(227 days)
The NobelDesign Software uses patient-specific data from scanners and defines the shapes of dental prosthetic devices such as dental abutments, copings, crowns and bridges through the use of a CAD tool. The software also serves as a means of creating, ordering and managing cases.
The software is intended to be used at a dental clinic or a dental laboratory. The software uses patient-specific data, e.g., a scanned geometry, and facilitates the CAD of an individual restorative solution based on these data, e.g., a supporting dental framework. The design data of the individualized design and any associated standardized components are transmitted via a network to a production facility where the individualized prosthetic device is manufactured.
NobelDesign software is an integrated CAD Software in a cockpit application that enables the user to scan, design and order the designed products (prosthetics).
The software is a solution that allows the use of the NobelProcera 2G scanner and integrates a third party software which is used to perform the actual design (CAD design) for the previously cleared Nobel Biocare patient specific dental implant abutments and implant bridges.
Using this system it is possible for the user to create and track cases, scan models and order NobelProcera prosthetics from the global production resources of Nobel Biocare. NobelDesign v1.0 includes to the following modules:
- Design Setup. Module to allow users to define restoration type and design on which tooth position
- Scan Center. Module to assist in acquiring all required scans for the design . setup.
- Case Management. Module with tools to help manage designs.
- Basic Application. Basic functionality (About, Help framework) are in place.
- Order Manager. Module for ordering designed components and order tracking.
- CAD Design. Module to allows designing of different types of restorations . virtually with easy to use tools.
Each physical product (prosthetic) which is designed in the software has been previously FDA cleared with its own specifications and requirements.
The provided document is a 510(k) premarket notification for the NobelDesign Software, a CAD tool for designing dental prosthetic devices. The document clarifies that this is a new device and seeks substantial equivalence to a predicate device (Procera Software, K053602).
Here's an analysis of the acceptance criteria and study information based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance:
The document doesn't provide a specific table of numerical acceptance criteria or quantitative performance metrics in the way one might expect for a diagnostic or therapeutic device. Instead, the acceptance criteria are implicitly defined by the successful completion of verification and validation testing to demonstrate that the software meets its design requirements and is substantially equivalent to the predicate device.
The "reported device performance" is described qualitatively as:
- Ability to define shapes of dental prosthetic devices: The software uses patient data from scanners to define shapes of dental abutments, copings, crowns, and bridges via a CAD tool.
- Case management: It creates, orders, and manages cases.
- Integration and functionality: It integrates scanning, design (CAD), and ordering modules effectively.
- Adherence to standards: It conforms to Quality System Regulation (21 CFR § 820), ISO 13485, and IEC 62304 (2006).
The focus of this 510(k) is on the functional equivalence and safety in the context of a CAD software, rather than quantifiable clinical performance metrics. The clearance explicitly states that "The final abutment design and all other screw retained restorations (e.g., abutments, crowns and bridges) will continue to be sent to a Nobel Biocare facility for validation of the design for appropriate dimensions and manufacturing." This implies that the software's primary role is to create a design file, and the ultimate "performance" in terms of fit and function is validated downstream by the manufacturer.
2. Sample Size Used for the Test Set and Data Provenance:
The document mentions "an entire end-to-end validation (scan, design, and order) for multiple worst-case scenario scan and design requirements." However, it does not specify the exact sample size (e.g., number of cases, number of scan datasets, number of designs).
The data provenance is not explicitly stated in terms of country of origin. The test data are described as "worst-case scenario scan and design requirements," so it's likely synthetic or carefully selected real-world data designed to challenge the software's capabilities. It's not specified whether the data were retrospective or prospective, but for software validation, it often involves pre-existing or engineered datasets.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:
The document does not specify the number of experts or their qualifications used to establish ground truth for the test set. Given that the software designs dental prosthetics, "ground truth" would likely involve adherence to dental anatomical principles, manufacturing specifications, and clinical best practices for prosthetic design. This would typically be assessed by experienced dental professionals or engineers specializing in dental prosthetics.
4. Adjudication Method for the Test Set:
The document does not explicitly state an adjudication method. For software validation, ground truth is often established by comparing the software's output against predefined engineering specifications, expert-reviewed designs, or reference physical models. The statement "The final abutment design... will continue to be sent to a Nobel Biocare facility for validation of the design for appropriate dimensions and manufacturing" suggests an internal validation process at the manufacturing facility for the final product, which would involve expert review.
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 software validation process focusing on functionality, safety, and substantial equivalence to a predicate device, not a comparative study of human reader performance with and without AI assistance.
6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done:
Yes, a standalone validation of the algorithm (software) was primarily done. The testing described is focused on the software's ability to fulfill its design requirements: "The performance of the NobelDesign software was verified and validated following the guidance provided in FDA Guidance General Principles of Software Validation... This documentation includes testing which demonstrates that the requirements for the features have been met." This implies evaluating the software's functions (scanning, designing, ordering modules) in isolation to ensure they work as intended. While humans use the software, the validation focuses on the software's correct execution of its programmed functions. The "human-in-the-loop" aspect exists in the sense that a human operator is using the CAD tool, but the validation is of the tool itself.
7. The Type of Ground Truth Used:
The ground truth used for the software validation is based on engineering specifications and functional requirements. The phrase "testing which demonstrates that the requirements for the features have been met" indicates that the software's output (e.g., the generated prosthetic design files) was compared against predetermined specifications for accuracy, completeness, and adherence to design principles. Given the context of dental prosthetic design, this would implicitly involve principles derived from dental anatomy, biomechanics, and manufacturing constraints.
8. The Sample Size for the Training Set:
The document does not provide information on the sample size for a training set. The NobelDesign Software is described as a CAD tool. CAD software typically relies on algorithms based on geometric principles and design rules, rather than machine learning models that require a "training set" of data in the same way an AI diagnostic algorithm would. If any machine learning components were embedded, their training data is not mentioned.
9. How the Ground Truth for the Training Set Was Established:
Since there's no mention of a traditional "training set" for a machine learning model, this question is not directly applicable. For CAD software, the "ground truth" for its underlying rules and capabilities is established by engineering and dental design principles coded into the software by developers and subject matter experts.
§ 872.3630 Endosseous dental implant abutment.
(a)
Identification. An endosseous dental implant abutment is a premanufactured prosthetic component directly connected to the endosseous dental implant and is intended for use as an aid in prosthetic rehabilitation.(b)
Classification. Class II (special controls). The guidance document entitled “Class II Special Controls Guidance Document: Root-Form Endosseous Dental Implants and Endosseous Dental Implant Abutments” will serve as the special control. (See § 872.1(e) for the availability of this guidance document.)