Search Results

The MRIdian Linac system, with magnetic resonance imaging capabilities, is intended to provide stereotactic radiosurgery and precision radiotherapy for lesions, tumors, and conditions anywhere in the body where radiation treatment is indicated.

This submission describes a modification to the currently marketed multi-leaf collimator (MLC) cleared as part of the MRIdian Linac system (K162393) intended to enhance beam resolution. The current MLC is comprised of 60 tungsten leaves in two opposing 30-leaf banks. The design modification discussed in this submission essentially divides the two existing leaf banks in half splitting them along the leaf motion direction and shifting the banks by one half of a leaf width. The "tonque and groove" is eliminated from the leaves and the shifted banks act effectively as better focused "tongue and groove" mechanism. As with the predicate, each MLC layer of the modified MLC is focused to a point beyond the source point of the Linac radiation to improve interleaf leakage performance. In this system, the focus point is 15 mm further away than the source point. The minimum projected leaf width then becomes one half the physical projected leaf width. The projected leaf width is 8.3 mm and the projected half leaf width is 4.15 mm. The modified MLC design has two layers of opposing tungsten leaves; the top and bottom layers have 34 and 35 leaves in each bank respectively for a total of 138 leaves. There have been no other changes to the cleared MRIdian Linac system (K162393).

My apologies. I cannot fulfill this request. The provided text is a US FDA 510(k) Premarket Notification document, which primarily focuses on demonstrating substantial equivalence of a modified medical device (MRIdian Linac System with 138-leaf Collimator) to a legally marketed predicate device (MRIdian Linac System with 60-leaf Collimator).

The document does not describe an AI/ML-driven device or study parameters typical for such devices, such as:

• Acceptance criteria related to AI performance metrics (e.g., sensitivity, specificity, AUC)
• Sample sizes for test sets in the context of AI model evaluation (e.g., number of images, patients)
• Data provenance for AI model training or testing (e.g., retrospective vs. prospective data cohorts, country of origin)
• Expert involvement in establishing ground truth for AI model testing
• Adjudication methods for AI performance evaluation
• MRMC comparative effectiveness studies with AI assistance
• Standalone AI algorithm performance studies
• Ground truth types specific to AI applications (e.g., pathology reports, long-term outcomes)
• Training set details for AI models

Instead, the document focuses on:

• Changes to a hardware component: The modification is to the Multi-Leaf Collimator (MLC), specifically increasing the number of leaves from 60 to 138 to improve beam resolution.
• Engineering and Safety Testing: The "Summary of Performance Testing" section details various IEC and ISO standards for electrical safety, EMC, usability, software lifecycle, and general medical device performance, verifying that the modified hardware system still meets these standards.
• Substantial Equivalence Argument: The core of the 510(k) is to show that despite the MLC change, the device's fundamental functionality and safety remain equivalent to the predicate.

Therefore, I cannot extract the requested information about AI model acceptance criteria and study design from this document because it is about a hardware modification to a radiation therapy system, not an AI/ML medical device.

Ask a specific question about this device