FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The ControlRad® Select Model Z with Siemens Artis zee is indicated to provide fluoroscopic imaging of the patient during diagnostic, surgical, and interventional procedures while reducing patients' and clinicians' radiation exposure (Dose Area Product) outside of the ControlRad Select Model Z's region of interest (ROI) as compared to Artis zee non-collimated image area.1 The ControlRad Select Model Z semi-transparent filter should not be used in lieu of the Artis zee's collimators, as they block the most radiation, but can help physicians balance dose reduction with the need to visualize structures outside the ROI when it is considered clinically advantageous to do so. Clinical applications may include cardiac angiography, neuro-angiography, rotational angiography, multipurpose angiography and whole body radiographic/fluoroscopic procedures as well as procedures next to the table for i.e. patient extremities.

Additional procedures that can be performed include angiography in the operating room, image quided surgery by Xray, by image fusion, and by navigation systems. The examination table as an integrated part of the system can be used for X-ray imaging, surgery and interventions.

1Relative to open Field of View (FOV), the ControlRad Select Model Z reduces at least 85% of the Dose Area Product at 65 kVp and ROI with width and length that are smaller than 1/5 the edge size of the full FOV.

Device Description

The ControlRad® Select Model Z is a set of components mounted on the Artis zee system and cannot be used independent of the Artis zee system.

The ControlRad® Select Model Z consists of a ControlRad filter mounted onto the Siemens Medical Artis zee (K181407) C-arm. The ControlRad filter is installed semipermanently (i.e., the filter may be removed to return the C-arm to its original condition) to aid in reducing both patient and clinicians' radiation exposure while providing fluoroscopic imaging of the patient during diagnostic, surgical and interventional procedures. ControlRad® hereby submits this Traditional 510(k) to request clearance to market the ControlRad® Select Model Z with SW version V1.0.0.7.1 and a dose reduction claim: 85% of the Dose Area Product (DAP) at 65 kVp with width and length that are smaller than 1/5 the edge size of the full Field Of View (FOV).

The ControlRad Filter is an optional component installed on the Siemens Artis zee's collimator to further reduce radiation emissions. The ControlRad Select Model Z is only compatible with the 20x20 detector and associated collimator. Use of this system on other sizes is prohibited.

The additional radiation reduction provided by the ControlRad Filter will be outside the clinician-selected ROI and within the un-collimated region/image FOV.

The main components of the ControlRad® Select Model Z which are used with the Artis zee are:

ControlRad Tablet
ControlRad Filter
ControlRad Hardware
ControlRad Software and Firmware Modules
ControlRad Communication Interface

The ControlRad® Select Model Z is a system used to assist trained clinicians which is used to provide X-ray images when the clinician performs a medical procedure while reducing the patients and clinicians' radiation exposure (Dose Area Product) outside of the ControlRad® Select Model Z's reqion of interest (ROI) as compared to the Artis zee non-collimated image area. The ControlRad® Select Model Z can help physicians balance dose reduction with the need to visualize structures outside the ROI when it is considered clinically advantageous to do so.

ControlRad® Select Model Z contains a titanium filter. Two filter sizes are available with nominal thicknesses of 2.5mm or 3mm, both of which are partially transparent to X-ray radiation.

Based on the user selection of the Artis zee collimator the ControlRad Filter region of interest (ROI) the radiation will be reduced. The X-ray beam inside the ROI is not impacted by the ControlRad Filters. All radiation outside the ROI and inside the Artis zee collimated area will be filtered. This can help physicians balance dose reduction with the need to visualize structures outside the ROI when it is considered clinically advantageous to do so.

The ControlRad® Select Model Z allows physicians to select a customizable region of interest (ROI) using a ControlRad dedicated screen Tablet. The proprietary technology then adjusts semi-transparent titanium filters to deliver the designed high-quality image. This allows the Artis zee to generate an image in the selected physician's ROI while providing a lower radiation dose to the periphery. The result is a reduction in the overall radiation dose and exposure to the patient and the healthcare team while providing the physician the contextual information needed outside the ROI.

The workflow is therefore supported with lower radiation than with conventional imaging settings.

The ControlRad Filter is designed to always include the center of the FOV in the ROI. Therefore, when selecting an ROI by the user, the actual ROI might be expanded to include the center of the FOV.

The ControlRad® Select Model Z is a product that can be mounted only on the following configurations of the Artis zee: floor, ceiling, and bi-plane systems.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) summary for the ControlRad® Select Model Z:

Acceptance Criteria and Device Performance

The provided document details various performance tests conducted to demonstrate the device's safety and effectiveness. The acceptance criteria are implicit in the "Verification" or "Accuracy" statements, and the "Reported Device Performance" is the stated outcome of these tests.

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

The document does not specify exact sample sizes (e.g., number of images, patients, or test runs) for each specific test mentioned in the "Performance Data" section. It primarily relies on bench testing, system validation testing, and simulated clinical use tests.

