Search Results

uDR Aurora CX is intended to acquire X-ray images of the human body by a qualified technician, examples include acquiring two-dimensional X-ray images of the skull, spinal column, chest, abdomen, extremities, limbs and trunk. The visualization of such anatomical structures provide visual evidence to radiologists and clinicians in making diagnostic decisions. This device is not intended for mammography.

uDR Aurora CX is a model of Digital Medical X-ray Imaging System developed and manufactured by Shanghai United Imaging Healthcare Co., Ltd(UIH). It includes X-ray Generator, X-ray Imaging System. The X-ray Generator produces controlled X-rays by high-voltage generator and X-ray tube assembly, ensuring stable energy output for human body penetration. The X-ray Imaging System converts X-ray photons into electrical signals by detectors, and generates DICOM-standard images by workstation to reflecting density variations of human body.

This document describes the acceptance criteria and study details for two features of the uDR Aurora CX device: uVision and uAid.

1. Acceptance Criteria and Reported Device Performance

2. Sample Size and Data Provenance for the Test Set

3. Number and Qualifications of Experts for Ground Truth (Test Set)

4. Adjudication Method (Test Set)

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

• uVision: No MRMC comparative effectiveness study was done to compare human readers with and without AI assistance. The study evaluates the AI's direct assistance in positioning, measured by compliance with clinical criteria, rather than comparing diagnostic performance of human readers.
• uAid: No MRMC comparative effectiveness study was done. The study focuses on the standalone performance of the algorithm in identifying image quality issues, not on how it impacts human reader diagnostic accuracy or efficiency.

6. Standalone Performance (Algorithm Only)

• uVision: Yes, a standalone performance study was done. The "95% compliance" rate refers to the algorithm's direct ability to set system position and FOV that meet clinical technician criteria without a human actively adjusting or guiding the initial AI-generated settings during the compliance evaluation. Technicians could manually adjust those settings if needed.
• uAid: Yes, a standalone performance study was done. The algorithm processes images and outputs a quality classification (Green, Yellow, Red) and identifies specific issues (foreign object, incomplete lung fields, etc.). Its sensitivity and specificity metrics are standalone performance indicators.

7. Type of Ground Truth Used

• uVision: Expert Consensus/Clinical Criteria: The ground truth for uVision's performance (i.e., whether the automatically set position/FOV was "compliant") was established by "clinical experts" based on "clinical technicians' criteria" for proper positioning and shooting requirements.
• uAid: Expert Consensus/Manual Labeling: The ground truth for uAid's evaluation (e.g., presence of foreign objects, complete lung fields, open scapula, centerline deviation) was established through a "classification" process, implying manual labeling or consensus by experts after data collection and cleaning. The document mentions "negative" and "positive" data distributions for each criterion.

8. Sample Size for the Training Set

• uVision: Not explicitly stated in the provided text. The testing data was confirmed to be "collected independently from the training dataset, with separated subjects and during different time periods."
• uAid: Not explicitly stated in the provided text. The document mentions "The data collection started in October 2017, with a wide range of data sources" for training, but does not provide specific numbers for the training set size.

9. How Ground Truth for Training Set was Established

• uVision: Not explicitly stated for the training set. It can be inferred that a similar process to the test set, involving expert review against clinical criteria, would have been used.
• uAid: Not explicitly stated for the training set. Given that the data was collected from "different cooperative hospitals," "multiple cleaning and sorting" was performed, and the study was "approved by the institutional review board," it is highly likely that the ground truth for the training set involved manual labeling by clinical experts/radiologists, followed by a review process (potentially consensus-based or single-expert) to establish the labels for image characteristics and quality.

Ask a specific question about this device

The intended use of the device YSIO X.pree is to visualize anatomical structures of human beings by converting an X-ray pattern into a visible image.

The device is a digital X-ray system to generate X-ray images from the whole body including the skull, chest, abdomen, and extremities. The acquired images support medical professionals to make diagnostic and/or therapeutic decisions.

YSIO X.pree is not for mammography examinations.

