(126 days)
The KLS Martin Individual Patient Solutions (IPS) Planning 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 IPS Planning 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, guides, splints, and case reports for use in maxillofacial surgery. The IPS Planning System is also intended as a pre-operative software tool for simulating / evaluating surgical treatment options.
The KLS Martin Individual Patient Solutions (IPS) Planning System is a collection of software and associated additive manufacturing (rapid prototyping) equipment intended to provide a variety of outputs to support reconstructive and orthognathic surgeries. The system uses electronic medical images of the patients' anatomy (CT data) with input from the physician, to manipulate original patient images for planning and executing surgery. The system processes the medical images and produces a variety of patient specific physical and/or digital output devices which include anatomical models, guides, splints and case reports.
Here's an analysis of the acceptance criteria and study information based on the provided text, focusing on what is present and what is not:
The document (K181241 510(k) Summary) describes the KLS Martin Individual Patient Solutions (IPS) Planning System, which is a software system for image segmentation and pre-operative planning, and also provides physical outputs like anatomical models, guides, and splints for maxillofacial surgery. The submission focuses on demonstrating substantial equivalence to a predicate device (VSP System K120956) and uses several reference devices.
Acceptance Criteria and Study Information:
Based on the provided text, the device itself is a planning system that ultimately produces physical outputs. The "performance" being evaluated relates to the characteristics of these physical outputs (tensile strength, biocompatibility, sterility, pyrogenicity) and the software's functionality. There isn't a direct "device performance" metric in the traditional sense of an AI diagnostic device's sensitivity, specificity, or accuracy against a clinical outcome.
1. Table of Acceptance Criteria and Reported Device Performance
Given the nature of the device (planning system with physical outputs, not a diagnostic AI), the performance metrics are primarily related to safety and manufacturing quality.
| Acceptance Criteria Category | Specific Acceptance Criteria (as implied) | Reported Device Performance (as summarized) |
|---|---|---|
| Tensile & Bending | Polyamide guides maintain 85% of initial tensile strength after multiple sterilization cycles. Demonstrate a 6-month shelf life. Titanium devices are equivalent or better than traditional methods. | Polyamide guides met the criteria, demonstrating resistance to degradation after sterilization and supporting a 6-month shelf life. Titanium test results were leveraged from a reference device (K163579) and confirmed equivalence or superiority to traditional manufacturing. |
| Biocompatibility | Meet pre-defined acceptance criteria for cytotoxicity, sensitization, irritation, and chemical/material characterization (according to ISO 10993-1). | All conducted tests (cytotoxicity, sensitization, irritation, chemical/material characterization) for subject devices (polyamide and titanium) were within pre-defined acceptance criteria. Titanium results also leveraged from K163579. |
| Sterilization | Achieve a Sterility Assurance Level (SAL) of 10^-6 for each output device using the BI overkill method for steam sterilization (according to ISO 17665-1:2006 for dynamic-air-removal cycle). | All test method acceptance criteria were met, achieving the specified SAL of 10^-6. Validations for titanium were leveraged from K163579. |
| Pyrogenicity | Meet pyrogen limit specifications, with endotoxin levels below USP allowed limit for medical devices (according to AAMI ANSI ST72 for LAL endotoxin testing). | The devices contain endotoxin levels below the USP allowed limit for medical devices and meet pyrogen limit specifications. Testing for titanium was leveraged from K163579. |
| Software Verification & Validation (V&V) | All software requirements and specifications are implemented correctly and completely, traceable to system requirements. Conformity with pre-defined specifications and acceptance criteria based on risk analysis and impact assessments. Mitigation of potential risks and performance as intended based on user requirements. | Quality and on-site user acceptance testing provided objective evidence of correct and complete implementation of software requirements, traceability, and conformity with specifications and acceptance criteria. Software documentation demonstrated risk mitigation and intended performance. (Note: Specific quantitative metrics for software performance are not provided in this summary). |
2. Sample size used for the test set and the data provenance
- Test Set Sample Size: The document does not specify a "test set" in the context of a dataset for evaluating AI performance (e.g., medical images for segmentation accuracy). Instead, "testing" refers to non-clinical performance evaluations of the physical outputs and software.
- For Tensile & Bending, Biocompatibility, Sterilization, and Pyrogenicity, the sample sizes are not explicitly mentioned, but the tests were performed on "the subject polyamide guides" and "titanium" components. The provenance is internal testing performed by the manufacturer, or results leveraged from previous KLS Martin device submissions.
- For Software V&V, no specific numerical "test set" of software inputs is given, but testing was performed on "each individual software application."
- Data Provenance: The data provenance for non-clinical testing is internal to the manufacturer or relied upon previous regulatory submissions for similar materials/processes. It is not patient or country-specific data as would be for clinical studies. The data used by the IPS Planning System itself (CT data) would be patient data, but the evaluation here is of the system's outputs, not its interpretation of patient data in a diagnostic manner. The document states the system "transfers imaging information from a medical scanner such as a CT based system."
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- This question is not applicable in the context of this 510(k) submission. The "ground truth" for the non-clinical tests (tensile strength, biocompatibility, etc.) is established by standard scientific and engineering methodologies, not by expert medical review of images.
- For the "Software Verification and Validation," the "ground truth" for software functionality is defined by the established software requirements and specifications, validated by internal quality and user acceptance testing, not by external experts in the medical domain. The document mentions "input from the physician" for manipulation of original patient images, suggesting physicians set the clinical goals for the plan, but the validation of the system's performance is not described as involving experts establishing a "ground truth" concerning image interpretation.
4. Adjudication method for the test set
- Not applicable. There is no adjudication method described as would be used for clinical interpretation or diagnostic performance evaluation by multiple experts. The non-clinical tests follow established standards and protocols.
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, a multi-reader multi-case (MRMC) comparative effectiveness study was not done.
- This device is a planning system for maxillofacial surgery, not a diagnostic AI that assists human readers in image interpretation or diagnosis. It aids in surgical planning and creates physical outputs. The submission explicitly states: "Clinical testing was not necessary for the determination of substantial equivalence."
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Given the device's function, it is inherently a "human-in-the-loop" system. The description states: "The system uses electronic medical images of the patients' anatomy (CT data) with input from the physician, to manipulate original patient images for planning and executing surgery." And, "The physician provides input for model manipulation and interactive feedback through viewing of digital models...that are modified by the trained employee/engineer during the planning session."
- Therefore, performance of the algorithm without human intervention is not the intended use or focus of this submission. The "software verification and validation" (Section 11) is the closest thing to an "algorithm-only" evaluation, but it's about software functionality, not standalone image interpretation performance.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- For the non-clinical performance tests of the physical outputs (Tensile & Bending, Biocompatibility, Sterilization, Pyrogenicity), the "ground truth" is defined by established engineering and scientific standards (e.g., ISO 10993-1, ISO 17665-1:2006, AAMI ANSI ST72, and internal specifications).
- For Software Verification & Validation, the "ground truth" is adherence to predefined software requirements and specifications (functional and non-functional, related to image transfer, manipulation, and output file generation). It is not based on medical "ground truth" like pathology or clinical outcomes.
8. The sample size for the training set
- This question is not applicable. The KLS Martin IPS Planning System is described as using "validated commercially off-the-shelf (COTS) software applications" for image manipulation. There is no mention of a "training set" in the context of machine learning or AI model development within this summary. It appears to be a rule-based or conventional algorithmic system rather than a deep learning/machine learning model that would require a distinct training set.
9. How the ground truth for the training set was established
- This question is not applicable, as no training set for machine learning was mentioned or identified in the document (see point 8).
§ 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.