Data Provenance: The studies appear to be non-clinical bench and system validation tests conducted by the manufacturer, rather than studies involving human subjects in a real-world clinical setting. The "clinical simulations" suggest a controlled environment. There is no information regarding the country of origin of data in terms of patient data, as no patient data appears to have been formally collected for this 510(k) submission. These are engineering and performance validation tests.

3. Number of Experts and Qualifications for Ground Truth

The document does not specify the number of experts or their qualifications used to establish ground truth for the test set.

For "Comparative image quality inside the ROI" and "Image quality evaluation via clinical simulations," it states that these validate the image quality, implying expert assessment. However, the specific number and qualifications of these "experts" (e.g., radiologists) are not provided.
The overall evaluation is based on meeting technical performance standards and demonstrating substantial equivalence to a predicate device, which primarily involves engineering validation rather than human interpretation assessment of image quality.

4. Adjudication Method for the Test Set

The document does not describe any specific adjudication method (e.g., 2+1, 3+1) for the test set. Given that the studies are primarily technical performance validations (bench and system tests, not reader studies), such adjudication methods would typically not be applicable.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not conducted. The submission focuses on demonstrating substantial equivalence through technical performance tests and comparing the device's technical specifications to a predicate device. The goal is to show the device functions as intended and safely, not to quantify improvement in human reader performance with AI assistance.

6. Standalone (Algorithm Only) Performance

The device is an add-on component to an existing fluoroscopic x-ray system (Siemens Artis zee). It is described as a "set of components mounted on the Artis zee system and cannot be used independent of the Artis zee system." Therefore, a standalone performance study in the sense of an "algorithm only without human-in-the-loop" would not be applicable as the device itself is a physical and integrated system component, not just a software algorithm for interpretation. Its function is to modify the X-ray field and subsequent image processing, which inherently involves human operation and interaction with the Artis zee system.

7. Type of Ground Truth Used

The ground truth used for these tests is primarily engineering specifications and established performance standards (e.g., 21 CFR 1020.32, IEC 60601 series).

For radiation dose metrics, the ground truth is derived from quantitative measurements verifying compliance with specified tolerances (e.g., ±35% accuracy for AK, AKR, DAP).
For image quality, the ground truth for "inside the ROI" is the image quality of the predicate Artis zee system alone. For "outside the ROI," it's about validating that the degraded image quality remains "clinically relevant" through "clinical simulations," suggesting a qualitative assessment against a clinical utility standard.
For mechanical, electrical, and software aspects, the ground truth is conformance to relevant international standards (ISO, IEC) and FDA guidance documents.

8. Sample Size for the Training Set

The document does not mention a "training set" or "training data" in the conventional sense associated with machine learning or AI models. This device primarily involves hardware modifications and associated software that controls the physical filter and image processing. It is not an AI algorithm that learns from a dataset like an image classification model.

9. How the Ground Truth for the Training Set was Established

As no training set is discussed or implied to be part of the device's development or validation process (as it's not a machine learning model), this question is not applicable.

Summary

Regulation Number and Section

§ 892.1650 Image-intensified fluoroscopic x-ray system.

(a)
Identification. An image-intensified fluoroscopic x-ray system is a device intended to visualize anatomical structures by converting a pattern of x-radiation into a visible image through electronic amplification. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II (special controls). An anthrogram tray or radiology dental tray intended for use with an image-intensified fluoroscopic x-ray system only is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9. In addition, when intended as an accessory to the device described in paragraph (a) of this section, the fluoroscopic compression device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9.