The YSIO X.pree is a radiography X-ray system. It is designed as a modular system with components such as a ceiling suspension with an X-ray tube, Bucky wall stand, Bucky table, X-ray generator, portable wireless, and fixed integrated detectors that may be combined into different configurations to meet specific customer needs.

The following modifications have been made to the cleared predicate device:

• -New Camera Model in Collimator
• -New Auto Collimation Function: Auto Long-Leg/Full-Spine
• -Two new wireless detectors

The provided text is a 510(k) summary for the YSIO X.pree X-ray system. It describes the device, its intended use, and comparisons to predicate and reference devices. However, it does not contain the detailed clinical study information typically required to directly answer all aspects of your request regarding acceptance criteria and performance metrics for an AI/CADe device.

Specifically, the document mentions:

• A "Customer Use Test (CUT)" was performed at the "Universitätsklinikum Augsburg, Germany," focusing on "System function and performance-related clinical workflow, Image quality, Ease of use, Overall performance and stability."
• "The results of the clinical test stated that the intended use of the system was met, and the clinical need covered."
• "All images acquired with the new detectors were sufficiently acceptable for radiographic usage."

This summary indicates that new features, particularly the "Auto Collimation Function: Auto Long-Leg/Full-Spine" which is AI-based (taken from the MULTIX Impact algorithm, K213700), underwent testing. However, the FDA 510(k) summary does not include the specific acceptance criteria with reported performance against those criteria, nor detailed information about the study design (sample size, ground truth establishment, expert qualifications, etc.) for the AI-based auto collimation feature. The "Customer Use Test" appears to be a general usability and performance test for the overall system and new detectors, rather than a rigorous performance study for an AI algorithm with specific quantitative metrics.

Therefore, I cannot fully complete the table and answer all questions with the provided text. I can only extract what is present.

Here's a breakdown of what can be extracted and what cannot:

1. Table of Acceptance Criteria and Reported Device Performance:

The document does not provide a table of explicit acceptance criteria for the AI-based auto collimation function with corresponding quantitative performance metrics (e.g., accuracy, precision for delimiting regions of interest). It only states that the overall system and new detectors' images were "sufficiently acceptable for radiographic usage" and that the "intended use of the system was met, and the clinical need covered."

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

• Test set sample size for AI-based auto collimation: Not specified in the provided text.
• Data Provenance: The Customer Use Test (CUT) was performed at "Universitätsklinikum Augsburg, Germany." This suggests prospective data collection in a clinical setting in Germany for the general system and new detectors. It is not explicitly stated if the AI-based auto collimation performance was evaluated on this specific dataset, or if a separate dataset (and its provenance) was used for validating the AI. Given the AI algorithm was "taken over" from the MULTIX Impact (K213700) and that previous 510(k) for that device might contain more details.

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

• Not specified for any specific ground truth establishment (especially for the AI-based auto collimation). The "Customer Use Test" involved clinical evaluation, implying healthcare professionals (presumably radiologists or radiographers) were involved, but their number and specific qualifications for establishing ground truth for AI performance are not detailed.

4. Adjudication method (e.g., 2+1, 3+1, none) 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:

• No MRMC comparative effectiveness study is described for the AI-based auto collimation. The document focuses on device safety and substantial equivalence to a predicate, not enhancement of human reader performance.

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

• Not explicitly detailed. The AI auto collimation feature is integrated into the workflow, implying it assists, but a standalone technical performance study for the AI component itself is not described with quantitative results. The statement that the "Multix Impact algorithm has been taken over" suggests that its performance characteristics might have been established during the clearance of the MULTIX Impact (K213700), but those details are not in this document.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

• Not specified for the AI-based auto collimation. For the general system usability and image quality, the "Customer Use Test" implies a clinical assessment, likely representing expert (clinician) judgment.

8. The sample size for the training set:

• Not specified. The document states that the AI algorithm was "taken over" from the MULTIX Impact. This implies the training was done previously for the MULTIX Impact, but the size of that training set is not provided here.

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

• Not specified, for the same reasons as in point 8.

