deemed them to be of acceptable quality for the intended use." Specific specialties (e.g., radiologist, dentist with specific experience) or years of experience are not provided beyond "licensed practitioner."

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

• "A licensed practitioner reviewed the sample clinical images." This suggests an individual review, potentially without formal adjudication unless the "two licensed practitioners" independently reviewed and concurred, which is not detailed. No specific adjudication method (e.g., consensus, majority vote) is mentioned. The primary assessment seems to be a qualitative review for "acceptable quality for the intended use."

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 "substantial equivalence" submission for an imaging device, not an AI-assisted diagnostic tool. The purpose was to show the new device produces images comparable to the predicate for diagnostic use. No AI component is mentioned, and therefore no assessment of human reader improvement with AI assistance.

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

• Not applicable. This device is an X-ray imaging system, not a diagnostic algorithm. Its performance is related to image acquisition parameters and image quality, not an output from an algorithm in the typical sense of standalone AI.

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

• The "ground truth" for the clinical images appears to be the qualitative assessment of "acceptable quality for the intended use" by licensed practitioners. It is not based on pathology, outcomes data, or a formal expert consensus process as would be typically seen for a diagnostic performance study. The images serve to "show that the complete system works as intended."

8. The sample size for the training set

• Not applicable. This is an X-ray imaging system, not a machine learning model, so there is no "training set." The software validation refers to standard software development practices, not AI model training.

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

• Not applicable, as there is no training set for an AI model.

In summary, the provided document focuses on demonstrating that the updated RAYSCAN a-Expert3D device is substantially equivalent to previously cleared predicate devices, primarily through non-clinical bench testing and a limited qualitative review of clinical images by licensed practitioners. It does not contain the detailed, quantitative clinical study data (such as MRMC, standalone algorithm performance, or specific metrics with acceptance thresholds) typically associated with AI/CADe device submissions.

Ask a specific question about this device

RCT800 is CBCT and panoramic x-ray imaging system with cephalometric. Which is intended to radiographic examination of the dento-maxillofacial, sinus, TMJ, Airway for diagnostic support for adult and pediatric patients. And a model scan is included as an option. Cephalometric image is also includes wrist to obtain carpus images for growth and maturity assessment for orthodontic treatment. The device is to be operated and used by dentists or other legally qualified health care professionals.

The system's purpose is RCT800, a 3D computed tomography scanner for scanning hard tissues like bones and teeth. By rotating the c-arm, which is embedded with an all-in-one x-ray tube and a detector on each end, CBCT images of the dental maxillofacial area can be attained by recombining data from the same level that is scanned from different angles. Additionally, the system includes a panoramic image scanning function for attaining images of the whole teeth, a cephalometric scanning option for attaining a cephalic image, and a Model Scan option for attaining a dental model CBCT image.

The provided text describes the 510(k) summary for the Ray Co., Ltd.'s RCT800 device. While it mentions various tests and compliance with standards, it does not explicitly provide a table of acceptance criteria and reported device performance in the typical format of a clinical study report with specific metrics and thresholds. Instead, it states that "All test results were satisfactory" for performance (imaging performance) testing conducted according to IEC 61223-3-4 and IEC 61223-3-7.

It also mentions "Clinical images were provided" and that "A licensed practitioner reviewed the sample clinical images and deemed them to be of acceptable quality for the intended use," but it doesn't detail a formal clinical study with specific acceptance criteria beyond subjective expert opinion.

Therefore, much of the requested information regarding the study that proves the device meets the acceptance criteria in a quantifiable manner (e.g., statistical significance, specific performance numbers) is not present in the provided document. The document primarily focuses on demonstrating substantial equivalence to a predicate device through technical similarities, bench testing, and compliance with general safety and performance standards.

However, based on the provided text, I can infer and extract some information:

Acceptance Criteria and Device Performance (Inferred from compliance statements)

Since the document states that "All test results were satisfactory" for imaging performance tests and that a licensed practitioner found clinical images to be of "acceptable quality for the intended use," the implied acceptance criteria were met. However, the specific quantitative criteria are not listed.