Acceptance Criteria Category	Specific Acceptance Criteria	Reported Device Performance
Radiation Dose	Dose Area Product (DAP) Reduction: Reduces at least 85% of the Dose Area Product at 65 kVp and ROI with width and length that are smaller than 1/5 the edge size of the full FOV, relative to open Field of View (FOV).	The device is able to reduce DAP as noted in the indications for use, meeting the "at least 85% reduction" criteria.
	Filter Attenuation: Shows a 44% to 98% attenuation outside the ROI.	The device demonstrated a 44% to 98% attenuation outside the ROI. (Note: The "Comparison of Technological Characteristics" table also states "Reduces radiation outside the aperture typically by 44% to 98%").
	Impact of Air Kerma: Does not significantly increase Air Kerma when compared to the Artis zee system alone.	Verified that the ControlRad components on the reference system do not significantly increase Air Kerma when compared to the Artis zee system alone.
	Air Kerma and Air Kerma Rate Accuracy (per 21 CFR 1020.32): AK and AKR of the subject system (ControlRad components integrated with the Artis zee system) is ± 35% of the Artis zee system alone.	Verified that the AK and AKR of the subject system is ± 35% of the Artis zee system alone.
	Radiation Dose Structure Report (RDSR) AK Accuracy: Cumulative AKR as referenced in the RDSR of the subject system (ControlRad components integrated with the Artis zee system) is ± 35% of the Artis zee system alone.	Verified that the cumulative AKR as referenced in the RDSR of the subject system is ± 35% of the Artis zee system alone.
	Reference Air Kerma Warning Functionality: Reference Air Kerma Warning Functionality is not impacted on the subject system when compared to the Artis zee system alone.	Verified that the Reference Air Kerma Warning Functionality is not impacted.
	Dose Area Product (DAP) Accuracy (per IEC 60601-2-43): DAP measurements of the subject system (ControlRad components integrated with the Artis zee system) is ± 35% of the Artis zee system.	Demonstrated the DAP measurements of the subject system is ± 35% of the Artis zee system. (Note: The "Comparison of Technological Characteristics" table specifies "+35%*" for DAP Accuracy, likely referring to the upper limit of the acceptable deviation).
Image Quality	Comparative Image Quality Inside the ROI: Image quality inside the ROI is at least the same quality as the image that would be gathered with the Artis zee alone.	Verified that image quality inside the ROI is at least the same quality as the image that would be gathered with the Artis zee alone.
	Comparative Image Quality Outside the ROI: Quantify the level of image quality degradation outside the region of interest as a result of the CR filter blades. (Implied acceptance is that degradation is acceptable for clinical context).	Quantified the level of image quality degradation outside the region of interest as a result of the CR filter blades.
	Image Quality Evaluation via Clinical Simulations: Validate that the image quality outside the ROI is clinically relevant following degradation due to use of the ControlRad filters.	Validated that the image quality outside the ROI is clinically relevant following degradation due to use of the ControlRad filters.
Safety & Mechanical	Leakage Radiation Evaluation: Evaluate the impact of the ControlRad components on the reference device leakage radiation measurements (Implied acceptance is no unacceptable increase).	Evaluated the impact of the ControlRad components on the reference device leakage radiation measurements.
	Stray Radiation Evaluation: Evaluate the impact of the ControlRad components on the reference device stray radiation measurements (Implied acceptance is no unacceptable increase).	Evaluated the impact of the ControlRad components on the reference device stray radiation measurements.
	Recovery Management (per IEC 60601-2-43, section 201.4.101): Verify recovery management.	Verified recovery management.
	Mechanical Impact on Filter Cover (per IEC 60601-1): Verify that impact to the filter / collimator cover does not create unacceptable risk.	Verified that impact to the filter / collimator cover does not create unacceptable risk.
	Collision Sensor Functionality Evaluation: Verify that the collision sensor functionality is maintained when using the ControlRad filter / collimator covers.	Verified that the collision sensor functionality is maintained.
	Focal spot to patient distance (per IEC 60601-2-54, section 203.9): Verify the focal spot to skin distance implementation.	Verified the focal spot to skin distance implementation.
	Tensile Strength Evaluation (per IEC 60601-1): Verify that additional mass from the Select Model Z filter does not create unacceptable risk from tensile strength.	Verified that additional mass from the Select Model Z filter does not create unacceptable risk from tensile strength.
	Filter Motion Reliability Testing: Verify the durability of the mechanical filter assembly in frame of a random motion stress test.	Verified the durability of the mechanical filter assembly.
Usability & Software	Touch-In-Glove Bench Test: Verify the sensitivity of the ControlRad tablet when using sterile radiation reducing gloves.	Verified the sensitivity of the ControlRad tablet when using sterile radiation reducing gloves.
	Wireless Devices and Cybersecurity Evaluation: Compliance with FDA Guidance documents: "Radio Frequency Wireless Technology in Medical Devices" and "Postmarket Management of Cyber Security in Medical Devices".	Evaluated compliance with wireless and cybersecurity guidance. Implemented a process of preventing unauthorized access, modifications, misuse or denial of use, or the unauthorized use of information that is stored, accessed or transferred from a medical device to an external recipient.
	Software Documentation for Moderate Level of Concern: Conformance with special controls for medical devices containing software (per FDA's Guidance Documents "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices" and "Off-The-Shelf Software Use in Medical Devices").	Documentation included; performance data demonstrates continued conformance; risk analysis completed, and risk control implemented; testing results support all software specifications met acceptance criteria; verification and validation found acceptable.
	Usability Testing (per IEC 60601-1-6): Usability related hazards are addressed in system tests according to the operator's manual and in simulated clinical use tests with customer feedback. Customer employees are adequately trained.	Usability testing showed that usability related hazards are addressed in the system test and in simulated clinical use tests with customer report and feedback form. Customer employees are adequately trained.
DAP Chamber	DAP Chamber Change Justification: Justify the change in DAP chamber from the original chamber used in the Artis zee system as a result of the addition of the CR filter.	Justified the change in DAP chamber.