Ask a specific question about this device

MULTIX Impact E is a radiographic system used in hospitals, clinics, and medical practices. MULTIX Impact E enables radiographic exposures of the whole body including: skull, chest, abdomen, and extremities and may be used on pediatric, adult and obese patients. Exposures may be taken with the patient sitting, standing, or in the prone position. MULTIX Impact E uses digital detectors for generating diagnostic images by converting X- rays into image signals. MULTIX Impact E is also designed to be used with conventional film/screen or Computed Radiography (CR) cassettes. MULTIX Impact E is not intended for mammography.

The MULTIX Impact E Radiography X-ray system is a modular system of X-ray components (floor-mounted x-ray tube, bucky wall stand, bucky table, x-ray generator, portable wireless detector) based on the predicate device, the MULTIX Impact E (VB10) (K220919). The following modifications have been made to the predicate device: 1) A new elevating table (option) 2) Upgraded software version to VB20 to support hardware modifications. The modified system will be branded as the MULTIX Impact E.

Please note, this document pertains to the MULTIX Impact E, which is a stationary X-ray system, and not an AI-powered diagnostic device. The submission focuses on demonstrating substantial equivalence to a predicate device, highlighting hardware modifications and software upgrades to support these changes. Therefore, the questions related to AI device performance, such as MRMC studies, ground truth establishment for AI, and training/test set sample sizes, are not applicable to this specific submission.

The acceptance criteria and performance are based on general safety and effectiveness of X-ray systems, primarily through compliance with recognized standards and non-clinical testing.

Here's a breakdown based on the provided document, addressing the applicable points:

Acceptance Criteria and Device Performance for MULTIX Impact E (X-Ray System)

1. A table of acceptance criteria and the reported device performance

The document does not present explicit "acceptance criteria" in a quantitative performance table for a diagnostic outcome (like sensitivity/specificity for a disease). Instead, the acceptance is based on demonstrating substantial equivalence to a predicate device through meeting regulatory standards, functional performance of components, and safety considerations. The performance is reported in terms of comparison to the predicate device for various attributes.

Here's an inferred "acceptance criteria" based on the document's structure, which is "Substantially Equivalent to Predicate Device (MULTIX Impact E VB10, K220919) and Reference Device (MULTIX Impact VA21, K213700) with no new safety risks":

2. Sample sized used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

This document describes a premarket notification (510(k)) for a conventional X-ray system, not an AI/ML-driven device. The "test set" here refers to non-clinical verification and validation testing of the device hardware and software against engineering specifications and regulatory standards.

• Sample Size: Not applicable in the context of patient imaging data for an AI algorithm. The testing involves system-level and component-level verification, functional tests, and safety tests performed on representative units of the device. The document does not specify a "sample size" of devices tested, but rather indicates that such testing was "successfully completed."
• Data Provenance: Not applicable in the context of patient data. The "data" comes from engineering tests and measurements on the device itself, not from clinical images.

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)

Not applicable. Ground truth as typically defined for medical image analysis (e.g., diagnosis of disease) is not established for this type of device submission. The "ground truth" for the performance of an X-ray system revolves around its physical characteristics, image quality parameters, and safety compliance, which are measured and evaluated by engineers and quality assurance personnel against established technical specifications and regulatory standards.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Not applicable. Adjudication methods are relevant for clinical studies where expert consensus is needed to establish ground truth for image interpretation. This submission is based on engineering and manufacturing verification and validation, not clinical image interpretation.

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. This is not an AI-powered device, and no MRMC study was performed or required for this type of submission.

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

Not applicable. This is not an AI algorithm.

7. The type of ground truth used (expert concensus, pathology, outcomes data, etc)

The "ground truth" for this device's performance is established by engineering specifications, compliance with recognized industry standards (e.g., IEC, ISO, NEMA for X-ray systems), and adherence to manufacturing quality controls. This is demonstrated through physical measurements, electrical tests, safety circuit validation, software functionality tests, and image quality measurements, rather than clinical outcomes or diagnostic interpretations.

8. The sample size for the training set

Not applicable. This document does not describe an AI/ML device with a training set.

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

Not applicable. This document does not describe an AI/ML device with a training set.

