Search Results

Ask a specific question about this device

Navident is a computerized dental navigational system intended to assist preoperative planning and to guide drilling in a patient jaw during implantation surgery, using pre-acquired CT scan of the jaw. The device is intended for use by a qualified dental surgeon in the treatment of partially or fully edentulous jaws.

Navident is an image-guided dental navigational system intended to assist with preoperative planning and real-time positioning of drilling tools during implantation surgery. In particular, Navident provides visual, real-time feedback on the location of the working tip of a dental handpiece. It shows the location and direction of the tip relative to a volumetric CT image of the patient's jaw registered to that anatomy, and, when available, relative to a path planned on that image. Navident is comprised of the following parts: The main system is comprised of a cart that carries a stereoscopic video camera and a laptop with pre-installed proprietary software. The Navident system also includes several types of accessories: Jaw motion Tracking Accessories, Dental Handpiece Tracking Accessories, Registration accessories, Calibrator. Navident's four core functions are: Model, Plan, Register, Guide.

Here's an analysis of the provided text to extract information about the acceptance criteria and the study proving the device meets them:

Device: Navident

Product Code: PLV
Regulation Number: 21 CFR 872.4120
Regulation Name: Bone Cutting Instrument And Accessories
Regulatory Class: Class II

1. Table of Acceptance Criteria and Reported Device Performance

The provided text focuses on the performance data of the device rather than explicit "acceptance criteria" presented as pass/fail thresholds against specific metrics, except for the accuracy at the drill tip. Instead, it describes various validation and testing activities that collectively demonstrate the device's acceptable performance.

2. Sample Sizes and Data Provenance

• Test Set (Clinical Data from Literature):

  • Study 1 (Trace Registration Accuracy): 136 implants placed in 59 partially edentulous patients.
  • Study 2 (Pterygoid Implants): 63 pterygoid implants in 39 partially edentulous patients (31 with dynamic navigation/trace registration, 32 with freehand surgery).

• Data Provenance: Clinical studies were conducted outside of the US. The document states that "patient populations, user profiles, use environment, and clinical practices are considered equivalent and applicable to the US population." The studies are retrospective, as they are "published scientific literature" that was collected and analyzed.

• Human Factors/Usability Study: 15 representative users.

• Bench Testing: Sample sizes are not explicitly stated for individual bench tests (reprocessing, electrical safety, EMC, full system accuracy bench testing, stability/repeatability of attachments), but it mentions "representative devices" and "representative master devices."

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

The clinical studies referenced are "published scientific literature" and involve "qualified dental surgeon(s)" performing the procedures. The ground truth for the clinical accuracy studies (Study 1 & 2) appears to be derived from actual implant placement compared to the pre-planned position (likely measured from post-operative imaging).

• Number of Experts: Not explicitly stated as "number of experts for ground truth establishment." The studies were performed by treating surgeons. Study 2 mentions "the same surgeon" for comparative data, implying single-surgeon data collection for parts of the comparison.
• Qualifications of Experts: "Qualified dental surgeon(s)." One of the authors in the referenced papers (Stefanelli LV, Mandelaris GA) are typically oral and maxillofacial surgeons or periodontists who are experts in implant surgery.

4. Adjudication Method for the Test Set

No explicit adjudication method (e.g., 2+1, 3+1) is described for the ground truth of the clinical studies. The accuracy metrics appear to be derived from quantitative measurements comparing planned vs. actual outcomes, likely without a separate expert adjudication panel.

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

An MRMC study was not explicitly performed in the context of comparing human readers with and without AI assistance, as Navident is a surgical navigation system, not an AI-assisted diagnostic imaging device for human reading. The "comparison" is between different surgical techniques (dynamic navigation with trace registration vs. freehand surgery, or dynamic navigation with new trace registration vs. old fiducial registration) or system accuracy benchmarks, not human interpretation of images. However, the studies demonstrate the clinical effectiveness of the device-assisted procedure.

