Search Results

The MedCAD® AccuStride™ System is intended to be used as a surgical instrument to assist in preoperative planning and/or in guiding the marking of bone and/or guide surgical instruments in nonacute, non-joint replacing osteotomies in the foot for adult and pediatric patients 12 years of age and older. The MedCAD® AccuStride™ System surgical guides are intended for single use only. The MedCAD® AccuStride™ System surgical guides should only be used when the anatomic landmarks necessary for pre-operative planning can be clearly identified on the patient's radiographic images (i.e., CT).

The MedCAD® AccuStride™ System is a collection of two individual pieces of software and associated additive manufacturing equipment intended to provide a variety of outputs to support non-acute, nonjoint replacing osteotomies in the foot. The system uses electronic medical images of the patient's anatomy or 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 guides, and patient-specific case reports.

Following the MedCAD® Quality System and specific Work Instructions, trained employees utilize Commercial Off-The-Shelf (COTS) software to manipulate 3-D medical Computed Tomography (CT) images to create patient-specific physical and digital outputs. The process requires clinical input and review from the physician during and prior to delivery of the final outputs. It should be noted that the system is operated only by trained MedCAD employees, and the physician does not directly input information. The physician provides input for model manipulation and interactive feedback through viewing of digital models of system outputs that are modified by the engineer during the planning session.

The FDA 510(k) summary for the MedCAD® AccuStride™ System provides information about the device's acceptance criteria and the studies conducted. However, it does not provide detailed acceptance criteria and performance metrics for the software's ability to perform preoperative planning or guide marking, as typically expected for pure AI/ML software. Instead, the performance testing focuses on the physical components (surgical guides and anatomical models) produced by the system.

Here's an attempt to extract and infer the requested information based on the provided text, while also highlighting the limitations:

1. Table of Acceptance Criteria and Reported Device Performance

Important Note: The document focuses on the outputs of the system (surgical guides and anatomical models) rather than the standalone performance of the software in performing planning or marking. The "acceptance criteria" presented here are high-level outcomes of physical tests related to these outputs. There are no explicit metrics for the accuracy or efficacy of the "surgical instrument to assist in preoperative planning and/or in guiding the marking of bone" as a software component itself.

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

• Wear Debris Testing: No specific sample size is provided beyond "Cutting / drilling instruments were used on a worst-case titanium surgical guide."
• Simulated Use Cadaver Validation Testing: No specific sample size is given for the number of cadavers used. It mentions "Guides were created to support bunion and metadductus, metatarsal 2+3, akin, and calcaneal osteotomies," implying a variety of cases were tested.
• Data Provenance: The cadaver validation testing is experimental ("simulated surgeries in cadavers") rather than derived from retrospective or prospective patient data. There is no information on the country of origin.

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

The document states, "The process requires clinical input and review from the physician during and prior to delivery of the final outputs." However, it does not specify the number of experts, their qualifications, or how they established the ground truth for the performance testing. The input and review from physicians appear to be part of the operational process rather than a formal ground truth establishment for a validation study.

4. Adjudication Method for the Test Set

No specific adjudication method (e.g., 2+1, 3+1) is mentioned for the performance testing. The "clinical input and review from the physician" is described as part of the normal workflow.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No MRMC comparative effectiveness study is mentioned. The performance testing focuses on the physical guides in a simulated cadaver environment, not on human readers' improvement with or without AI assistance.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

The device description explicitly states: "the system is operated only by trained MedCAD employees, and the physician does not directly input information. The physician provides input for model manipulation and interactive feedback through viewing of digital models of system outputs that are modified by the engineer during the planning session." This indicates that the system is not a standalone AI algorithm without human-in-the-loop. It's a system where trained MedCAD employees (a human element) use software to develop outputs based on physician input and feedback. The performance testing reflects this human-in-the-loop process for generating the physical guides.

7. The Type of Ground Truth Used

• Wear Debris Testing: The ground truth for this test is implicit in "less than that reported in the literature to be safe." This refers to established safety thresholds for wear debris.
• Simulated Use Cadaver Validation Testing: The ground truth for this test appears to be "pre-surgical plans" for angle and position data, and subjective evaluation of "usability and fit." This is an engineering/design validation approach rather than pathology, expert consensus, or outcomes data from clinical cases.

