Search Results

Software: The Precision AI Planning Software is intended to be used as a pre-surgical planner for simulation of surgical interventions for shoulder joint arthroplasty. The software is used to assist in the positioning of shoulder components by creating a 3D bone construct of the joint and allows the surgeon to visualize, measure, reconstruct, annotate and edit presurgical plan data. The software leads to the generation of a surgery report along with a pre-surgical plan data file which can be used as input data to design the Precision AI Shoulder Guide and Biomodels.

Hardware: The Precision AI Planning System Guides and Biomodels are intended to be used as patient-specific surgical instruments to assist in the intraoperative positioning of shoulder implant components used with total and reverse shoulder arthroplasty by referencing anatomic landmarks of the shoulder that are identifiable on preoperative CT-imaging scans. The Glenoid Guide is used to place the k-wire and the Humeral Guide is used to place humeral pins for humeral head resection. The Precision AI Guides and Biomodels are indicated for single use only. The Precision AI Surgical Planning System is indicated for use on adult patients that have been consented for shoulder joint arthroplasty. Both humeral and glenoid guides are suitable for a delto-pectoral approach only. The Precision AI Surgical Planning System is indicated for total and reverse shoulder arthroplasty using the following implant systems and their compatible components: Enovis and Lima.

The Precision AI Surgical Planning System is a patient-specific medical device that is designed to be used to assist the surgeon in the placement of shoulder components during total anatomic and reverse shoulder replacement surgery. This can be done by generating a pre-surgical shoulder plan and, if requested by the surgeon, by manufacturing a patient-specific guides and models to transfer the plan to surgery. The subject device is a system composed of the following: The Precision AI Surgical Planning System Software will create a 3D construct/render of the patient's shoulder joint for the surgeon to plan the operatively then create a physical Patient Specific Instrument (or Guide), using 3D printing by selective laser sintering. The patient's CT scan images are the design input for this to be created and are auto segmented via a locked, or static, artificial intelligence algorithm. The surgeon can visualise the deformity of the diseased joint, on this 3D render and CT scan images, and determine the inherent deformity of the joint. They are then able to virtually place the artificial implants in an optimal position to correct the measured deformity for that specific patient. The Precision AI Guides, which are a patient-specific guide and models that are based on a pre-surgical plan. This pre-surgical plan is generated using the software component. Patient-specific guide and models will be manufactured if the surgeon requests patient-specific guides to transfer the plan to surgery.

The provided text is an FDA 510(k) summary for the Precision AI Surgical Planning System (PAI-SPS). It describes the device and its intended use, but it does NOT contain specific acceptance criteria, detailed study designs, or performance results in terms of metrics like sensitivity, specificity, accuracy, or effect sizes for human reader improvement.

The document states that "Software verification and validation were performed, and documentation was provided following the 'Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices.' This includes verification against defined requirements and validation against user needs." It also mentions that "Design verification and validation testing demonstrated that the PAI-SPS meets all design requirements and is as safe and effective as its predicate device (K233992)."

However, it does not explicitly lay out the acceptance criteria (e.g., "model must achieve 90% accuracy") or the results against those criteria. It focuses more on demonstrating substantial equivalence to a predicate device (PAI-SPS K233992) by showing that the technology and intended use are similar, with the main differences being the addition of compatibility with more implant systems.

Therefore, I cannot fully complete the requested table and answer all questions based solely on the provided text. I will provide information based on what is available and indicate where information is missing.

Description of Acceptance Criteria and Study to Prove Device Meets Criteria

Based on the provided FDA 510(k) summary, the PAI-SPS device is being cleared primarily through demonstrating substantial equivalence to an already cleared predicate device (PAI-SPS, K233992). The key claim for equivalence rests on similar intended use, fundamental scientific technology, design, functionality, operating principles, and materials, with the primary difference being expanded compatibility with additional implant systems.

The document implicitly suggests that the "acceptance criteria" are tied to demonstrating that these technological differences "do not raise any different questions of safety and effectiveness." The studies cited are primarily focused on software verification and validation, ensuring the new compatible implant systems do not negatively impact the established safety and effectiveness.

Here is a breakdown of the available information:

1. Table of Acceptance Criteria and Reported Device Performance

Note: The provided document does not explicitly state quantitative acceptance criteria (e.g., minimum accuracy percentages, specific error bounds) or detailed performance metrics. The performance is largely framed as demonstrating that the device "meets all design requirements and is as safe and effective as its predicate device."

