(182 days)
The ViewRay (MRIdian) Linac System, with magnetic resonance imaqinq capabilities, is intended to provide stereotactic radiosurgery and precision radiotherapy for lesions, tumors, and conditions anywhere in the body where radiation treatment is indicated.
The MRIdian Linac system delivers ionizing radiation using a magnetic resonance imaging system (MRIS) unit for image quidance and a 6MV linear accelerator to deliver radiation therapy. The system is designed so that the imaging and radiotherapy fields of view coincide permitting imaging of the patient at the radiotherapy isocenter before and during treatment. The MRIdian Linac system is used with the ViewRay Treatment Planning and Delivery System (TPDS) (K102915). As with the predicate MRIdian System (K111862), the MRIdian Linac System consists of three primary subsystems: The Treatment Planning and Delivery System (TPDS), The Magnetic Resonance Imaging System (MRIS), and The Radiation Therapy Delivery System (RDS). These three subsystems are designed to operate concurrently for accurate targeted administration of radiation therapy.
The provided document is a 510(k) summary for the ViewRay (MRIdian) Linac System (K162393), which is a medical device for radiation therapy. The purpose of this document is to demonstrate "substantial equivalence" to a predicate device, not necessarily to detail comprehensive acceptance criteria and a study to prove meeting them in the way a clinical trial or performance study for a novel device would.
Based on the provided text, the device itself is a radiation therapy system, not an AI/ML algorithm. Therefore, many of the typical AI/ML-specific questions (like sample size for test set, number of experts for ground truth, adjudication method, MRMC study, standalone performance) are not directly applicable or explicitly stated in this type of regulatory submission for this device. The "performance" being described is primarily related to the physical and functional characteristics of the radiation delivery system.
However, I can extract information related to demonstrating substantial equivalence and the performance characteristics presented.
Here's an interpretation based on the document:
1. Table of Acceptance Criteria and Reported Device Performance
The document compares the "Device with Change" (K162393, the MRIdian Linac System) to the "Cleared Device" (K111862, the predicate MRIdian System). The "acceptance criteria" can be inferred from the predicate device's performance, as the goal is to show the new device is "substantially equivalent." Where specific quantitative metrics are provided, those can be considered the performance reported for each device.
| Feature | Acceptance Criteria (Predicate: K111862) | Reported Device Performance (K162393) |
|---|---|---|
| Radiation Source | Cobalt-60 Sources (qty. 3) | 6MV Linear Accelerator |
| Beam | 2.0 cm dia. Cobalt 60 Gamma Ray Source, 1.332 & 1.172 MeV | 6 MV Bremsstrahlung X-Rays produced by Linear Accelerator |
| Max Dose Rate | 600 cGy/min. total (200 cGy/min. per head at installation) at Dmax at 105cm isocenter for a 10.5 cm x 10.5 cm field (three sources utilized) | 600 cGy/min. at Dmax at a 90 cm isocenter for a 10 cm x 10 cm field (Single Source) |
| Static Dose Accuracy | 90% of points evaluated in a treatment volume pass a relative gamma criteria of 3%/3mm and a high dose, low gradient absolute point measurement is within 5% of the planned dose (per AAPM TG 119) | 90% of points evaluated in a treatment volume pass a relative gamma criteria of 3%/3mm and a high dose, low gradient absolute point measurement is within 5% of the planned dose (per AAPM TG 119) |
| Moving Target Dose Accuracy | Dose delivery on a moving target is consistent within ≤ 2% to that of a stationary target with real time tumor tracking (RealTarget). | Dose delivery on a moving target is consistent within ≤ 2% to that of a stationary target with real time tumor tracking (RealTarget). |
| Collimation | Field shaping, Multi Leaf Collimator (MLC), Quantity of 3 | Field shaping, Multi Leaf Collimator (MLC), Quantity of 1 |
| Range of MLC collimated beam size | 1.05cm x 1.05cm to 27.3cm x 27.3cm projected at isocenter | 0.72 cm x 1.43 cm to 25.71 cm x 25.71 cm projected at isocenter |
| Number of leaves per MLC | 60 | 60 |
| MLC material | Tungsten Alloy | Tungsten Alloy |
| Isocenter distance | 105 cm | 90 cm |
| Isocenter accuracy | 0.5mm radius (1 mm diameter) | 0.5mm radius (1 mm diameter) |
| Minimum Room Dimensions (Height/Length/Width) | 2.9 m x 7.6 m x 5.9 m | 2.9 m x 7.6 m x 5.9 m |
| Environment Line Voltage | 380-480V | 480V |
| Ambient Room Temp. | 65 °F to 72 °F | 65 °F to 72 °F |
| Relative Humidity | 40 to 60% | 40 to 60% |
| Power Distribution Isolation | Transformer | Transformer |
| Radiation Head Shielding | Depleted Uranium and Tungsten Alloy shield with stainless steel shell, 15,000 Curies max. capacity | Lead, Tungsten Alloy, and Steel shielding |
| Source control mechanism | Redundant timers controlling pneumatically driven linear source movement mechanisms | Redundant ion chambers and dose monitoring cards |
| Radiation Leakage when OFF | In the fully shielded BEAM OFF position, measured at survey points, is in accordance with NCRP #102. | Not applicable, no leakage when OFF |
| Radiation Transmission through head | With the source in the fully exposed BEAM ON position is less than 0.1% of the primary beam. | Less than 0.1% of the primary beam. |