8. The Sample Size for the Training Set

The document does not provide any information regarding a training set size. The system uses "Commercial Off-The-Shelf (COTS) software to manipulate 3-D medical Computed Tomography (CT) images." This suggests the core image manipulation algorithms might be pre-trained components within the COTS software, or the system does not involve machine learning in a way that requires a specifically defined "training set" for the AccuStride™ System's unique functionalities.

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

Since no training set is described for the MedCAD® AccuStride™ System functionality, there is no information on how its ground truth was established.

Ask a specific question about this device

The MedCAD® AccuPlate® 3DTi Patient-Specific Plating System is intended for use in the stabilization, fixation, and reconstruction of the maxillofacial / midface and mandibular skeletal regions in adolescents (greater than 12 to 21 years of age) and adults.

The MedCAD AccuPlate® 3DTi Patient-Specific Plating System is a metal bone plate used in conjunction with commercially available, non-locking metal bone screws for the fixation to bone in orbital, midface / maxillofacial, and non-continuity mandibular operations. The design and dimensions of each plate within the envelope specification is based upon the patient's anatomical data (CT scan, CBCT scan, or MRI), and the intended anatomy to be fixated as determined from input provided by the surgeon. The subject device is not intended to be bent or modified in surgery. If for any reason the surgeon chooses not to use the subject device in surgery, they may use any of the commercially available plates to complete the surgery. The subject device is additively manufactured from Ti-6AL-4V Extra Low Interstitial (ELI) titanium alloy, provided non-sterile, must be sterilized prior to use, and is intended for single use only. Plates are fastened to bone using commercially available non-locking bone screws with diameters ranging from 1.5mm to 2.7mm and lengths ranging from 3.5mm to 22mm.

Here's a summary of the acceptance criteria and study information for the MedCAD® AccuPlate® 3DTi Patient-Specific Plating System, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The FDA clearance letter mentions the following non-clinical performance tests. The acceptance criteria are implied to be "equivalent" to identified reference devices.

2. Sample Sizes Used for the Test Set and Data Provenance

The document does not explicitly state the sample sizes (number of devices or tests) for the non-clinical performance tests (bending, pushout, fit and form validation). It also does not specify the country of origin of the data or whether the studies were retrospective or prospective, as these are non-clinical (mechanical and material) performance tests rather than clinical studies with patient data.

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

This information is not explicitly provided in the document. For non-clinical performance tests, "ground truth" is typically established by engineering standards and validated testing procedures, not by human experts in the same way as clinical image interpretation. The "Fit and Form Validation" mentions alignment with an "approved surgical plan," which would involve surgical expertise in its creation, but the number and qualifications of individuals involved in approving these plans or assessing the fit are not detailed.

4. Adjudication Method for the Test Set

Not applicable for these types of non-clinical, objective performance tests. Adjudication methods like 2+1 or 3+1 are typically used for subjective clinical interpretations by multiple readers (e.g., radiologists) in diagnostic accuracy studies.

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 such study was mentioned. The MedCAD® AccuPlate® 3DTi Patient-Specific Plating System is a patient-specific surgical implant, not an AI-assisted diagnostic tool.

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

This concept is not directly applicable. The device itself is an implant. However, the "patient specific design process" for creating the implant is an "algorithm only" type of process in that it uses patient anatomical data to generate the device design. The "Fit and Form Validation" likely assessed the output of this design process against the intended surgical plan. The document states that the design process "was shown to produce devices that aligned with the approved surgical plan," indicating successful standalone performance of the design software.

7. The Type of Ground Truth Used

• Biocompatibility: Established by adherence to ISO 10993 standards and FDA guidance.
• Sterilization Validation: Established by adherence to ISO 17665 standards and FDA guidance, with a specific SAL target.
• Static and Dynamic Bending: Established by ASTM F382 standard, with comparison to a reference device's known performance.
• Axial Screw Pushout: Established by ASTM F543 standard, with comparison to a reference device's known performance.
• Fit and Form Validation: Established by alignment with an "approved surgical plan." This likely involves a comparison of the 3D-printed plate geometry with the virtual surgical plan derived from patient imaging data, which can be considered "design ground truth."