• Effect Size for AI vs. Without AI Assistance:
  • Study 1 (Trace Registration vs. Fiducial-based): The 95th percentile values for the trace registration method were "favorable" compared to the fiducial-based registration method (cleared predicate). This implies an improvement, but specific effect sizes are not quantified here beyond the descriptive "favorable." This is a comparison of system performance, not human performance.
  • Study 2 (Dynamic Navigation with Trace Registration vs. Freehand Surgery): Dynamic navigation was "more accurate." The mean deviations for dynamic navigation were significantly lower across all metrics (e.g., 0.66 mm coronal vs. 1.54 mm for freehand; 2.64° angular vs. 12.49° for freehand). This provides a clear quantitative effect size for the improvement in surgical accuracy when using the device compared to freehand.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

The "Full system accuracy bench testing" and the clinical accuracy studies (comparing planned vs. actual implant positions) inherently represent the performance of the algorithm and system, as they quantify the accuracy of the guidance provided by the device. While a human surgeon provides the input and executes the drilling, the accuracy metrics are a direct measure of the system's ability to guide to the planned position. The "Accuracy at the drill tip: ≤1.0mm" is a standalone performance metric.

7. Type of Ground Truth Used

• Clinical Studies: Ground truth for implant accuracy was based on comparison of the pre-planned position (from CT scan) with the actual post-operative implant position (derived from post-operative imaging). This is a form of outcomes data or objective measurement of surgical execution.
• Bench Testing: Ground truth for system accuracy was established through controlled measurements conducted on test setups designed to measure the mapping accuracy and stability.

8. Sample Size for the Training Set

The document does not specify a separate "training set" or its size. This is a surgical navigation system, not typically a machine learning or AI model trained on a large imaging dataset in the conventional sense. The "training" for the device's development would be more akin to algorithm development and calibration rather than data-driven machine learning, although some components might use statistical models or optimizations. The clinical literature cited pertains to validation of the device's performance, not training data for an AI.

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

Given that a specific "training set" for an AI algorithm is not mentioned, the method for establishing its ground truth is also not described. The device's core functionality relies on stereoscopic optical tracking and registration algorithms, whose "ground truth" and performance are verified through precise engineering and bench testing rather than large-scale data labelling for AI training.

Ask a specific question about this device

Navident is a computerized dental navigational system intended to assist preoperative planning and to guide drilling in a patient jaw during implantation surgery, using pre-acquired CT scan of the jaw. The device is intended for use by a qualified dental surgeon in the treatment of partial edentulism.

Navident is a cart-based, computerized, image-guided dental navigational system, operating as a combined package for performing both pre-operative planning and guided implant insertion. Using preacquired CT scan of the jaw, Navident provides the dentist (usually a GP dentist, prosthodontist, periodontist or dental surgeon) with implantation planning and real-time guidance during dental implants insertion.

The guidance function of Navident is primarily provided using a visualization of the drill pose (tip location and shaft axis direction) relative to the desired pose of the implant, as planned on a preacquired CT image of the jaw. This visualization assists the dentist in implementing their implantation plan.

Navident consists of four main components:

• 1. A notebook computer positioned above the patient's chest. The Navident software running on the computer provides both planning and navigation functionalities.
• 2. A handpiece attachment. The attachment consists of a universal removable metal adapter, and a specially marked plastic part, named DrillTag, which latches on the adapter.
• 3. A customizable patient jaw attachment consisting of a moldable stent part, named NaviStent, and a matching specially marked plastic tag, named JawTag. NaviStent can be formed, customized and tested for fit directly on the patient's jaw or on a plaster model. The patient wears the NaviStent during the CT scan and again during the surgery. Navident is able to continuously register NaviStent's position during surgery with its appearance in the CT scan automatically.
• 4. An optical position sensor which detects high contrast patterns printed on the DrillTag and JawTag and constantly reports their relative positions to the Navident software.