| Method of IMRT | MLC based cone-beam delivery | MLC based cone-beam delivery |
| Gantry | Ring Gantry, collision with patient not possible | Ring Gantry, collision with patient not possible |
| Motion synchronized treatment | Yes | Yes |
| Integrated imaging for planning, positioning, gating | Magnetic resonance imaging system | Magnetic resonance imaging system |
| MR Physical Characteristics | (Consistent with predicate for Bore Diameter, DSV, Patient table degrees of freedom, MRI Frequency, Field Strength, Field of View, Field Homogeneity, Field Stability, 3D Imaging Volumes/Resolution, 2D Imaging Planes/Resolution, Geometric Accuracy, Signal to Noise) | (Consistent with predicate for Bore Diameter, DSV, Patient table degrees of freedom, MRI Frequency, Field Strength, Field of View, Field Homogeneity, Field Stability, 3D Imaging Volumes/Resolution, 2D Imaging Planes/Resolution, Geometric Accuracy, Signal to Noise) |
| Temporal Integrity | 0.01s or better | 0.01s or better |
| Signal to Noise | 30 | 30 |
| Dose per treatment | None | None |
| Treatment Planning and Delivery System Dose Algorithm (K102915) | Monte Carlo Dose Computation Radiation Source Model for Cobalt photons. Measurement shows fundamental radiation-transport algorithm works correctly. | Monte Carlo Dose Computation Radiation Source Model for Bremsstrahlung X-Rays, fundamental radiation-transport algorithm is unchanged. Measurement shows fundamental radiation-transport algorithm works correctly. |
| Dose Output Modeling | Dose output modeled with beam-on time | Dose output modeled with monitor units |
| Dose Display | Display of Cobalt delivery parameters | Display of Linac delivery parameters |
2. Sample size used for the test set and the data provenance:
- Sample Size: Not explicitly stated as a number of patients or cases. The "testing" referred to appears to be engineering and physics verification tests on the device itself, rather than studies involving patient data or a "test set" of images/cases in the AI/ML sense. The testing verified conformance to design requirements and safety standards.
- Data Provenance: Not applicable in the context of this device's validation as described (i.e., not a data-driven AI/ML device in the document). The testing is on the physical system's performance.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable/mentioned. "Ground truth" in this context refers to established engineering/physics standards (e.g., AAPM TG 119) and design specifications, not expert interpretations of medical images for an AI/ML diagnostic or therapeutic algorithm.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Not applicable/mentioned. This refers to consensus methods for expert labeling, which is not relevant for the type of device validation presented here.
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. The device is a radiation therapy system, not an AI-assisted diagnostic or therapeutic interpretation tool requiring human-in-the-loop performance measurement.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Not applicable as an "algorithm-only" performance in the AI/ML sense. The "standalone" performance is the device's operational characteristics as detailed in the table (e.g., dose accuracy, geometric accuracy). The document states, "Testing executed on the System verified conformance to design requirements and ensured all identified risks and hazards were mitigated, and demonstrated conformance to relevant safety standards. The MRIdian Linac system described in this premarket notification passed all verification testing..."
7. The type of ground truth used (expert concensus, pathology, outcomes data, etc.):
- The "ground truth" for the performance characteristics are established engineering and physics standards, such as:
- AAPM TG 119 for Static Dose Accuracy.
- IEC 60601-1, IEC 60601-1-2, IEC 60601-2-33, IEC/EN 60601-2-1, IEC/EN 60976, IEC 60601-1-6, IEC 61217, IEC 62083, EN 62304, EN 62366, ISO 10993-1.
- The design specifications of the ViewRay (MRIdian) System (K111862) serve as the benchmark for substantial equivalence.
8. The sample size for the training set:
- Not applicable. This is not an AI/ML device that uses a "training set" in the conventional sense. The "training" for the device's design implicitly comes from decades of medical linear accelerator and MRI technology development, and experience with the predicate device.
9. How the ground truth for the training set was established:
- Not applicable. As this is not an AI/ML device with a training set, the concept of establishing ground truth for it is not relevant here.
§ 892.5050 Medical charged-particle radiation therapy system.
(a)
Identification. A medical charged-particle radiation therapy system is a device that produces by acceleration high energy charged particles (e.g., electrons and protons) intended for use in radiation therapy. This generic type of device may include signal analysis and display equipment, patient and equipment supports, treatment planning computer programs, component parts, and accessories.(b)
Classification. Class II. When intended for use as a quality control system, the film dosimetry system (film scanning system) included as an accessory to the device described in paragraph (a) of this section, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9.