8. The Sample Size for the Training Set

The document does not describe the development or training of an AI algorithm in the traditional sense, so there is no mention of a training set sample size. The "patient-specific design software" is mentioned as being the same as a reference device (K192282), implying it's a previously validated system rather than a newly trained AI model.

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

As no training set for an AI algorithm is described, this information is not applicable.

Ask a specific question about this device

MedCAD® AccuPlan® Orthopedics 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 system and the result is an output data file. This file may then be provided as digital models or used as input to a rapid prototyping portion of the system that produces physical outputs including anatomical models of the fibula and ilium. surgical guides for harvesting bone grafts from the fibula or ilium, and surgical planning case reports for use in maxillofacial reconstructive surgeries. MedCAD® AccuPlan® Orthopedics System is also intended as a pre-operative software tool for simulating / evaluating surgical treatment options. The MedCAD® AccuPlan® Orthopedics System is indicated for use in adolescents (greater than 12 to 21 years of age) and adults.

The MedCAD® AccuPlan® Orthopedics System is a collection of software and associated additive manufacturing equipment intended to provide a variety of outputs to support harvesting of bone to support maxillofacial reconstructive surgeries. The system uses electronic medical images of the patient's anatomy with input from the physician to manipulate original patient images for planning and executing surgery. The patient specific outputs from the system include anatomical models, surgical guides, and patient-specific case reports.

Following the MedCAD® Quality System and specific Work Instructions, trained employees utilize Commercial Off-The-Shelf (COTS) software to manipulate 3-D medical Computed Tomography (CT) images to create patient-specific physical and digital outputs. The process requires clinical input and review from the physician during and prior to delivery of the final outputs. While the process and dataflow vary somewhat based on the requirements of a given patient and physician, the following description outlines the functions of key sub-components of the system, and how they interact to produce the defined system outputs. It should be noted that the system is operated only by trained MedCAD employees, and the physician does not directly input information. The physician provides input for model manipulation and interactive feedback through viewing of digital models of system outputs that are modified by the engineer during the planning session.

The MedCAD® AccuPlan® Orthopedics System is made up of two individual pieces of software for the design and various manufacturing equipment integrated to provide a range of anatomical models (physical and digital), surgical guides, and patient-specific planning reports for harvesting of bone from the fibula and ilium for use in maxillofacial reconstructive surgeries.

The MedCAD® AccuPlan® Orthopedics System requires an input 3-D image file from medical imaging systems (i.e., CT). This input is then used, with support from the prescribing physician to provide the following potential outputs to support maxillofacial reconstructive surgery. Each system output is designed with physician inout and reviewed by the physician prior to finalization. All outputs are used only with direct physician involvement to reduce the criticality of the outputs.

System outputs include:

• Anatomical Models
• Surgical Guides
• Patient-Specific Case Reports

Here's a breakdown of the acceptance criteria and study information for the MedCAD® AccuPlan® Orthopedics System, based on the provided FDA 510(k) summary:

This device is primarily a software system for surgical planning and the creation of physical outputs (anatomical models and surgical guides) based on CT scan data. The performance testing detailed here focuses on the physical outputs rather than an AI-driven diagnostic or assistive algorithm's accuracy.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Wear Debris Testing: Not explicitly stated, but "worst-case titanium surgical guide" indicates at least one such guide was tested.
• Fit and Form Validation: "All samples" were tested. The exact number of samples is not explicitly stated.
• Data Provenance: Not specified in the provided text (e.g., country of origin, retrospective/prospective). However, since the system processes patient-specific CT imaging, the "data" would be the CT scans themselves.

