(232 days)
TMS is used to plan radiation treatments with linear accelerators and other similar teletherapy devices with x-ray energies from 4 to 50MV, as well as Cobalt-60, and electron energies from 4 to 50 MeV. TMS will plan the 3D radiotherapy treatment approaches of combined modality plans, asymmetric and noncoplanar fields; total body irradiation (TBI); multi-leaf collimators (MLC); motorized and dynamic wedges; customized blocking; compensating filters (CF); and bolus,
The Siemens Medical Systems, Oncology Care Systems Group TMS is a 3D Radiotherapy Treatment Planning (RTP) system for radiation dose planning of patients undergoing external beam treatment in the Oncology clinic. TMS is a 3-D system, using modern algorithms for dose calculations. A convolution/superposition pencil beam algorithm is used for photons and a generalized Gaussian pencil beam model is used for electrons. The system software is designed to lead the user through a logical flow planning process.
The provided document (K953391) is a 510(k) summary for a Treatment Planning System (TMS) and primarily focuses on demonstrating substantial equivalence to a predicate device. It does not contain the detailed information required to fill out all aspects of your request regarding acceptance criteria and a specific study proving the device meets those criteria for AI/performance evaluation in the modern sense.
This document predates widespread AI development and the common practice of defining detailed performance metrics like sensitivity, specificity, F1-score, or AUC for diagnostic/predictive AI devices. Instead, it focuses on safety, effectiveness, and equivalence in the context of a treatment planning software.
Here's an attempt to extract what information is present, and explicitly state where information is not available in the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Derived from "Performance tests were conducted and the results indicated that the system consistently performed within the design parameters and equivalently to the predicate device.") | Reported Device Performance |
|---|---|
| Consistent performance within design parameters (e.g., accurate dose calculations, logical workflow, correct beam and patient data verification, correct export capabilities) | System consistently performed within design parameters. |
| Performance equivalent to the predicate device (SCANDIPLAN 3-Dimensional Radiation Therapy Treatment Planning System K914926) | System performed equivalently to the predicate device. |
| Compliance with IEC 601-1 (Medical electrical equipment - General requirements for safety) | System complies with IEC 601-1. |
| Compliance with IEC 601-1.1 (Safety requirements for medical electrical systems) | System complies with IEC 601-1.1. |
| Compliance with IEC 878 (Graphical symbols for electrical equipment in medical practice) | System complies with IEC 878. |
| Compliance with FDA, CDRH, ODE, August 29, 1991, Reviewer Guidance for Computer Controlled Medical Devices Undergoing 510(k) Review | System complies with the FDA guidance document. |
2. Sample size used for the test set and the data provenance:
- Not available. The document states "Performance tests were conducted" but does not specify the sample size of cases, patients, or scenarios used in these tests.
- Data provenance: Not explicitly stated. It's highly probable that the testing would have involved internal company data or standard test cases, but no country of origin or retrospective/prospective nature is mentioned.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not available. The concept of "ground truth" established by external experts for a performance study (as understood in AI/diagnostic device contexts) is not detailed. The performance tests would have likely involved comparing the TMS's output (e.g., dose distributions) against known physics principles, simulated scenarios, or outputs from the predicate device, possibly validated by internal experts or engineers.
4. Adjudication method for the test set:
- Not applicable / Not available. Adjudication methods like "2+1" or "3+1" are typically used for establishing ground truth in diagnostic studies where there's subjectivity. For a treatment planning system, performance is likely evaluated against objective criteria (e.g., adherence to physical laws, calculation accuracy, UI functionality). No such adjudication method is 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:
- Not applicable / Not available. This document describes a traditional treatment planning system, not an AI-assisted diagnostic or interpretative device. Therefore, an MRMC study comparing human readers with and without AI assistance was not performed and is not mentioned.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, implicitly. The "Performance tests were conducted and the results indicated that the system consistently performed within the design parameters" implies evaluating the software's functionality, calculations, and output on its own (standalone). However, the specific metrics and methodology of this "standalone" evaluation are not detailed beyond a general statement of compliance and equivalence.
7. The type of ground truth used:
- Implicitly: Objective engineering standards, physical models, and predicate device outputs. For a treatment planning system, "ground truth" for calculations would be derived from known physics principles, validated algorithms, and comparison with a well-established predicate device. For user interface and workflow, it would be against design specifications and usability goals. Pathology or outcomes data are not relevant for validating the dose calculation and planning capabilities of this type of system.
8. The sample size for the training set:
- Not applicable / Not available. This device is a rule-based or algorithm-based treatment planning system, not a machine learning or AI system that requires a "training set" in the modern sense. Its algorithms are developed based on physics and mathematical models, not trained on a dataset.
9. How the ground truth for the training set was established:
- Not applicable / Not available. As explained above, there is no "training set" for this type of system. The ground truth for its underlying physics models and algorithms would be established through scientific research, established principles of radiation physics, and engineering validation.
In summary, this 510(k) document is from an era and for a type of device where the evaluation criteria and methodologies were very different from what would be expected for a modern AI-powered medical device. The focus is on demonstrating safety, effectiveness, and substantial equivalence to a predicate device through adherence to standards and general performance tests, rather than detailed statistical performance metrics against a rigorously established ground truth dataset using AI-specific methodologies.
§ 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.