LogoFDA 510(k) Search
Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K230144
    Device Name
    Denti.AI Detect
    Manufacturer
    Denti.AI Technology, Inc.
    Date Cleared
    2023-10-06

    (261 days)

    Product Code
    MYN
    Regulation Number
    892.2070
    Type
    Traditional
    Panel
    Radiology
    Reference & Predicate Devices
    K210365,K210187
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    Denti. Al Detect is a Computer-Assisted Detection (CADe) software device intended to be used by dental professionals. comprising dentists and dental specialists, while reading extraoral and intraoral 2D dental radiographs. The device aims to assist in detecting and highlighting uncategorized regions of interest (ROIs) within the teeth area, which include cares and periapical radiolucency, as a second reader. The device is also intended to aid in the measurements of mesial and distal bone levels associated with each tooth.

    The device is aimed to be used with images from the patients of 22 years age and older without remaining primary dentition. The device is not intended to replace a complete clinical judgment that considers other relevant information from the image or patient history.

    Device Description

    Denti.AI Detect is a prescription use Computer-Assisted Detection (CADe) device aimed to assist dentists and dental specialists in detecting and highlighting regions of interest (ROIs) within the teeth area, which include caries and periapical radiolucency. The device is also intended to aid in the measurements of medial and distal bone levels associated with each tooth.

    AI/ML Overview

    Denti.AI Detect - Acceptance Criteria and Study Details

    1. Acceptance Criteria and Reported Device Performance

    For Detection of Caries and Periapical Radiolucency (Stand-alone Study):

    EndpointAcceptance CriteriaReported Device Performance (95% CI)
    Across-category wAFROC AUCLower bound of 95% CI > 0.60.737 (0.713, 0.761)

    For Bone Level Measurement (Stand-alone Study):

    EndpointAcceptance CriteriaReported Device Performance (95% CI)
    Bitewing: CEJ-Bone: SensitivityNot explicitly defined, but implied to be high from results98.1% (96%, 99.5%)
    Bitewing: CEJ-Bone: SpecificityNot explicitly defined, but implied to be high from results93% (88.7%, 96.7%)
    Bitewing: CEJ-Bone: Mean Absolute Error (MAE)Not explicitly defined, but implied to be low from results0.513 mm (0.444 mm, 0.593 mm)
    Bitewing: CEJ-Bone/CEJ-Root Ratio: MAENot explicitly defined, but implied to be low from results3.8% (3.3%, 4.4%)
    Periapical: CEJ-Bone: SensitivityNot explicitly defined, but implied to be high from results98.2% (96.2%, 99.7%)
    Periapical: CEJ-Bone: SpecificityAcceptance Criterion was met with the exception of the lower CI bound being slightly less than the Acceptance Criterion.88.5% (82.8%, 93.4%)
    Periapical: CEJ-Bone: MAENot explicitly defined, but implied to be low from results0.572 mm (0.497 mm, 0.653 mm)
    Periapical: CEJ-Root: SensitivityNot explicitly defined, but implied to be high from results96.9% (94.2%, 99%)
    Periapical: CEJ-Root: SpecificityNot explicitly defined, but implied to be high from results92.2% (87.9%, 96%)
    Periapical: CEJ-Root: MAENot explicitly defined, but implied to be low from results0.735 mm (0.612 mm, 0.868 mm)
    Periapical: CEJ-Bone/CEJ-Root Ratio: MAENot explicitly defined, but implied to be low from results4.3% (3.7%, 4.9%)
    Extraoral: CEJ-Bone-Root: SensitivityNot explicitly defined, but implied to be high from results91.6% (89.6%, 93.6%)
    Extraoral: CEJ-Bone-Root: SpecificityNot explicitly defined, but implied to be high from results84.3% (70.8%, 95.8%)
    Extraoral: CEJ-Bone/CEJ-Root Ratio: MAENot explicitly defined, but implied to be low from results4.7% (4%, 5.5%)

    For Reader Study (Multi-Reader Multi-Case):

    EndpointAcceptance CriteriaReported Device Performance (95% CI)
    Across-category reader performance (wAFROC)Statistically significant improvement with AI assistance (p < 0.05)Aided wAFROC: 0.809 Unaided wAFROC: 0.784 Difference: 0.025 p-value: 0.029

    2. Sample Sizes and Data Provenance

    Detection of Caries and Periapical Radiolucency (Stand-alone Study Test Set):

    • Sample Size: 709 images
    • Data Provenance: Obtained from 6 sites in the US. Retrospective (implied, as the study involves established ground truth and does not mention prospective data collection).
    • Patient Demographics: Aged 22 to 92, roughly split between males and females.

    Bone Level Measurement (Stand-alone Study Test Set):

    • Sample Size: 193 images
    • Data Provenance: Obtained from 9 sites in the US. Retrospective (implied).
    • Patient Demographics: Aged 22 to 88, roughly split between males and females.

    Multi-Reader Multi-Case (MRMC) Study Test Set:

    • Sample Size: 154 images
    • Data Provenance: Taken from 5 sites across the US. Retrospective (implied).

    3. Number of Experts and Qualifications for Test Set Ground Truth

    Detection of Caries and Periapical Radiolucency (Stand-alone Study):

    • Number of Experts: 3
    • Qualifications: "Independent dentists" (specific experience level not provided).

    Bone Level Measurement (Stand-alone Study):

    • Number of Experts: 3
    • Qualifications: "Independent dentists" (specific experience level not provided).

    Multi-Reader Multi-Case (MRMC) Study:

    • The document implies that the ground truth for this study was established prior to the readers evaluating the images, likely by experts, although the explicit number and qualifications for this specific study's ground truth are not detailed.

    4. Adjudication Method for the Test Set

    Detection of Caries and Periapical Radiolucency (Stand-alone Study):

    • Adjudication Method: Consensus rule applied to establish final reference standard. (This is typically considered a (3:0) or 2+1 if 2 agree and a third resolves discrepancy method).

    Bone Level Measurement (Stand-alone Study):

    • Adjudication Method: Majority rule applied to establish final reference standard. (This is a (2:1) or (3:0) method).

    5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

    • Yes, a fully crossed Multi-Reader Multi-Case (MRMC) study was conducted.
    • Effect Size of Human Readers' Improvement with AI vs. without AI Assistance:
      • The study demonstrated a statistically significant improvement in across-category reader performance when assisted by Denti.AI Detect as compared to when unaided.
      • Aided wAFROC: 0.809
      • Unaided wAFROC: 0.784
      • Difference (effect size): 0.025
      • P-value: 0.029

    6. Standalone (Algorithm Only) Performance Study

    • Yes, standalone studies were conducted to assess the performance of the Denti.AI Detect system for:
      1. Detecting suspicious ROIs (caries and periapical radiolucencies).
      2. Making measurements of distal and mesial bone levels associated with each tooth.

    7. Type of Ground Truth Used

    • Expert Consensus: For both the detection of caries and periapical radiolucency and the bone level measurement studies, ground truth was established by independent dentists using consensus or majority rule. This is a form of expert consensus.

    8. Sample Size for the Training Set

    • The document does not explicitly state the sample size for the training set. It only describes the test sets used for validation.

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

    • The document does not explicitly state how the ground truth for the training set was established. It only details the ground truth establishment for the test sets used in the performance evaluation studies.
    Ask a Question

    Ask a specific question about this device

    Page 1 of 1