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

• The document states that "The process requires clinical input and review from the physician during and prior to delivery of the final outputs" and "All outputs are used only with direct physician involvement to reduce the criticality of the outputs."
• For the performance testing itself, the identity and number of "experts" (e.g., those determining if a fit was "PASS") are not specified. The ground truth for fit and form seems to be established by physical/optical measurement against a 3D model and representative anatomical model.
• For the broader system design and physician input, "the prescribing physician" provides input and review, but their qualifications and number are not detailed beyond "physician."

4. Adjudication Method for the Test Set

• No specific adjudication method (e.g., 2+1, 3+1 consensus) for the performance tests is described. The "PASS" result suggests a clear acceptance/rejection criteria was applied.
• For the general operation of the system, it notes that the "physician provides input for model manipulation and interactive feedback through viewing of digital models of system outputs that are modified by the engineer during the planning session." This implies an iterative, interactive process rather than a formal adjudication of a test set.

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

• No, an MRMC comparative effectiveness study was not performed or described. The device is a surgical planning and manufacturing system, not an AI diagnostic aid evaluated for human reader improvement. The performance testing focuses on the physical properties and accuracy of the manufactured outputs.

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

• The performance testing performed (wear debris, fit and form, sterilization, biocompatibility) relates to the outputs of the system (surgical guides, anatomical models) rather than the standalone algorithmic accuracy of image segmentation or planning.
• The system itself relies on "trained MedCAD employees" and "clinical input and review from the physician." Therefore, it's not a purely standalone AI algorithm without human involvement in its operational workflow.

7. The Type of Ground Truth Used

• For Fit and Form Validation: The ground truth appears to be the digital 3D model ("alignment with the 3D model") and a representative anatomical model ("fitting the guide over the corresponding defect"). These are objective measurements and physical fit evaluations.
• For Wear Debris Testing: The ground truth for safety is published literature values of safe wear debris.
• For Sterilization and Biocompatibility: Ground truth is established by international standards (ISO 17665, ISO 10993-1) and FDA guidance documents, which define accepted levels and methodologies.

8. The Sample Size for the Training Set

• Not applicable / Not provided. This device is described as using "Commercial Off-The-Shelf (COTS) software to manipulate 3-D medical Computed Tomography (CT) images." There is no mention of a machine learning model that requires a dedicated training set. The "software" component appears to be tools for engineers and physicians to interact with and refine 3D models.

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

• Not applicable. As no training set for a machine learning model is mentioned, there's no ground truth establishment process to describe for a training set.

Ask a specific question about this device

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 physical outputs including anatomical models, surgical guides, and dental splints for use in maxillofacial surgery. The surgical 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 pre-operative 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, dental splints, surgical guides, and patient-specific case reports.

Following the MedCAD® Quality System and specific Work Instructions, trained employees utilize Commercial Off-The-Shelf (COTS) software to manipulate 3-D medical scan images which can include Computed Tomography (CT), Cone Beam CT (CBCT), and/or 3-D scan images from patient physical models (stone models of the patient's teeth) to create patient-specific physical and digital outputs. The process requires clinical input and review from the physician during planning and prior to delivery of the final outputs. While the process and dataflow vary somewhat based on the requirements of a given patient and physician, the following description outlines the functions of key sub-components of the system, and how they interact to produce the defined system outputs. It should be noted that the system is operated only by trained MedCAD employees, and the physician does not directly input information. The physician provides input for model manipulation and interactive feedback through viewing of digital models of system outputs that are modified by the engineer during the planning session.

The MedCAD® AccuPlan® System is made up of 4 individual pieces of software for the design and various manufacturing equipment integrated to provide a range of anatomical models (physical and digital), dental splints, surgical guides, and patient-specific planning reports for reconstructive surgery in the maxillofacial region.

The MedCAD® AccuPlan® System requires an input 3-D image file from medical imaging systems (i.e. - CT) and/or implant file. This input is then used, with support from the prescribing physician to provide the following potential outputs to support reconstructive surgery. Each system output is designed with physician input and reviewed by the physician prior to finalization. All outputs are used only with direct physician involvement to reduce the criticality of the outputs.