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

• The document does not specify the sample size for any test set (e.g., for software validation or hardware accuracy).
• Data Provenance: Not explicitly stated for specific test sets. Given the company is "Precision AI Pty Ltd" in Australia, and the document discusses "previous cadaver testing and composite bone model testing," it's likely a mix of lab-based/simulated data and potentially some retrospective clinical imaging data for initial AI development/testing, but this is not detailed. The document implies that new testing was not extensively conducted for this submission, relying heavily on the predicate device's prior validation and the minor changes to compatibility.

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

• The document does not specify the number of experts or their qualifications used to establish ground truth for any test set.
• It mentions that the software allows a "qualified surgeon" to approve pre-surgical plan data, implying that expert surgical review is part of the workflow.

4. Adjudication Method for the Test Set

• The document does not specify any adjudication method (e.g., 2+1, 3+1) for establishing ground truth or evaluating test results.

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

• No, an MRMC comparative effectiveness study was not explicitly mentioned or described. The focus of this 510(k) is substantial equivalence based on technological similarity and expanded compatibility, not a comparative study against human readers or performance improvement with AI assistance.

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

• The document states that the AI algorithm performs "auto segmentation via a locked, or static, artificial intelligence algorithm." While this indicates a standalone AI component, the document does not provide standalone performance metrics for this AI segmentation. The overall system is described as a "pre-surgical planner" where the surgeon can "visualize, measure, reconstruct, annotate and edit pre-surgical plan data," suggesting a human-in-the-loop workflow.

7. The Type of Ground Truth Used

• For software, the implicit ground truth appears to be expert consensus or approved surgical plans for judging the accuracy of the software's representations and planning capabilities. The document states "The software allows a qualified surgeon to visualize, measure, reconstruct, annotate, edit and approve pre-surgical plan data."
• For hardware, "previous cadaver testing and composite bone model testing" were used, implying physical measurements against a known standard or "true" position established in these models.

8. The Sample Size for the Training Set

• The document does not specify the sample size used for the training set for the AI segmentation algorithm.

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

• The document does not specify how the ground truth for the AI training set was established. It only mentions that the AI algorithm for auto-segmentation is "locked, or static," implying it was trained previously.

Ask a specific question about this device

Hardware:
The Materialise Shoulder Guide and Models are intended to be used as a surgical instrument to assist in the intraoperative positioning of glenoid components used with total and reverse shoulder arthroplasty by referencing anatomic landmarks of the shoulder that are identifiable on preoperative CT-imaging scans.

The Materialise Shoulder Guide and Models are single use only.

The Materialise Shoulder Guide and Models can be used in conjunction with the following total and reverse shoulder implants systems and their respective compatible components:

• · DePuy Synthes'
• GLOBAL® APG+ Shoulder System (K052472)
• DELTA XTEND™ Reverse Shoulder System (K120174, K062250, K183077, K203694)
• GLOBAL® STEPTECH® APG Shoulder System (K092122)
• INHANCE™ Anatomic Shoulder System (K202716)1
• INHANCE™ Reverse Shoulder System (K212737)
• INHANCE™ Hybrid Anatomic Glenoid Implant (K212933)
• INHANCE™ Reverse Glenoid Peripheral Posts (K221467)
• INHANCE Convertible Glenoid (K230831)
• · Enovis'2 (DJO)
• Reverse® Shoulder Prosthesis (K051075, K111629, K092873)
• Turon® Shoulder System (K080402)
• AltiVate™ Anatomic Shoulder System (K162024)
• AltiVate™ Anatomic Augmented Glenoid (K213387, K222592)
• AltiVate™ Reverse Glenoid (K233481)
• · Smith+Nephew's3
• Titan™ Total Shoulder System (K100448, K112438, K142413, K152047)
• Titan™ Reverse Shoulder System (K130050, K161189, K173717, K181999)
• AETOS Total Shoulder System (K220847, K230572)
• AETOS Reverse Shoulder System (K220847, K230572)
• · Lima's
• SMR™ Shoulder System (K100858)
• SMR™ Reverse Shoulder System (K110598)
• SMR™ Modular Glenoid (K113254) (K143256)
• SMR™ 3-Pegs Glenoid (K130642)
• SMR™ TT Metal Back Glenoid (K133349)
• SMR TM 40mm Glenosphere (K142139)
• SMR™ TT Augmented 360 Baseplate (K220792)
• SMR™ TT Hybrid Glenoid (K220792)
• PRIMA TT Glenoid (K222427)

Software:
SurgiCase Shoulder Planner is intended to be used as a pre-surgical planner for simulation of surgical interventions for shoulder orthopedic surgery. The software is used to assist in the positioning of shoulder components. SurgiCase Shoulder Planner allows the surgeon to visualize, measure, reconstruct, annotate and edit pre-surgical plan data. The software leads to the generation of a surgery report along with a pre-surgical plan data file which can be used as input data to design the Materialise Shoulder Guide and Models.

