(434 days)
The MedCAD® AccuPlan® System is intended for use as a software system and image segmentation system for the transfer of imaging information from a medical scanner such as a CT based system. The input data file is processed by the MedCAD® AccuPlan® System and the result is an output data file that may then be provided as digital models or used as input to a rapid prototyping portion of the system that produces including anatomical models, surgical guides, and dental splints for use in maxillofacial guides and dental splints are intended to be used for the maxillofacial bone in maxillofacial surgery. The MedCAD® AccuPlan® System is also intended as a preoperative software tool for simulating / evaluating surgical treatment options.
The MedCAD® AccuPlan® System is a collection of software and associated additive manufacturing equipment intended to provide a variety of outputs to support orthognathic or reconstructive surgery. The system uses electronic medical images of the patient's anatomy or stone castings made from the patient anatomy with input from the physician, to manipulate original patient images for planning and executing surgery. The patient specific outputs from the system includes anatomical models, surgical quides, dental splints, and patient-specific case reports.
This document (K192282) is a 510(k) premarket notification for the MedCAD® AccuPlan® System. It describes the device, its intended use, and argues for its substantial equivalence to a predicate device (VSP® System, Medical Modeling, Inc. K120956).
The document does NOT contain information about specific acceptance criteria, a detailed study proving the device meets those criteria, or the types of quantitative performance metrics typically associated with AI/ML device approval (e.g., sensitivity, specificity, AUC). This is likely because the MedCAD® AccuPlan® System, as described, is primarily a software system for image processing and generating physical models/guides, rather than an AI/ML diagnostic or prognostic tool that would require such performance evaluations. The performance data section focuses on manufacturing process validation, dimensional analysis, mechanical performance, and software system validation for basic functionality, rather than clinical performance metrics based on a test set.
Therefore, many of the requested details related to "acceptance criteria" and "study that proves the device meets the acceptance criteria" in the context of AI/ML performance metrics cannot be extracted from this document. The document focuses on demonstrating that the device meets design inputs and is substantially equivalent to a predicate device in its intended function and manufacturing quality.
Below, I will extract relevant information that is present in the document, acknowledging where the requested information is absent or not applicable based on the nature of this device's submission.
1. A table of acceptance criteria and the reported device performance
The document does not explicitly present a table of quantitative acceptance criteria for clinical performance (e.g., sensitivity, specificity for disease detection) and corresponding reported device performance metrics in the way an AI/ML diagnostic device would.
Instead, the "Performance Data" section describes categories of validation performed:
| Acceptance Criteria (Implied / Type of Validation) | Reported Device Performance (Summary) |
|---|---|
| Device Performance Validation | "Successfully demonstrates that design outputs meet design inputs." |
| Process Validation (IQ, OQ, PQ) | "Ensure that the manufacturing process can effectively produce patient-matched devices." "Equipment used for production purposes have been qualified to ensure the equipment used for manufacturing... meet production needs." |
| Dimensional Analysis | "Performed to ensure the proper fit of the final output." (No specific metric provided, just that it was done and presumably passed). |
| Mechanical Performance | "Assessing dynamic compressive strength and ligature wire pullout testing were conducted on final, finished devices demonstrating they are equivalent to the predicate device." (Implies meeting equivalence thresholds, but no specific values are given.) |
| Software System Validation | "Off-the-shelf software packages are operating correctly and any necessary file conversions will not negatively impact the final output." |
| Independent Subsystem Verification | "Verification of each independent software subsystem against defined requirements" |
| Subsystem Compatibility Verification | "Verification of compatibility between software subsystems against defined requirements" |
| Integrated System Validation | "Validation of fully integrated system including all subsystems against overall system requirements" |
| Sterilization Validation | "Conducted in accordance with international standard ISO 17665 and FDA guidance document... to a Sterility Assurance Level (SAL) of 1x10-9. All test method acceptance criteria were met." |
| Biocompatibility Validation | "Conducted in accordance with international standard ISO 10993-1 and FDA guidance document... The results of the testing adequately address biocompatibility for the output devices and their intended use." (No specific metrics, but attests to meeting standard requirements). |
2. Sample sizes used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)
This information is not provided in the document. The document describes validation efforts in terms of process, software, and material properties, rather than an evaluation of clinical performance on a "test set" of patient data for diagnostic accuracy. The input data comes from "medical scanners such as a CT based system." There is no mention of the origin or type of patient data used for any performance validation beyond its source (medical scanners).
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 applicable/provided in the document. The device is a "software system and image segmentation system" for planning and producing physical outputs. It's not a diagnostic AI/ML tool requiring expert-established ground truth for disease detection for a test set. The process involves "clinical input and review from the physician" for planning, but this is part of the operational workflow rather than establishing a "ground truth" for an algorithm's performance evaluation.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
This information is not applicable/provided in the document, for the same reasons as #3.
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 applicable/provided in the document. An MRMC study is typically performed for AI/ML diagnostic aids that assist human readers. The MedCAD® AccuPlan® System's stated function does not involve assisting human readers in interpreting images for diagnosis. It's a tool for planning and manufacturing.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
This information is not applicable/provided as a formal standalone performance study with clinical endpoints. The software aspects are validated for correctness and compatibility of data processing and file conversions, as noted in the "Software System Validation" section. However, this is related to its functional accuracy as a tool, not its diagnostic accuracy in a clinical context. The device is described as being "operated only by trained MedCAD employees" and requiring "clinical input and review from the physician during planning." This implies it's always used with human oversight.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
This information is not applicable/provided in the context of traditional "ground truth" for diagnostic AI. For the dimensional analysis of the physical outputs (anatomical models, surgical guides, dental splints), the "ground truth" would be the intended dimensions/specifications derived from the original medical images and planning. The "proper fit of the final output" is assessed against these design inputs.
8. The sample size for the training set
This information is not provided in the document. The MedCAD® AccuPlan® System is described as using "Commercial Off-The-Shelf (COTS) software to manipulate 3-D medical scan images." There is no indication of a machine learning (ML) model being "trained" in the conventional sense, as is common with many AI devices. The validation focuses on the correct functioning and integration of these COTS software components and the manufacturing process.
9. How the ground truth for the training set was established
This information is not applicable/provided since there is no mention of a training set or an ML model being trained for this device as it is characterized in the 510(k) submission.
§ 872.4120 Bone cutting instrument and accessories.
(a)
Identification. A bone cutting instrument and accessories is a metal device intended for use in reconstructive oral surgery to drill or cut into the upper or lower jaw and may be used to prepare bone to insert a wire, pin, or screw. The device includes the manual bone drill and wire driver, powered bone drill, rotary bone cutting handpiece, and AC-powered bone saw.(b)
Classification. Class II.