(87 days)
RayStation is a software system for radiation therapy and medical oncology. Based on user input, RayStation proposes treatment plans. After a proposed treatment plan is reviewed and approved by authorized intended users, RayStation may also be used to administer treatments.
The system functionality can be configured based on user needs.
RayStation is a treatment planning system for planning, analysis and administration of radiation therapy and medical oncology treatment plans. It has a modern user interface and is equipped with fast and accurate dose and optimization engines.
RayStation consists of multiple applications:
- . The main RayStation application is used for treatment planning.
- . The RayPhysics application is used for commissioning of treatment machines to make them available for treatment planning and used for commissioning of imaging systems.
- The RayTreat application is used for sending plans to treatment delivery devices for treatment and receiving records of performed treatments.
The device to be marketed, RayStation 11B, contains modified features compared to RayStation 11.0 as indicated below:
A simplified license configuration of RayStation is marketed as RayPlan. RayPlan has a limited set of purchasable licenses and some modules will not be accessible. RayPlan is marketed as RayPlan 11B.
EQD2 dose computation (new) - From photon and/or braction doses, it is possible to compute, deform and accumulate the two Gray equivalent (EQD2) dose. The computation of the EQD2 dose uses the biological linear quadratic model, which is also the basis for the already released biological optimization and evaluation functionality.
Generation of synthetic CT from CBCT (new) - Two new methods (algorithms) for synthetic CT generation will be included. The synthetic CT images are created by combining information in the CBCT image and a CT image for the specific patient to allow for dose computation using the HU values in the image, as for regular CT images. In RayStation 11.0 it is possible to compute dose on CBCT images for photons using bulk density assignments. The added functionality will improve the photon dose calculation accuracy on CBCT images. Handling of LET and other RBE components (new) - This functionality enables possibility to compute and evaluate the dose weighted LET (Linear Energy Transfer) for proton and light ion plans. LET is an additional dosimetric measure that can be used to assess the radiobiological effect of the proton and light ion radiation. Radiobiological equivalent (RBE) dose is a derived quantity with dependence on both the physical dose and the LET. In RayStation 11.0, it is possible to compute and evaluate RBE doses.
The provided text describes a 510(k) submission for RayStation 11B, a radiation therapy treatment planning system. However, the document does not contain the detailed information required to answer many of the questions about acceptance criteria and the study that proves the device meets them, especially regarding the performance of a machine learning component if one exists and its specific acceptance criteria.
The document states: "Related to machine learning, there is no change compared to the predicate device." This implies that while the RayStation product family might have machine learning components, RayStation 11B does not introduce new machine learning features that would necessitate specific performance studies for acceptance as part of this 510(k) submission. The focus of the provided text is on the validation of new dose computation methods and synthetic CT generation.
Therefore, many sections below will indicate "Information Not Provided in Text" or state that the question is not applicable given the document's content.
Acceptance Criteria and Reported Device Performance
Given the information provided, the acceptance criteria are largely related to the successful verification and validation of the new features (EQD2 dose computation, synthetic CT generation from CBCT, and handling of LET and other RBE components) demonstrating that the dose computations are "adequate for clinical use" and that the system "has met specifications and is as safe, as effective and performs as well as or better than the legally marketed predicate device."
Without specific numerical thresholds or performance metrics in the provided text, a formal table of acceptance criteria and reported performance cannot be fully constructed for these new features. The document suggests that the previous predicate device's performance benchmarks likely served as the implicit standard for "as well as or better than."
