(203 days)
The Sentinel system is intended for use in radiation therapy clinics to accurately position patients in a reproducible way, prior to treatment and to monitor the patient continuously during treatment. The system provides information about a patient's position and the adjustments required in order to position the patient as close as possible to a reference setup. During monitoring, the system reports deviations in the patient's position during treatment.
The Sentinel system is a new version of the Positioner system (previously registered with 510(k) Number K063839), updated with an application module for patient monitoring, cMotion (the design control documents are included in Appendices A and B).
Sentinel is an advanced system for surface contour localization and patient monitoring during the radiotherapy treatment process. The Sentinel platform is based on advanced laser technology with multipurpose software modules covering different tasks in the treatment procedure. The c4D multi-application software supports all modes of operation in one integrated package. The software is user friendly and requires a minimum of user interaction in the daily clinical workflow, while providing the advanced user with sophisticated data management, analysis and reporting functionalities. The software is designed to integrate with existing systems at the clinic, such as CT, linacs and R&V systems, and with motorized couch tops.
Sentinel includes two application modules, cPosition for fast and accuracy patient positioning and cMotion for motion detection during the treatment delivery procedure.
The Sentinel hardware consists of a single scanner unit containing the laser and camera, mounted in the ceiling in front of the gantry. The scanner is connected to the PC running the c4D software.
During patient surface acquisition, a laser line is swept along the patient while the camera records a number of images. From the data acquired, a complete 3D surface of the patient can be reconstructed using laser line triangulation. For patient positioning, the acquired surface is captured in a few seconds and can contain several hundred contours. For motion detection the number of contours is typically lowered so that the desired frame rate is achieved.
cPosition
Once the treatment planning has been performed, the resulting plan can be transferred to the Sentinel system through import from the industry-standard DICOM format, creating the reference data necessary for patient positioning. Reference data can also be by created using the Sentinel laser scanner. In the treatment room, synchronization with the LINAC or R & V (Record and Verify) system ensures that the correct reference data is called up automatically when the patient is selected for treatment, and also eliminates the need for any manual selection of the patient in the Sentinel system.
By advanced surface registration algorithms the actual patient position is compared to the predefined reference, suggesting within seconds a correction in six degrees of freedom of the patient's position. With interface to major accelerator vendors the suggested patient position is transferred to the respective couch control system and fast and accurate alignment is achieved.
cMotion
cMotion monitors the movement of the patient during treatment delivery and automatically warns if the patient moves outside the allowed tolerances.
The provided text describes the C-RAD Sentinel system, a radiotherapy positioning system. However, it does not contain the detailed information necessary to fully answer all aspects of your request regarding acceptance criteria and the study that proves the device meets them. This document is a 510(k) summary, which focuses on demonstrating substantial equivalence to a predicate device rather than providing a comprehensive study report with specific performance metrics and methodologies.
Here's a breakdown of what can be extracted and what is missing:
1. A table of acceptance criteria and the reported device performance:
The document broadly states the intended use, which implies certain performance goals, but it does not specify quantitative acceptance criteria or detailed reported performance figures in a table format.
| Acceptance Criteria (Implied) | Reported Device Performance (Implied/Stated) |
|---|---|
| Accurate patient positioning | Achieves "fast and accurate alignment" |
| Reproducible patient setup | Positions patients "in a reproducible way" |
| Continuous patient monitoring | Monitors patient movement during treatment, warns if movement is outside allowed tolerances |
| Integration with clinic systems | Designed to integrate with CT, linacs, R&V systems, and motorized couch tops |
| Fast surface acquisition | "Acquired surface is captured in a few seconds" |
| Suggests corrections in 6 DoF | "Suggesting within seconds a correction in six degrees of freedom" |
What's missing for point 1: Specific numerical values for accuracy (e.g., within X mm or degrees), reproducibility (e.g., standard deviation of Y mm), latency of motion detection, or specific thresholds for "allowed tolerances."
2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective):
This information is not available in the provided 510(k) summary. The document mentions "performance data has been submitted," but does not detail the nature of this data, its sample size, or its provenance.
3. 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):
This information is not available in the provided 510(k) summary. Given the device's function (positioning and motion detection), the "ground truth" would likely involve physical measurements or phantoms, rather than expert interpretation of images.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
This information is not available in the provided 510(k) summary.
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:
This is not applicable to the Sentinel device as described. The Sentinel system is a radiotherapy positioning system that uses laser technology for surface contour localization and patient monitoring. It is not an AI-assisted diagnostic imaging device that human "readers" would interpret or benefit from in the way an MRMC study evaluates. The system provides objective positional data and alerts, rather than complex interpretations requiring human "reading."
6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
The device inherently operates with "human-in-the-loop" for the positioning and treatment process, but its core function of detecting patient position and movement is an algorithm-only task. The system "suggests within seconds a correction" and "automatically warns if the patient moves outside the allowed tolerances." This implies a standalone capability for sensing and alerting based on its algorithms. However, the study specific to this standalone performance is not described with details of methodology or metrics.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
The document does not explicitly state the type of ground truth used for performance testing. Given the nature of the device (positioning and motion detection), the ground truth would likely be established through:
- Physical measurements: Using calibrated phantom devices, external measurement systems (e.g., optical tracking systems, micrometers, or other highly accurate instruments) to verify the accuracy of the Sentinel's reported positions and movements.
- Known displacements: Inducing precise, known movements in phantoms or test setups and verifying if the Sentinel system accurately detects and quantifies these movements.
8. The sample size for the training set:
This information is not available in the provided 510(k) summary. The Sentinel system uses "advanced surface registration algorithms." While such algorithms often involve training data (e.g., to learn surface features or model variations), the document does not specify any training set size. It's possible that the algorithms are based on fundamental physics and geometry rather than large-scale machine learning, in which case a "training set" might not be applicable in the same way as for AI image analysis.
9. How the ground truth for the training set was established:
This information is not available in the provided 510(k) summary for the reasons mentioned in point 8.
In summary: The provided 510(k) document serves its purpose of demonstrating substantial equivalence by outlining the device's function, intended use, and stating that performance data was submitted. However, it specifically lacks the detailed study methodology, quantitative acceptance criteria, specific performance metrics, sample sizes, and ground truth establishment methods that would be present in a comprehensive study report or a more detailed technical file.
§ 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.