Materialise Shoulder System™ is a patient-specific medical device that is designed to be used to assist the surgeon in the placement of shoulder components during total anatomic and reverse shoulder replacement surgery. This can be done by generating a pre-surgical shoulder plan and, if requested by the surgeon, by manufacturing a patient-specific glenoid guide and models to transfer the glenoid plan to surgery. The device is a system composed of the following:

• a software component, branded as SurgiCase Shoulder Planner. This software is a planning tool used . to generate a pre-surgical plan for a specific patient.
• Materialise Shoulder Guide and Models, which are a patient-specific quide and models that are based ● on a pre-surgical plan. This pre-surgical plan is generated using the software component. Patientspecific glenoid guide and models will be manufactured if the surgeon requests patient-specific guides to transfer the glenoid plan to surgery. The Materialise Shoulder Guide is designed and manufactured to fit the anatomy of a specific patient. A bone model of the scapula is delivered with the Materialise Shoulder Guide. A graft model can be delivered with the Materialise Shoulder Guide. The graft model visualizes the graft-space between implant and bone, based on the pre-operative planning of the surgeon. The graft model serves as a visual reference for the surgeon in the OR.

The provided FDA 510(k) summary (K242813) for the Materialise Shoulder System™ describes a submission seeking substantial equivalence to a previously cleared device (K241143). This submission is primarily for adding compatibility with new implant components rather than introducing a completely new AI capability or significant software change that would necessitate extensive new performance data. Therefore, the document does not contain the detailed information typically found in a study proving a device meets acceptance criteria for an AI/ML product.

Specifically, the document states:

• "The non-clinical performance data has demonstrated that the subject software technological differences between the subject and predicate devices do not raise any different questions of safety and effectiveness." (Page 9)
• "Software verification and validation were performed, and documentation was provided following the 'Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices.' This includes verification against defined requirements, and validation against user needs." (Page 10)
• "Previous testing for biocompatibility, sterility, cleaning, debris, dimensional stability and packaging are applicable to the subject device. Testing verified that the accuracy and performance of the system is adequate to perform as intended. The stability of the device placement, surgical technique, intended use and functional elements of the subject device are the same as that of the predicate device of Materialise Shoulder System™ K241143 and previously cleared devices... therefore previous simulated surgeries using rapid prototyped bone models and previous cadaver testing on previously cleared devices K153602 and K131559 are considered applicable to the subject device." (Page 10)

Given this, I cannot provide detailed answers to many of your questions as the submission relies on the substantial equivalence principle and prior testing rather than new, extensive performance studies for AI/ML.

However, I can extract what is available:

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

The document does not provide a specific table of quantitative acceptance criteria and reported device performance for the current submission (K242813), as it relies on the previous clearance and the assessment that the changes (adding implant compatibility) do not raise new safety or effectiveness concerns.

The general acceptance criterion mentioned is that the "accuracy and performance of the system is adequate to perform as intended." This was verified through previous testing, including "simulated surgeries using rapid prototyped bone models and previous cadaver testing."

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

The document mentions "previous simulated surgeries using rapid prototyped bone models and previous cadaver testing on previously cleared devices K153602 and K131559." It does not specify the sample size for these tests, nor the country of origin of the data or whether it was retrospective or prospective.

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 provided in the document.

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

This information is not provided in the document.

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 MRMC study is mentioned. The device is a "pre-surgical planner" and "surgical instrument" designed to assist the surgeon, but the provided text does not detail comparative effectiveness studies of human readers (surgeons) with and without the AI (planning software) assistance.

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

The software (SurgiCase Shoulder Planner) generates a pre-surgical plan which the "qualified surgeon" can "visualize, measure, reconstruct, annotate, edit and approve" (Page 9). This indicates a human-in-the-loop process. Standalone performance of the algorithm without human interaction is not discussed as it's not the intended use.