Ask a specific question about this device

MOBILETT Impact is a mobile device intended to visualize anatomical structures of human beings by converting an Xray pattern into a visible image. MOBILETT Impact is not intended for mammography examinations.

MOBILETT Impact is a complete X-ray imaging system on wheels. It contains a single tank high voltage generator with an X-ray tube and collimator attached to the end of a telescopic support arm connected to a swiveling column. The system is a manually driven system, with no motor support for movement. The system includes a digital image acquisition system with an image display and graphical user interface. The digital detector, Max wi-D can be stored in the built-in docking station in the system. In addition, the system can be used with detectors Max mini and Core-L. All three detectors are equipped with rechargeable batteries, which can be charged by external battery chargers. The system can perform X-Ray when it is connected to mains. Exposure can be released by a hand switch or remote control. The included system batteries only power the imaging system if it is not connected to the mains. Besides the detectors and image system, the hardware of the same as for the predicate device SEDECAL SM-V, which was cleared on 11/21/2022 with K222951.

The provided text describes the Siemens MOBILETT Impact, a mobile X-ray system. However, it does not detail acceptance criteria or a study proving the device meets specific performance metrics for image quality or diagnostic accuracy in the way an AI/CADe device would.

Instead, the document focuses on demonstrating substantial equivalence to a predicate device (SEDECAL SM-V, K222951) and a reference device (MULTIX Impact, K213700) based on regulatory compliance, technological characteristics, and safety.

The summary highlights:

• Regulatory Compliance: Adherence to various IEC, ANSI AAMI, ISO, NEMA, and FDA standards for medical electrical equipment, radiation protection, usability, risk management, software life cycle, and digital imaging.
• Technological Equivalence: Comparison of features like regulation description, product code, indications for use (similar), high voltage generator, X-ray tube, tube voltage/current, collimator, touch screen control, movement (non-motorized), US Performance Standard, and power source (all same as predicate or similar).
• Safety and Effectiveness Concerns: General statements about IFU, safety features (visual/audible warnings), error monitoring, and adherence to recognized industry practices to minimize electrical, mechanical, and radiation hazards.

Regarding the specific information requested in your prompt for an AI/CADe device:

1. Table of acceptance criteria and reported device performance: This information is not provided in the document. The document confirms regulatory compliance with various standards, but it doesn't state specific device performance metrics like sensitivity, specificity, or AUC for diagnostic tasks, as would be expected for an AI device.

2. Sample size used for the test set and data provenance: This information is not provided. A "Customer Use Test (CUT)" was performed, but no details on sample size, data type, or provenance are given.

3. Number of experts used to establish the ground truth for the test set and qualifications: This information is not provided. The CUT involved "gathering feedback on the device's usability in the clinical environment," but it doesn't describe a ground truth establishment process by experts for a diagnostic task.

4. Adjudication method for the test set: This information is not provided.

5. Multi-reader multi-case (MRMC) comparative effectiveness study: This information is not provided. The device is a mobile X-ray system, not an AI/CADe system designed to improve human reader performance.

6. Standalone performance (algorithm only without human-in-the-loop performance): This information is not provided. The device is a hardware system for X-ray imaging, not a standalone algorithm.

7. Type of ground truth used: For the "Customer Use Test (CUT)," the "ground truth" implicitly involved qualitative feedback on "system function and performance-related clinical workflow, image quality, ease of use, and overall performance and stability." This is not a quantitative ground truth for a diagnostic task like pathology or outcomes data.

8. Sample size for the training set: This information is not provided. There is no mention of a training set as this is a hardware device.

9. How the ground truth for the training set was established: This information is not provided.

In summary, the provided FDA 510(k) clearance letter and its associated summary are for a conventional mobile X-ray imaging system, not an AI/CADe product. Therefore, the detailed performance metrics and study design elements typically associated with AI/CADe device validation (e.g., sensitivity, specificity, reader studies, ground truth establishment by experts) are not present in this document. The "study" mentioned is a "Customer Use Test (CUT)" focused on usability and overall system performance in a clinical workflow, rather than a diagnostic performance study.

Ask a specific question about this device