Table 1: Implied Acceptance Criteria and Reported Device Performance

Study Details (Based on the provided text)

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

   • Test Set (Clinical Images): The document states "Clinical imaging samples were collected from new detectors on the proposed device at the two offices where the predicate device was installed for the clinical test images. These images were gathered from all detectors installed with RCT800 using protocols with random patient age, gender, and size."
     • Specific Number: Not specified. It indicates "samples" and "images gathered from all detectors," implying a collection across multiple patient demographics, but no numerical count is provided.
     • Provenance: Clinical images obtained from "two offices where the predicate device was installed." The country of origin is not explicitly stated for these clinical images, but the applicant (Ray Co., Ltd.) is from SOUTH KOREA, so it is plausible these were collected within South Korea or countries where the predicate device was installed.
     • Retrospective/Prospective: Not explicitly stated, but the mention of collecting "new detectors" and "random patient age, gender, and size" suggests some form of prospective or concurrent collection for evaluation.

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

   • Number of Experts: "Two licensed practitioners/clinicians" clinically tested and approved the features. "A licensed practitioner reviewed the sample clinical images." It's unclear if these refer to the same "two" or one specific "licensed practitioner."
   • Qualifications: "Licensed practitioners/clinicians." No further details on their years of experience, subspecialty (e.g., specific type of radiologist/dentist), or formal board certifications are provided.

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

   • The document states "A licensed practitioner reviewed the sample clinical images and deemed them to be of acceptable quality for the intended use." This suggests a qualitative assessment, but no formal adjudication method (like 2+1 consensus) is described. It implies that the single or (if two practitioners were involved) potentially un-adjudicated opinion of the expert(s) served as the "truth."

4. 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 mentioned. The document focuses on demonstrating that the device itself produces images of acceptable quality for human interpretation, not on the improvement of human readers with AI assistance. This device is a source of images (an X-ray system), not an AI algorithm for image analysis.

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

   • The device is a medical imaging system (CT/panoramic/cephalometric X-ray system), not an AI algorithm for image analysis. Therefore, the concept of "standalone performance" for an algorithm does not apply in the context of this device. The performance refers to the image acquisition capabilities of the system itself.

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

   • For clinical image quality, the ground truth was expert opinion/assessment by "licensed practitioners/clinicians" who deemed the images "of acceptable quality for the intended use."
   • For bench testing, the ground truth was based on compliance with standards (e.g., IEC 61223-3-4, IEC 61223-3-7) and "designated tolerance" parameters.

7. The sample size for the training set:

   • This device is an X-ray imaging system. It takes images (presumably raw data) and reconstructs them into clinically useful images using reconstruction algorithms such as FBP (Filtered Back Projection) (as mentioned for the predicate device). The document does not describe the use of machine learning that would require a "training set" in the conventional AI sense. If any internal algorithms are adaptable or "learn," details are not provided. The phrase "training set" is typically applicable to AI/ML devices, which this is not identified as in the provided text.

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

   • Not applicable, as no training set (for AI/ML) is mentioned or implied for this device.

Ask a specific question about this device

When properly worn, the Ear-Friendly Masks are intended to protect and healthcare workers from transfer of microorganisms, body fluids and particulate material. This device is non sterile and for single use only.

The Ear-Friendly Masks are non-sterile, single use, 3 layers, flat-pleated style with ear loops and nose piece. The outer layer and inner facing layer of face mask consist of Spunbond Polypropylene, and the middle layer consists of Melt Blown Polypropylene Filter. Each mask contains ear loops to secure the mask over the user's face and mouth with nose piece to firmly fit over the nose. This device is not made from any natural rubber latex.

Here's a breakdown of the acceptance criteria and study that proves the device meets them, based on the provided FDA 510(k) Summary for the "Ear-Friendly Mask."