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

The document generally refers to "verification against defined requirements, and validation against user needs" and "accuracy and performance of the system is adequate to perform as intended" based on "simulated surgeries using rapid prototyped bone models and previous cadaver testing." This suggests a ground truth established through expert-defined surgical planning parameters and comparison to physical outcomes in the simulated/cadaveric environment, but specifics on how this ground truth was formalized (e.g., expert consensus on optimal planning, precise measurement validation) are not detailed.

8. The sample size for the training set

This device is a surgical planning tool and guides, not a deep learning AI model that requires a "training set" in the conventional sense for image classification or similar tasks. It is based on algorithms that process CT-imaging scans and anatomical landmarks to generate personalized plans and guides. Therefore, the concept of a "training set" for AI/ML is not applicable here in the way it would be for a pattern recognition AI. The software's robustness and accuracy are likely validated through extensive testing against various patient anatomies and surgical scenarios.

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

As explained above, the concept of a training set as typically understood for AI/ML models is not directly applicable to this device based on the provided information.

Ask a specific question about this device

Hardware:
The Materialise Shoulder Guide and Models are intended to be used as a surgical instrument to assist in the intraoperative positioning of glenoid components used with total and reverse shoulder arthroplasty by referencing anatomic landmarks of the shoulder that are identifiable on preoperative CT-imaging scans.
The Materialise Shoulder Guide and Models are single use only.
The Materialise Shoulder Guide and Models can be used in conjunction with the following total and reverse shoulder implants systems and their respective compatible components:

• DePuy Synthes'
• GLOBAL® APG+ Shoulder System (K052472)
• DELTA XTEND™ Reverse Shoulder System (K120174, K062250, K183077, K203694)
• GLOBAL® STEPTECH® APG Shoulder System (K092122)
• INHANCE™ Anatomic Shoulder System (K202716)1
• INHANCE™ Reverse Shoulder System (K212737)
• INHANCE Hybrid Anatomic Glenoid Implant (K212933)
• INHANCE Reverse Glenoid Peripheral Posts (K221467)
• Enovis'2 (DJO)
• Reverse® Shoulder Prosthesis (K051075, K111629, K092873)
• Turon® Shoulder System (K080402)
• AltiVate™ Anatomic Shoulder System (K162024)
• AltiVate™ Anatomic Augmented Glenoid (K213387)
• Smith+Nephew's3
• Titan™ Total Shoulder System (K100448, K112438, K142413, K152047)
• Titan™ Reverse Shoulder System (K130050, K161189, K173717, K181999)
• AETOS Total Shoulder System (K220847, K230572)
• AETOS Reverse Shoulder System (K220847, K230572)
• Lima's
• SMR™ Shoulder System (K100858)
• SMR™ Reverse Shoulder System (K110598)
• SMR™ Modular Glenoid (K113254) (K143256)
• SMR™ 3-Pegs Glenoid (K130642)
• SMR™ TT Metal Back Glenoid (K133349)
• SMR™ 40mm Glenosphere (K142139)
• SMR™ TT Augmented 360 Baseplate (K220792)
• SMR™ TT Hybrid Glenoid (K220792)
• PRIMA TT Glenoid (K222427)

Software:
SurgiCase Shoulder Planner is intended to be used as a pre-surgical planner for simulation of surgical interventions for shoulder orthopedic surgery. The software is used to assist in the positioning of shoulder components. SurgiCase Shoulder Planner allows the surgeon to visualize, measure, reconstruct, annotate and edit pre-surgical plan data. The software leads to the generation of a surgery report along with a pre-surgical plan data file which can be used as input data to design the Materialise Shoulder Guide and Models.

Materialise Shoulder System™ is a patient-specific medical device that is designed to be used to assist the surgeon in the placement of shoulder components during total anatomic and reverse shoulder replacement surgery. This can be done by generating a pre-surgical shoulder plan and, if requested by the surgeon, by manufacturing a patient-specific glenoid guide and models to transfer the glenoid plan to surgery. The device is a system composed of the following:

• a software component, branded as SurgiCase Shoulder Planner. This software is a planning tool used to generate a pre-surgical plan for a specific patient.
• Materialise Shoulder Guide and Models, which are a patient-specific guide and models that are based on a pre-surgical plan. This pre-surgical plan is generated using the software component. Patient-specific glenoid guide and models will be manufactured if the surgeon requests patient-specific guides to transfer the glenoid plan to surgery. The Materialise Shoulder Guide is designed and manufactured to fit the anatomy of a specific patient. A bone model of the scapula is delivered with the Materialise Shoulder Guide. A graft model can be delivered with the Materialise Shoulder Guide. The graft model visualizes the graft-space between implant and bone, based on the pre-operative planning of the surgeon. The graft model serves as a visual reference for the surgeon in the OR.