Implicit Acceptance Criteria and Reported Performance (based on text):
| Acceptance Criterion | Reported Device Performance |
|---|---|
| EQD2 dose computation is adequate for clinical use. | Validation for photon and/or braction doses were validated as part of the Clinical Evaluation for Brachy and User Site Validation demonstrates that the dose computation adequate for clinical use. |
| Improved photon dose calculation accuracy on CBCT images for synthetic CT from CBCT is adequate for clinical use. | Validation of improved photon dose calculation accuracy on CBCT images was performed in CBCT Conversion validation demonstrates that the dose computation adequate for clinical use. |
| Handling of LET and other RBE components for proton and light ion plans is adequate for clinical use. | Validation of dose weighted LET (Linear Energy Transfer) for proton and light ion plans were performed as part of the Proton PBS Monte Carlo validation. The validation demonstrates that the dose computation adequate for clinical use. |
| Overall system meets specifications and is as safe, as effective, and performs as well as or better than the predicate device. | "The data obtained from the verification show that system tests, unit and subsystem tests have passed, and the validations been completed successfully. The reviews of design, code and labeling are also passed." |
Study Information
-
A table of acceptance criteria and the reported device performance
See table above. Specific numerical thresholds are not provided in the text. -
Sample sizes used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Sample Sizes: Not explicitly stated for any of the validations (e.g., Clinical Evaluation for Brachy, CBCT Conversion validation, Proton PBS Monte Carlo validation). The text mentions "User Site Validation," which implies real-world data, but details on the number of cases or patients are absent.
- Data Provenance: Not specified (e.g., country of origin).
- Retrospective or Prospective: Not specified. "User Site Validation" might imply prospective or real-world use data, but this is not confirmed.
-
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)
- Number of Experts: Not specified.
- Qualifications of Experts: Not specified. Mention of "User Site Validation in cooperation with cancer clinics" suggests involvement of clinical professionals, but their specific roles, number, or qualifications are not detailed.
-
Adjudication method (e.g. 2+1, 3+1, none) for the test set
- Adjudication Method: Not specified.
-
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
- An MRMC study is not mentioned for the new features. The document explicitly states: "Related to machine learning, there is no change compared to the predicate device." This indicates that no new AI-assisted workflows requiring human reader improvement studies were part of this submission. The validation efforts focus on the accuracy of the new computational functions themselves, not human-AI interaction.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Yes, the validation activities described (e.g., "EQD2 dose computation (new) – Validation," "Generation of synthetic CT from CBCT (new) – Validation," "Handling of LET and other RBE components (new) - Validation") are, by their nature, evaluations of the algorithm's performance in generating accurate dose calculations or synthetic CTs. These are standalone evaluations of the new features. The "User Site Validation" is likely a real-world system test, but the core 'algorithm-only' performance is implied by the specific validation names.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- The term "ground truth" is not explicitly used, but the validation implies comparisons against established or expected physical and radiobiological models and potentially clinical standards of care.
- For dose computation, validation would typically involve comparing computed doses against known physics models, phantom measurements, or existing clinical systems deemed accurate.
- For synthetic CT, the "ground truth" would likely be the actual CT data that the synthetic CT is intended to replicate or approximate for dose calculation purposes.
- The text doesn't specify if clinical outcomes were used as ground truth for these specific validations.
- The term "ground truth" is not explicitly used, but the validation implies comparisons against established or expected physical and radiobiological models and potentially clinical standards of care.
-
The sample size for the training set
- Information Not Provided in Text. The document states "Related to machine learning, there is no change compared to the predicate device," and the new features are described as "algorithms" or "functionality," not explicitly machine learning models that would require a distinct training set for this submission. If the original RayStation product had ML components, their training set information is not part of this document.
-
How the ground truth for the training set was established
- Information Not Provided in Text. (See point 8).
In summary, the provided FDA 510(k) summary focuses on demonstrating the substantial equivalence of RayStation 11B to its predicate device by verifying and validating new computational features rather than new or modified AI/ML features requiring human-in-the-loop or specific training/test set performance metrics commonly associated with AI/ML device clearances. The acceptance criteria are implicitly tied to the successful completion of these verification and validation activities, proving the new functions are "adequate for clinical use" and the system performs "as well as or better than" the predicate.
§ 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.