Device Description:
The Ear-Friendly Mask is a non-sterile, single-use, 3-layer, flat-pleated surgical face mask with ear loops and a nose piece. It is intended to protect both patients and healthcare workers from the transfer of microorganisms, body fluids, and particulate material.

1. Acceptance Criteria and Reported Device Performance

The acceptance criteria for the Ear-Friendly Mask are based on the ASTM F2100 standard for the performance of materials used in surgical face masks, specifically for Level 3 performance. The table below outlines the acceptance criteria and the device's reported performance:

The document states, "All Result of testing met ASTM F2100 Level 3 acceptance Criteria."

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

The document does not explicitly state the specific sample sizes used for each of the performance tests (Fluid Resistance, PFE, BFE, Differential Pressure, Flammability, Biocompatibility).

Data Provenance: The studies were non-clinical (bench testing and biocompatibility), conducted by the manufacturer, RAY Co., Ltd., based in Anyang-si, Gyeonggi-do, Korea. The provenance is therefore prospective industrial testing aimed at regulatory submission.

3. Number of Experts Used to Establish Ground Truth and Qualifications

• Not applicable (N/A). This submission pertains to a physical medical device (surgical face mask) and its performance characteristics (filtration, fluid resistance, etc.), along with its biocompatibility. The "ground truth" for these tests is established through standardized laboratory methods (e.g., ASTM F tests, ISO 10993). There is no mention or requirement for human experts (e.g., radiologists) to establish ground truth as would be the case for an AI/CADe device interpreting medical images.

4. Adjudication Method for the Test Set

• Not applicable (N/A). As these are objective, standardized physical and chemical tests, there is no need for an adjudication method in the way one would adjudicate human expert interpretations in a clinical study. The tests yield quantitative results that are compared directly against the predefined numerical acceptance criteria.

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

• No, an MRMC comparative effectiveness study was not done. This device is a physical product (surgical face mask), not an AI-assisted diagnostic or imaging device. Therefore, studies involving human readers or their improvement with AI assistance are not relevant.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study Was Done

• Not applicable (N/A). This device is a physical medical product, not an algorithm or software device. The concept of "standalone performance" for an algorithm is not relevant here. The tests performed are purely bench and biocompatibility tests on the physical mask itself.

7. The Type of Ground Truth Used

The ground truth for this device's performance relies on objective, standardized laboratory test results as defined by:

• Consensus Standards: ASTM F2100 (for mask performance) and ISO 10993 (for biocompatibility).
• Specific Test Methods: ASTM F1862 (Fluid Resistance), ASTM F2299 (PFE), ASTM F2101 (BFE), MIL-M-36954C (Differential Pressure), 16 CFR part 1610 (Flammability), ISO 10993-5 (Cytotoxicity), ISO 10993-10 (Sensitization/Irritation).

The "ground truth" is whether the quantitative results obtained from these tests meet the specified numerical acceptance criteria for a Level 3 surgical mask.

8. The Sample Size for the Training Set

• Not applicable (N/A). This device is not an AI/ML product developed using training datasets. The 'training set' concept does not apply to the manufacturing and testing of a physical product like a surgical face mask.

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

• Not applicable (N/A). As there is no training set for an AI/ML model, there is no ground truth to be established for it. The product's compliance is demonstrated through a series of physical and chemical performance tests.

Ask a specific question about this device

RCT800 is CBCT and panoramic x-ray imaging system with cephalometric. Which is intended to radiographic examination of the dento-maxillofacial, sinus, TMJ, Airway for diagnostic support for adult and pediatric patients. And a model scan is included as an option. Cephalometric image is also includes wrist to obtain carpus images for growth and maturity assessment for orthodontic treatment. The device is to be operated and used by dentists or other legally qualified heath care professionals.