System outputs include:

• Anatomical Models
• Surgical Guides
• Dental Splints
• Patient-Specific Case Reports

The purpose of this submission was to add titanium cutting / drilling guides to the family of available patient specific outputs. Cutting and drilling instruments can only be used with titanium cutting / drilling guides. Polymer guides are to be used for marking and positioning of anatomy only.

The MedCAD® AccuPlan® System is cleared by the FDA as a software and image segmentation system for maxillofacial surgery. The primary purpose of this specific submission (K223024) was to add titanium cutting/drilling guides to the family of available patient-specific outputs, which were not part of the previous K192282 clearance.

Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The FDA clearance relies on performance testing to demonstrate substantial equivalence, particularly concerning the new titanium cutting/drilling guides. The document highlights two key performance tests:

The document also mentions:

• Sterilization Validation: In accordance with ISO 17665 and FDA guidance, to a Sterility Assurance Level (SAL) of 1x10^-6. All test method acceptance criteria were met.
• Biocompatibility Validation: In accordance with ISO 10993-1 and FDA guidance. Results adequately address biocompatibility for the output devices and their intended use.

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

• Wear Debris Testing: "Cutting / drilling instruments were used on a worst-case titanium surgical guide." The sample size is not explicitly stated but implies at least one worst-case guide was tested.
• Fit and Form Validation: "Subject devices from historical cases were manufactured." The sample size is not explicitly stated beyond "All samples met the predetermined acceptance criteria." The provenance is implied to be retrospective as it uses "historical cases." The country of origin of the data is not specified.

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

The document does not mention the use of experts to establish a "ground truth" in the traditional sense for these performance tests. The ground truth for the device's physical outputs (surgical guides) appears to be derived from engineered specifications and objective measurements (optical scanning, fit to master models). The system relies on trained MedCAD employees and physician input for planning, but this is part of the operational workflow rather than a ground truth establishment process for performance testing.

4. Adjudication Method for the Test Set

Not applicable. The performance testing described (wear debris, fit and form) does not involve human readers or a need for adjudication in the context of diagnostic agreement. It's a technical validation of physical properties.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done, What was the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

No MRMC comparative effectiveness study was mentioned or performed. This device is described as a software system for planning and manufacturing physical outputs (guides, splints, models), which are then used in surgery, rather than an AI-based diagnostic tool that directly assists human readers in interpreting medical images for diagnosis. The system is operated by "trained MedCAD employees" and involves "clinical input and review from the physician during planning," but it's not described as an AI assistance tool for human interpretation of images.

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

The performance tests described (wear debris and fit/form) evaluate the physical characteristics and dimensional accuracy of the manufactured outputs, which are direct results of the system's (algorithm's) processing and manufacturing. In that sense, aspects of "standalone" performance of the physical output are assessed. However, the system is explicitly stated as requiring "clinical input and review from the physician" for planning, meaning it's generally a human-in-the-loop system in its intended use, rather than a fully autonomous diagnostic algorithm.

7. The Type of Ground Truth Used

The ground truth for the performance tests appears to be:

• Engineered Specifications/Design Accuracy: For Fit and Form Validation, the manufactured devices are compared to the "3D model" (the digital design generated by the system based on patient imaging). This implies the 3D model itself serves as the ground truth for ideal form and alignment.
• Literature-based Safety Thresholds: For Wear Debris Testing, the acceptance criterion is quantitative: less than "that reported in the literature to be safe." This indicates a ground truth derived from existing scientific literature on safe levels of wear debris from similar materials/applications.

8. The Sample Size for the Training Set

The document does not provide information about a "training set" or "training data" for the MedCAD® AccuPlan® System. This suggests that the system's functionality is not based on a machine learning model that requires a training phase with labeled data in the way many AI/ML medical devices do. It appears to be a rule-based or engineering-based software for image processing, segmentation, and design for manufacturing.

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

Since no training set is mentioned (refer to point 8), this question is not applicable.

Ask a specific question about this device

The MedCAD® AccuShape® Titanium Patient-Specific Cranial Implant is designed individually for each patient and intended to correct defects / replace bony voids in the cranial skeleton.

