(94 days)
3D Endo™ Software is intended to aid in the visualization, diagnosis and planning of endodontic treatment and retreatment cases utilizing DICOM images.
The proposed 3D Endo™ Software is image processing software for simulating 3D images using Digital Imaging and Communication in Medicine (DICOM) images of the respective tooth and the surrounding tissue. The DICOM images used by the proposed 3D Endo™ Software are imported from those generated by Cone Beam Computed Tomography (CBCT) scanners. The proposed 3D Endo™ Software has no direct control or interface with the CBCT Scanner. This imaging is used to provide a means for pre-operative planning for endodontic root canal procedures. The proposed 3D Endo™ Software is stand-alone software, with no physical component. It is downloaded and locally installed on the end user's computer system, and updated automatically as required through an active internet connection. The proposed 3D Endo™ Software allows a user to create a 3D image from DICOM images using a 5 step process. The user can diagnose their case, visually isolate the tooth of interest and surrounding soft and osseous tissue, investigate the canal system, understand the 3D canal anatomy, and create a treatment plan by following the steps laid out in the software's wizard. The proposed 3D Endo™ Software functions by processing 2D images received in DICOM format from a CBCT scanning device, and processing multiple images into a virtual 3D image. The user can measure, assess, and visualize the anatomy of the root canal structure, both with and without simulated instruments, allowing the practitioner to plan the endodontic procedure.
The provided document, a 510(k) Premarket Notification for the 3D Endo™ Software, does not contain specific information about quantitative acceptance criteria for device performance or a detailed study proving the device meets these criteria in the context of diagnostic accuracy (e.g., sensitivity, specificity, AUC).
The FDA's review focuses on substantial equivalence to a predicate device (SimPlant 2011), primarily through comparison of intended use, technology, and functionalities, alongside verification and validation of the software itself. The document explicitly states: "No human clinical data is included in this premarket notification to support substantial equivalence."
Therefore, I cannot provide a table of acceptance criteria and reported device performance from this document, nor can I elaborate on the specifics of a study proving diagnostic performance, as such a study (with the mentioned elements like sample size, ground truth, expert opinions, MRMC studies) was not included or required for this particular 510(k) submission.
The "study" described in the document is primarily software verification and validation (V&V) and usability testing.
Here's what can be extracted based on the provided text regarding device "acceptance":
1. A table of acceptance criteria and the reported device performance:
Based on the document, the "performance" discussed is primarily related to software functionality and usability, not diagnostic accuracy metrics.
| Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|
| Compliance with software life cycle processes as per IEC 62304 | Satisfactorily met requirements; demonstrated compliance. |
| Functionality and compatibility of all system components | Verified through software unit, integration/functional, and regression testing. |
| Ease of use and mitigation of potential human usage errors (Usability) | No results demonstrating the presence of new or unexpected use errors which present a serious residual use-related risk to the user or patient. (Per IEC 62366-1:2015 and FDA guidance) |
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: Not specified for any diagnostic performance study, as none was conducted or reported. For software V&V, the "test data" involved internal testing scenarios, not patient cases in the sense of a clinical trial.
- Data Provenance:
- For software V&V: Refers to internal test data, no patient data specified.
- For usability testing: Not specified, but generally involves human users testing the software.
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):
- Not applicable/Not specified. No diagnostic ground truth was established by experts for a clinical performance study. The ground truth for software V&V would be the expected output or functionality as per design specifications.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not applicable/Not specified. No clinical test set requiring adjudication of diagnostic findings was used.
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 explicitly states: "No human clinical data is included in this premarket notification to support substantial equivalence." This is because the device is classified as a "Picture archiving and communications system" (CFR 892.2050), and its primary function is visualization and planning aid, not automated diagnosis that would typically warrant such a study for initial market clearance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- The device is software intended "to aid in the visualization, diagnosis and planning of endodontic treatment... utilizing DICOM images." It processes images to create 3D views and allows for measurements and planning. There is an implicit assumption that a human user (dentist/endodontist) will be "in-the-loop" to interpret and act upon the visualizations and planning tools provided by the software. While the software itself functions "standalone" (i.e., it doesn't require another piece of medical equipment to generate its output once given DICOM files), its intended use is as an aid for a human, not as an autonomous diagnostic algorithm. No study for "algorithm only" performance for diagnostic accuracy (e.g., sensitivity/specificity for detecting a condition) was reported.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- For software verification and validation, the ground truth is the expected behavior and output based on design specifications and engineering requirements. For usability, the "ground truth" is that trained users can operate the software effectively and without creating undue risk, as measured by usability metrics (e.g., SUS scores and qualitative observations).
- No clinical ground truth (expert consensus, pathology, outcomes data) for diagnostic accuracy was established or used, as no clinical study for diagnostic performance was conducted.
8. The sample size for the training set:
- Not applicable/Not specified. The document describes a "Picture archiving and communications system" that creates 3D images from existing 2D DICOM data. It does not describe an AI/machine learning algorithm that requires a "training set" in the context of pattern recognition or diagnostic classification. The processing described (e.g., volume rendering) is based on established computational geometry and image processing techniques.
9. How the ground truth for the training set was established:
- Not applicable/Not specified, as there is no mention of an AI/ML model with a training set.
In summary, the 510(k) submission for the 3D Endo™ Software primarily relies on a comparison to a predicate device and extensive software verification and validation, along with usability testing, rather than a clinical performance study measuring diagnostic accuracy against a ground truth dataset.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).