RCT800 is 3D computed tomography for scanning hard tissues like bone and teeth. By rotating the c-arm which is embedded with high voltage generator all-in-one x-ray tube and a detector on each end, CBCT images of dental maxillofacial is attained by recombining data from the same level that are scanned from different angle. Panoramic image scanning function for attaining image of whole teeth, cephalometric scanning option for attaining cephalic image, and Model Scan option for attaining dental model CBCT image are included.

The provided text describes the regulatory clearance of a medical device (RCT800), focusing on its substantial equivalence to predicate devices rather than a detailed study proving it meets specific acceptance criteria in the context of AI performance. Therefore, the information needed to fully answer the request, particularly regarding AI performance, ground truth establishment, expert qualifications, and sample sizes for training/test sets in an AI context, is largely absent.

It appears the RCT800 is a dental X-ray system (CBCT, panoramic, cephalometric) and not explicitly an AI-powered diagnostic device from the provided documentation. The "software" mentioned is for image generation, patient data management, and inquiry, with a moderate level of concern, suggesting it functions as control and viewing software rather than an AI-driven interpretation system.

However, I can extract and infer some information based on the request and the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Since this is a 510(k) submission focused on substantial equivalence for an imaging device, the "acceptance criteria" are generally aligned with demonstrating that the new device performs as safely and effectively as legally marketed predicate devices, and meets relevant performance standards.

2. Sample size used for the test set and the data provenance

• Test Set Sample Size:
  • Clinical Images: "random patient age, gender, and size" were used to gather clinical images from the new detector installed with RCT800. The number of patients or images is not specified.
  • Non-Clinical Performance: Test results for the new detectors (FXDD-0606CA, 1717SCC) were compared to predicates. This likely involved quantitative testing on phantoms or test objects, not patient data in the same sense as clinical images. The sample size for these non-clinical tests is not specified, but typically involves a defined set of measurements.
• Data Provenance: The new detector was installed at "2 offices where the predicate device is installed." This suggests prospective data collection (images gathered from newly installed RCT800 units in a clinical setting). The country of origin is not explicitly stated for the clinical data, but the manufacturer is in the Republic of Korea.

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

• Number of Experts: "two licensed practitioners/clinicians."
• Qualifications of Experts: "licensed practitioners/clinicians." No specific specialization (e.g., radiologist, dentist) or years of experience are provided, but the context of "dental X-ray system" strongly implies dental professionals.

4. Adjudication method for the test set

• The text states: "As licensed practitioners or clinician diagnoses of the images, it might be proved that the clinical diagnosis and structures are acceptable in the region of interests." This suggests an assessment of diagnostic acceptability by the two practitioners. An explicit "adjudication method" (like 2+1 or 3+1) is not detailed; it's more of a verification of intended function rather than a formal ground truth consensus process for diagnostic accuracy.

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 involving AI assistance was not described. The submission is for an imaging device itself, not an AI-powered diagnostic aid. The device helps acquire images; it does not independently interpret them or assist human readers in interpretation beyond providing the images.

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

• No, a standalone algorithm performance study was not described. This device is an X-ray system; it requires a human operator to acquire images and human practitioners to interpret them. The "software" mentioned supports device operation and image viewing, not autonomous diagnostic interpretation.

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

• For the clinical observations, the "clinical diagnosis and structures are acceptable in the region of interests" by "licensed practitioners or clinician diagnoses." This suggests expert diagnostic opinion as the basis for evaluating image acceptability. It does not refer to pathology or outcomes data.
• For non-clinical performance (e.g., image quality, dose), the ground truth would be physical measurements against established standards (IEC norms) or direct comparison to the physical properties/output of the predicate devices.

8. The sample size for the training set

• Not applicable / Not provided. The RCT800 is an X-ray imaging system, not an AI model requiring a training set in the conventional sense. The software mentioned is for control, patient data management, and image generation, not for learning from a training set to make diagnoses.

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

• Not applicable. As a non-AI imaging device, there is no "training set" or "ground truth for the training set" in the context of machine learning model development.

Ask a specific question about this device