The MedCAD® AccuShape® Titanium Patient-Specific Cranial Implant is a preformed non-alterable cranioplasty plate that cannot be altered or reshaped at the time of surgery and is designed to be implanted in a patient to repair a skull defect.
The subject device is composed of commercially pure (CP) Grade 2 titanium per ASTM F67. The manufacturing process is subtractive manufacturing (CNC milled) from models created and developed from patient specific CT Scan Data. The software used in this process is identical to the software used in the predicate device (K110684). The device is designed to have, as requested by the physician, drainage holes over the defect void area, fixation holes over an onlay area, and retractions and other features that fall within the approved design envelope. All designs must be approved by the physician prior to manufacture.

The provided document describes the MedCAD AccuShape Titanium Patient-Specific Cranial Implant and its substantial equivalence to a predicate device (MedCAD AccuShape PEEK Patient Specific Cranial Implant) based on non-clinical performance testing.

It is important to note that this document does not describe an AI/ML-driven device or study parameters typical for such devices (e.g., ground truth establishment for a training set, human reader studies, or expert consensus on clinical data). The device described is a physical cranial implant, and the study referenced in the document is a series of non-clinical performance tests designed to assess the physical and mechanical properties of the implant, not its diagnostic or predictive accuracy in an AI context.

Therefore, many of the requested bullet points, particularly those pertaining to AI/ML device evaluation (like sample size for test/training sets of data, number of experts for ground truth, MRMC studies, standalone performance), are not applicable to the information provided in this document.

However, I can extract information relevant to the device's acceptance criteria and the non-clinical performance testing performed for this physical device.

Here's an interpretation of the "acceptance criteria" and "study" as presented for a physical medical device, rather than an AI/ML diagnostic:

Device: MedCAD AccuShape Titanium Patient-Specific Cranial Implant (K220357)

Purpose of the "Study" (Non-Clinical Performance Testing): To demonstrate the substantial equivalence of the MedCAD AccuShape Titanium Patient-Specific Cranial Implant to its predicate device (MedCAD AccuShape PEEK Patient Specific Cranial Implant, K110684) by evaluating its physical and mechanical properties.

1. A table of acceptance criteria and the reported device performance

2. Sample size used for the test set and the data provenance

• Sample Size for Test Set:
  • Evaluation of Fit Testing: "3 large defect predicate historical cases (K110684)" were used to generate "worst case CT data". The number of manufactured implants tested is implied to be at least 3 (one for each case). The phrase "All samples" in the result suggests a specific number of manufactured implants were produced and tested, but the exact number isn't quantified beyond the 3 cases used for input data.
  • Comparative Strength: "Identical subject and predicate devices" were used, implying at least one (and likely more for statistical significance, though not stated) of each type (titanium and PEEK) for comparative testing.
  • Screw Fixation: Not explicitly stated, but implies multiple tests to determine "at least as strong as" criteria.
  • MR Compatibility: At least one device (or representative sample) would be tested.
• Data Provenance: The "worst case CT data" for the Evaluation of Fit testing came from "3 large defect predicate historical cases (K110684)". This suggests a retrospective use of previously acquired clinical data (CT scans) from actual patients. The country of origin is not specified but is implicitly USA, given this is an FDA submission.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

• This concept is not directly applicable in the context of this device's non-clinical testing. The "ground truth" (or reference standard) is based on engineering specifications, material properties, and established test methodologies (e.g., ASTM standards, previous FDA-cleared device performance).
• For the "Evaluation of Fit Testing," "qualified inspectors" performed the evaluation. Their qualifications (e.g., years of experience, specific certifications) are not detailed beyond "qualified".

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

• Not applicable. Adjudication methods are typically used in clinical studies involving multiple human readers interpreting medical images, where discrepancies need to be resolved. This document describes physical, non-clinical tests.

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

• Not applicable. This device is a physical implant, not an AI/ML diagnostic tool, and no human reader study was performed.