The provided text describes a 510(k) submission for the Materialise Shoulder System™, Materialise Shoulder Guide and Models, and SurgiCase Shoulder Planner. It indicates that this is a special 510(k) submission, meaning it's for a modification to a previously cleared device. Therefore, much of the performance data refers back to the predicate device and prior clearances.

Here's an analysis of the acceptance criteria and supporting study information based on the provided text:

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

The document does not explicitly state quantitative acceptance criteria or a direct table showing "acceptance criteria vs. reported device performance" for this specific 510(k) submission. Instead, for this special 510(k), the performance data mainly focuses on demonstrating that the changes (addition of new implant components to the software and hardware compatibility) do not raise new questions of safety and effectiveness compared to the predicate device.

The "performance data (non-clinical)" section highlights that:

• Hardware: Previous testing for biocompatibility, cleaning, debris, dimensional stability, and packaging is applicable. Accuracy and performance of the system were "adequate to perform as intended." Previous simulated surgeries and cadaver testing on earlier cleared devices are considered applicable.
• Software: Software verification and validation were performed "against defined requirements" and "against user needs," following FDA guidance.

Since this is a special 510(k) for an incremental change (adding compatibility with specific new implants), it's implied that the acceptance criteria are met if these additions do not negatively impact the established safety and effectiveness of the existing device, and the software development process meets regulatory standards.

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Hardware (previous testing cited): The document mentions "previous simulated surgeries using rapid prototyped bone models" and "previous cadaver testing on previously cleared devices K153602 and K131559." It does not specify the sample size for these previous studies (e.g., number of bone models or cadavers) or their provenance (country of origin), nor does it state if they were retrospective or prospective.
• Software (verification and validation): The document states "Software verification and validation were performed," but does not specify a separate "test set" in the context of clinical data or specific performance metrics with sample sizes for this particular submission. The V&V activities would involve testing against requirements and user needs, which could include various test cases and scenarios, but these are not quantified here as a "test set" size.

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 provided in the document. The studies cited for hardware ("simulated surgeries" and "cadaver testing") and software ("verification and validation") do not detail the involvement of experts in establishing ground truth, their number, or specific qualifications. The software's function is to assist surgeons in planning, implying surgeon input in its use, but not explicitly in establishing a ground truth for a test set described in this submission.

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

This information is not provided in the document.

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

The document does not mention any MRMC comparative effectiveness studies. The device (SurgiCase Shoulder Planner software component and Materialise Shoulder Guide and Models hardware component) is designed to assist surgeons in planning and component positioning, but the provided text does not contain data on whether human readers/surgeons improve with or without this specific AI assistance or effect sizes.

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

The document states that the SurgiCase Shoulder Planner is "intended to be used as a pre-surgical planner for simulation of surgical interventions for shoulder orthopedic surgery. The software is used to assist in the positioning of shoulder components." It also mentions, "SurgiCase Shoulder Planner allows the surgeon to visualize, measure, reconstruct, annotate and edit pre-surgical plan data." This indicates that the software is a human-in-the-loop device, where the surgeon is actively involved in the planning process and responsible for approving the plan. Therefore, a standalone (algorithm only) performance assessment, without human input, is unlikely to be the primary method of evaluation described or required for this type of device. The document does not provide such standalone performance data.

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

The document does not explicitly state the type of ground truth used for any specific test set related to this submission. For the hardware (guides and models), the "accuracy and performance" implies a comparison to a known standard or ideal, perhaps derived from anatomical models or surgical goals. For the software, "verification against defined requirements, and validation against user needs" suggests that the ground truth for V&V would be the successful adherence to these requirements and user expectations, which could involve internal expert review or adherence to pre-defined medical/engineering specifications. However, specific types of ground truth like pathology or long-term outcomes data are not mentioned.

8. The sample size for the training set

This information is not provided in the document. As the submission is for a special 510(k) updating compatibility, it's possible that the core algorithms were developed and trained previously, and details of their original training are not part of this specific submission. The focus here is on the impact of the changes to the device.

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

This information is not provided in the document. Similar to the training set size, the specifics of how the ground truth was established for the original training of any underlying algorithms are not included in this special 510(k).

Ask a specific question about this device