Here's a breakdown of the acceptance criteria and study details for the Navident device, based on the provided text:

Navident Device Performance and Acceptance Criteria

Study Details Proving Acceptance Criteria

Due to the nature of the provided document (a 510(k) summary), specific details for some categories (like distinct training vs. test sets or formal MRMC studies) are not fully separated as might be found in a later clinical trial report. However, the summary describes various performance and clinical evaluations.

1. Table of Acceptance Criteria and Reported Device Performance: (See table above)

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

   • Bench Testing: Performed with various settings and conditions (e.g., different image voxel sizes, distances, lighting, random jaw models). The exact number of "tests" or "samples" for each specific parameter is not quantified.
   • Human Factors/Usability Summative Evaluation: 15 representative users (oral surgeons, general dental practitioners).
   • Clinical Case Data: 21 cases with 36 guided implantations. The geographical provenance (country of origin) is not explicitly stated but implies international, as it mentions "a diverse international group of surgeons." This appears to be retrospective collected over time as "feedback on usage history."
   • Clinical Evaluation (Systematic Feedback): 7 surgeons who treated a total of 150 patients. The provenance is again "diverse international group of surgeons" and appears to be retrospective collection of "usage history and professional impressions."

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

   • For the human factors/usability evaluation, 15 "representative users (i.e., oral surgeons, general dental practitioners)" were observed. Their role was more as "test subjects" or "users" rather than establishing a separate "ground truth" in terms of clinical outcomes, though their feedback contributed to the "usability problems" assessment.
   • For the clinical case data and systematic feedback, the ground truth regarding "accuracy averaging 1.0mm deviation from plan at both entry and apex" was likely established through post-procedure analysis by the "experienced user" (qualified dental surgeon) themselves or by the study investigators. The qualifications of these 7 surgeons and the "diverse international group of surgeons" are defined as "qualified dental surgeon." No specific number of independent experts for ground truth adjudication is given for the accuracy assessment, rather it describes "an experienced user is able to achieve."

4. Adjudication Method for the Test Set:

   • No formal multi-reader adjudication method (e.g., 2+1, 3+1) is described for the clinical accuracy assessment. The summary states, "Accuracy assessment demonstrated that, with Navident's guidance, an experienced user is able to achieve an implant placement accuracy averaging 1.0mm deviation from plan at both entry and apex." This implies a measurement against the pre-planned position, which serves as the "ground truth" for deviation.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

   • No MRMC comparative effectiveness study, specifically comparing human readers with AI vs. without AI assistance, is mentioned. The device is a "computerized dental navigational system" providing real-time guidance, not an AI diagnostic aid that assists human readers in interpreting images. Its direct comparison is against a predicate surgical navigation system.

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

   • Yes, the "Full system bench testing" (point 3 under SUMMARY OF PERFORMANCE TESTING) evaluates the "Overall accuracy of the system in mapping the drill tip to the CT image" and other operational aspects without direct human surgical intervention as the primary performance metric. This represents the standalone performance of the navigation system's mapping and guidance precision.

7. The Type of Ground Truth Used:

   • Bench Testing: Engineering measurements against known physical standards or simulated conditions.
   • Clinical Cases (Accuracy): Deviation from the "planned pose of the implant" as visualized on the pre-acquired CT image. The pre-planned position serves as the ground truth for measuring positional / angular deviation.
   • Clinical Cases (Outcomes): "No significant patient adverse events or complications were reported." This relies on outcomes data from the surgeons.

8. The Sample Size for the Training Set:

   • The document describes a "510(k) summary," which focuses on demonstrating substantial equivalence and performance testing rather than algorithm development. It does not provide details on a specific "training set" for an AI algorithm. The device is a mechanical/software navigation system, not primarily an AI deep learning model that requires a large training dataset in the conventional sense. The "training" here refers to its design and calibration processes.

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

   • Not applicable in the context of an AI training set. The "ground truth" for the system's design and calibration would have been engineering specifications and physical measurements, ensuring its tracking capabilities (Stereoscopic triangulation of illuminated checker-board patterns) and mapping accuracy.

Ask a specific question about this device