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

• Not applicable. This is not an AI algorithm. The manufacturing software is mentioned (same as predicate device), but its performance in terms of design output is assessed through the physical device tests (e.g., Evaluation of Fit), not as a standalone AI model.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

• The "ground truth" for these tests is based on:
  • Engineering Specifications/Standards: e.g., ASTM F2503-20 for MR compatibility.
  • Predicate Device Performance: For comparative strength, the performance of the legally marketed predicate device (K110684 PEEK implant) served as the benchmark.
  • Established Biomechanical Principles: For screw fixation, comparison to "axial pullout forces measured in prior testing of FDA-cleared neuro screws in an established cortical bone model" serves as the reference.
  • 3D Digital Models/Physical Prototypes: For "Evaluation of Fit," the 3D digital model of the implant and representative anatomical models served as the reference for fit.

8. The sample size for the training set

• Not applicable. This device is not an AI/ML system that requires a "training set" of data in the machine learning sense. The manufacturing process uses patient-specific CT scan data as input for design, but this is not a training set for an AI model.

9. How the ground truth for the training set was established

• Not applicable, as there is no AI/ML training set in this context.

Ask a specific question about this device

The MedCAD® AccuPlate® Patient-Specific Plate is intended for prescription use in oral and maxillofacial surgery, trauma and reconstructive surgery.

Specific Indications for Use:

• Primary mandibular reconstruction with bone graft
• Temporary bridging until delayed secondary reconstruction
• Secondary mandibular reconstruction
• Comminuted mandibular fractures
• Fractures of edentulous and/or atrophic mandibles
• Unstable mandibular fractures
• Maxillary reconstruction with or without bone graft
• Maxillary trauma

MedCAD® AccuPlate® Patient-Specific Plates are metal bone plates used in conjunction with commercially available metal bone screws for the fixation to bone, specifically in the areas of the mandible and maxilla. The design and dimensions of each plate within the envelope specification is based upon the patient's anatomical data (CT scan, CBCT scan, or MRI), and the intended anatomy to be fixated as determined from input provided by the surgeon. The plates are designed by MedCAD in consultation with the surgeon and manufactured by MedCAD only. MedCAD® AccuPlate® Patient-Specific Plates are not intended to be bent or modified in surgery. The MedCAD® AccuPlate® Patient-Specific Plates are manufactured from commercially pure titanium, are provided non-sterilized prior to use, and are intended for single use only. Plates are fastened to bone using commercially available bone screws with diameters ranging from 2.0 mm to 2.7 mm and lengths ranging from 4.0 mm to 23.0 mm.

This document is a 510(k) Premarket Notification from the FDA regarding the MedCAD® AccuPlate® Patient-Specific Plate. It is a medical device clearance document, not a study report for an AI/software device. Therefore, it does not contain the information typically presented for acceptance criteria and study results of an AI or software as a medical device (SaMD).

The document concerns a patient-specific bone plate used in oral and maxillofacial surgery. The "performance data" section refers to non-clinical functional and material testing (e.g., static and dynamic bending, screw pushout testing, sterilization, and biocompatibility validation) to demonstrate substantial equivalence to predicate devices, not performance metrics of an AI or software.

Therefore, I cannot provide the requested information regarding acceptance criteria and study results for an AI/software medical device because this document does not describe such a device or study.

Specifically:

1. A table of acceptance criteria and the reported device performance: The document mentions "All test method acceptance criteria were met" for sterilization validation and that mechanical tests showed the device "meets the performance of the predicate / reference devices." However, it does not provide specific acceptance values or performance metrics for these non-clinical tests in a table format that would be relevant to an AI/software device.
2. Sample sizes used for the test set and the data provenance: Not applicable to this type of device. The "test set" would refer to physical prototypes in mechanical testing, not a dataset for software.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. Ground truth for physical device performance is established through engineering specifications and standardized testing protocols.
4. Adjudication method for the test set: Not applicable.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done: Not applicable. This type of study is for evaluating human performance with and without AI assistance for clinical tasks, which is not the scope of this bone plate device.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done: Not applicable. This is not an algorithm.
7. The type of ground truth used: For physical performance testing, the ground truth is defined by engineering standards (e.g., ASTM F382 for bending, ISO 10993-1 for biocompatibility).
8. The sample size for the training set: Not applicable. This is a manufactured physical device, not an AI model.
9. How the ground truth for the training set was established: Not applicable.

