Search Results

The RT Elements are applications for radiation treatment planning for use in stereotactic, conformal, computer planned, linac based radiation treatment of cranial, head and neck and extracranial lesions.
The Multiple Brain Mets SRS application as one RT Element provides optimized planning and display for cranial multimetastases radiation treatment planning.
The Cranial SRS application as one RT Element provides optimized planning and display for cranial radiation treatment planning.
The Spine SRS application as one RT Element provides optimized planning and display for single spine metastases.
The Dose Review application as one RT Element contains features for review of isodose lines, review of DVHs, dose comparison and dose summation.
RT QA is an accessory to the RT Elements and contains features for patient specific quality assurance. Use RT QA to recalculate patient treatment plans on a phantom to verify that the patient treatment plan fulfills the planning requirements.
The Adaptive Hybrid Surgery Analysis application as one RT Element simulates an automated template-based radiation treatment plan. The simulated plan is intended for treatment evaluation for example in tumor board meetings or operating rooms.

The RT Elements are applications for radiation treatment planning for use in stereotactic, conformal, computer planned, Linac based radiation treatment of cranial, head and neck, and extracranial lesions. There are several applications defined as an RT Element.
The RT Elements are released as a system. Each RT Element is released as a separate subsystem including risk analysis, verification and usability as well as design input an review activities. The system can be seen as container and documents compatibility between the elements. In addition, validation activities of the RT Elements are documented in the system as well as service and user documents.

The provided text does not contain specific acceptance criteria, study data, sample sizes, or details about expert consensus, MRMC studies, or standalone algorithm performance for the RT Elements device. The document primarily describes the device's indications for use, regulatory classification, and general information about its underlying dose calculation algorithms and the reason for the special 510(k) submission (a new optimization algorithm in Multiple Brain Mets SRS 2.0).

Here's what can be extracted and what information is missing based on your request:

1. Table of Acceptance Criteria and Reported Device Performance:

• Acceptance Criteria: Not explicitly stated in terms of quantitative metrics for performance (e.g., accuracy, sensitivity, specificity). The document mentions that the accuracy of both the pencil beam and Monte Carlo algorithms is tested according to IAEA-TECDOC-1540 to be "better than 3%." This is the closest to an acceptance criterion in terms of numerical performance.
• Reported Device Performance: The document states that "optimization results of the new algorithm is equivalent to the algorithm used in the predicate device." It also mentions that "All tests reports were rated as successful according to the acceptance criteria" during verification and validation. However, no specific performance metrics like those typically found in clinical studies (e.g., AUC, sensitivity, specificity, or specific error rates) are reported for the RT Elements as a whole or for its individual applications.

2. Sample size used for the test set and the data provenance:

• Test Set Sample Size: Not specified.
• Data Provenance: Not specified.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• Number of Experts: Not specified. The document mentions that "clinical experts evaluated the clinical suitability of radiation therapy planning using the RT Elements workflows," but doesn't quantify how many or their specific qualifications for establishing ground truth on a test set.
• Qualifications of Experts: It generally states "medical professionals who perform radiation treatment planning (medical physicists, radiation oncologists, dosimetrists, physicians, etc.)" were typical users and involved in validation.

4. Adjudication method for the test set:

• Not specified.

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 specified. The document focuses on the capabilities of the RT Elements applications for treatment planning, not on a direct comparison of human readers with and without AI assistance in a diagnostic or interpretive context.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• The document implies that the algorithms for dose calculation (pencil beam and Monte Carlo) and the new optimization algorithm in Multiple Brain Mets SRS 2.0 were tested for their inherent accuracy and equivalence to the predicate device. However, it doesn't refer to a standalone clinical performance study in the way it might be for a diagnostic AI. The "optimization results...is equivalent" suggests a standalone comparison of the new algorithm's output to the previous one.

7. The type of ground truth used:

• For the dose calculation algorithms, the ground truth appears to be based on established physical models and experimental verification detailed in scientific publications (Mohan et al., Kawrakow et al., Fippel et al.) and confirmed against IAEA-TECDOC-1540 (which outlines criteria for verifying dose calculation algorithms).
• For the new optimization algorithm, the ground truth for establishing "equivalence to the predicate device" would have been the performance/output of the predicate device's algorithm.
• For the validation by clinical experts, the "clinical suitability" likely involved expert consensus on the treatment plans generated, but the specific type of "ground truth" (e.g., patient outcomes, pathology confirmation) is not detailed.

8. The sample size for the training set:

• Not specified. The document refers to the dose calculation algorithms being based on published research and the new optimization algorithm being "rewritten," implying model training or development, but no details on training set size.

9. How the ground truth for the training set was established:

• Not specified. Given the nature of radiation treatment planning software, "training" in the traditional machine learning sense might not apply to all components. The dose calculation algorithms are based on physics models, not trained on labeled datasets in the same way an image classification AI would be. For the "optimization algorithm," the ground truth for its development would be the desired optimal treatment plan parameters or outcomes, likely derived from clinical best practices and physics principles, but specific details are not provided.

Ask a specific question about this device