(48 days)
The Oncentra system is a radiation treatment planning software designed to analyze and plan radiation treatments in three dimensions for the purpose of treating patients with cancer. The treatment plans provide estimates of dose distributions expected during the proposed treatment, and may be used to administer treatments after review and approval by qualified medical personnel.
The Oncentra system is a radiation treatment planning software designed to analyze and plan radiation treatments in three dimensions for the purpose of treating patients with cancer. The treatment plans provide estimates of dose distributions expected during the proposed treatment, and may be used to administer treatments after review and approval by qualified medical personnel. The system is intended for use by qualified medical personnel in radiotherapy clinics, suitably trained by Nucletron staff (or other competent people) in using the system. The Oncentra system should be configured locally and maintained by radiation physicists. It is a requirement that the person responsible for the local configuration has been suitably trained in configuring and maintaining the system. Oncentra 4.2 uses externally acquired medical images and user input. The Oncentra 4.2 software is based on a modular client/server design, with the treatment planning functions divided into "Activities".
This document describes the Oncentra 4.2 radiation treatment planning software. However, the provided text does not contain sufficiently detailed information about "acceptance criteria" and a "study that proves the device meets the acceptance criteria" in the format requested.
The document primarily focuses on demonstrating substantial equivalence to a predicate device (Oncentra MasterPlan 3.1, K081281) for regulatory purposes. It mentions "dosimetric validation" for brachytherapy and external beam calculations, which are relevant to performance, but it does not specify quantitative acceptance criteria or a formal study design that would typically be expected for a detailed evaluation of device performance against specific targets.
Here's an analysis based on the available information, highlighting what is and is not present:
Missing Information:
The document does not provide:
- A table of quantitative acceptance criteria with corresponding device performance metrics.
- Sample sizes for a dedicated test set.
- Data provenance for a test set (country of origin, retrospective/prospective).
- Number of experts or their qualifications for establishing ground truth for a test set.
- Adjudication method for a test set.
- Any information about a Multi-Reader Multi-Case (MRMC) comparative effectiveness study or human reader improvement with AI assistance.
- Details about a standalone (algorithm only) performance study against specific, pre-defined metrics.
- The sample size for a training set (as this is not an AI/ML device in the modern sense, but rather a treatment planning software).
- How the ground truth for a training set was established.
Information on "Acceptance Criteria" and "Study" (based on interpretation of "Dosimetric Validation"):
The closest the document comes to describing an "acceptance criterion" and a "study" is in sections 5.1 and 5.2 regarding Dosimetric Validation.
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Inferred) | Reported Device Performance |
|---|---|
| Brachytherapy: "Results well within expectable error margins" and "dose calculations are deemed good for clinical use" compared to: | Brachytherapy: |
| * Earlier Oncentra Brachy releases | * Comparisons made between Oncentra Brachy releases and Plato version 14.3.5. |
| * Plato version 14.3.5 | * Accuracy of applicator shielding calculations validated by comparisons with the established Plato planning system. |
| * Phantom measurements | * No specific quantitative results provided, but stated: "No changes have been made to the brachytherapy dose calculations, therefore only a comparison with the previous version has been performed to verify consistency. The results are well within expectable error margins and the dose calculations are deemed good for clinical use." |
| * Data published in the TG-43 report (AAPM Task Group No. 43 report update for brachytherapy dose calculations) | |
| External Beam: "Results well within expectable error margins" and "dose calculations are deemed good for clinical use" compared to: | External Beam: |
| * Earlier versions of Oncentra (to ensure no unexpected changes, as base algorithms have not been updated) | * Comparison of results with those from earlier versions has been made to ensure that no unexpected changes have taken place. |
| * No specific quantitative results provided, but stated: "The results are well within expectable error margins and the dose calculations are deemed good for clinical use." |
2. Sample size used for the test set and the data provenance:
- Sample Size: Not explicitly stated as a "test set" in the context of a formal study with specific case counts. The validation refers to comparisons (with other software, phantom measurements, and published data).
- Data Provenance: Not specified for a particular test set. The validation is internal ("evaluated by the physics team").
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Experts: "Evaluated by the physics team." Specific number and qualifications are not detailed beyond "physics team."
- Qualifications: Not specified.
4. Adjudication method for the test set:
- Not described.
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 MRMC comparative effectiveness study was done as this is a radiation treatment planning software, not an AI diagnostic aid for human readers. The document predates widespread discussion of "human readers improve with AI."
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- The "dosimetric validation" described for both brachytherapy and external beam calculations focuses on the algorithms' output (dose calculations). This can be considered a standalone assessment of the algorithms' accuracy in generating dose distributions. However, it's not presented as a standalone study against a defined ground truth in the way a modern AI/ML device might be evaluated for a specific diagnostic task. It's more of a verification of calculation consistency and agreement with established methods.
7. The type of ground truth used:
- Brachytherapy: Comparisons were made against:
- Results from the Plato version 14.3.5 planning system (another dose calculation algorithm).
- Phantom measurements (physical validation data).
- Data published in the TG-43 report (consensus-based physics model for brachytherapy dose calculations).
- External Beam: Comparisons were made against results from earlier versions of Oncentra (verifying consistency, implying the previous version's calculations served as a reference).
8. The sample size for the training set:
- This is not an AI/ML device in the context of machine learning training data. Therefore, a "training set" in that sense is not applicable or discussed. The software is based on physics models and algorithms, not data-driven machine learning.
9. How the ground truth for the training set was established:
- Not applicable for the reasons stated above.
In summary, the provided document offers a high-level overview of dosimetric validation performed to support the substantial equivalence claim for Oncentra 4.2. It emphasizes consistency with prior versions and established methods, rather than presenting a detailed performance study against explicit quantitative acceptance criteria for a novel AI/ML device.
§ 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.