Ask a specific question about this device

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:

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.

Ask a specific question about this device

The MedCAD AccuShape™ PEEK Patient Specific Cranial Implant is designed individually for each patient and intended to correct defects / replace bony voids in the cranial skeleton.

The MedCAD AccuShape™ PEEK Patient Specific Cranial Implant devices are individually sized and shaped implantable prosthetic cranioplasty plates intended to correct defects / replace voids in the cranial skeleton of a specific patient. The implants are designed using the patient's CT imaging data and precision manufactured from implantable grade polyether-ether-ketone (PEEK) material. The devices have a nominal thickness of 3mm, ranging from 2-5mm depending on the anatomical location. The device can be supplied as one or as multiple parts due to material constraints and/or the complexity of the device, with each part ranging in size from 10 x 10 (mm) to 200 x 200 (mm). The implants are provided with 2mm drainage holes spaced 10 mm apart from center to center with an edge margin of 10 mm. The devices are non-pyrogenic and are provided non-sterile for sterilization by the physician prior to implantation. The implants are attached to the native bone with commercially available cranioplasty fasteners.

This document describes the MedCAD AccuShape™ PEEK Patient Specific Cranial Implant, a device intended to correct defects or replace bony voids in the cranial skeleton. It is a Class II device (21 CFR 882.5330).

1. Table of Acceptance Criteria and Reported Device Performance

The provided text describes a 510(k) summary, which focuses on establishing substantial equivalence to predicate devices rather than setting specific performance acceptance criteria like diagnostic accuracy or sensitivity/specificity for an AI algorithm. In this context, "acceptance criteria" are related to demonstrating that the new device shares the same fundamental characteristics and safety/effectiveness profile as the legally marketed predicate devices.

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

This is not an AI/CADe type device that uses a "test set" in the conventional sense of evaluating an algorithm's performance on a dataset of patient images. Instead, this device is a patient-specific cranial implant. The "testing" involves demonstrating the physical and biological characteristics of the material and the manufacturing process, and confirming that the final product adheres to design specifications comparable to predicate devices.

• Sample Size for Non-Clinical Tests: Not explicitly stated as a number of patient cases. The description mentions "Biological laboratory tests including thorough sterilization validation have been conducted" and "Precision measurements have validated the dimensional accuracy and stability of the devices. In addition, material testing has been performed to demonstrate structural integrity." These are likely tests on material samples and manufactured devices, rather than a patient image test set.
• Data Provenance: Not applicable in the context of patient image data. The "data" here would be laboratory test results and manufacturing quality control data.

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

This is not applicable as there is no "test set" of patient cases for which an AI algorithm is generating a result that needs comparison to an expert-established ground truth. The device itself is the physical implant, designed and manufactured based on a patient's CT data. The "ground truth" for this type of device relates to engineering specifications, material properties, and biological compatibility.

4. Adjudication Method for the Test Set

Not applicable for a physical implant device where performance is established through non-clinical testing of materials and manufacturing processes, rather than an AI algorithm's output on a diagnostic test set.

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

Not applicable. This device is an implantable medical device, not an AI-assisted diagnostic or assistive tool for human readers interpreting medical images.

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

Not applicable. This is not an AI algorithm. Its design uses CAD (Computer-Aided Design) based on patient CT data, which is a common engineering process for custom implants, not an autonomous AI system delivering a diagnostic output.

7. The Type of Ground Truth Used

The "ground truth" for this device is based on engineering specifications, material science standards, and biological testing standards. Specifically:

• Confirmation of manufacturing precision and dimensional accuracy against design specifications.
• Validation of material properties (strength, stiffness, elasticity, density) against established PEEK material standards and predicate device characteristics.
• Demonstration of biocompatibility and sterility against recognized biological testing standards.

8. The Sample Size for the Training Set

Not applicable. This device does not involve machine learning or an AI algorithm that requires a "training set." The design process uses a patient's individual CT data, not a pre-trained model.

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

Not applicable, as there is no training set for this type of device.

Ask a specific question about this device