Search Results

The device is a permanently installed diagnostic x-ray system for general radiographic x-ray imaging including tomography. The tomography feature is not to be used for imaging pediatric patients.

The DRX-Evolution is a diagnostic x-ray system utilizing digital radiography (DR) technology. The DRX-Evolution is designed for horizontal and upright projection exams. The system consists of a high voltage x-ray generator, overhead tube crane with x-ray tube assembly, radiographic table with detector tray, Bucky image receptor on an upright wall stand, and x-ray controls containing a power distribution unit and operator PC (user interface).

This document describes the Carestream DRX-Evolution, a diagnostic x-ray system. The modifications to the device include firmware and mechanical changes to facilitate linear tomography exams, the addition of a new generator option, and software updates such as Bone Suppression and Pneumothorax Visualization.

Here's an analysis of the acceptance criteria and the studies that prove the device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Linear Tomography Accuracy (Bench Testing):
  • Sample Size: Not explicitly stated, but it involved "a tool phantom" and "four different anthropomorphic phantoms (chest, hand, knee and pelvis)."
  • Data Provenance: Bench testing, likely conducted internally by Carestream Health, Inc. The country of origin is not specified but the company is based in Rochester, New York, USA. This is retrospective for the purpose of regulatory submission.
• Detector Image Quality (Clinical Studies - DRX 2530C and DRX-1C):
  • Sample Size: Not explicitly stated in this document ("Results of these studies demonstrated..."). These studies refer to K130464 for DRX 2530C and K120062 for DRX-1C, where detailed sample sizes would be found.
  • Data Provenance: Clinical studies, in accordance with FDA guidance. Prospective studies are typical for such evaluations. The country of origin is not specified.
• DR Long Length Imaging Software Diagnostic Capability (Clinical Study):
  • Sample Size: Not explicitly stated in this document ("Results of the DR Long Length Imaging study demonstrated..."). It refers to K130567 for more details.
  • Data Provenance: Clinical study, likely prospective. The country of origin is not specified.
• Bone Suppression Software Effectiveness (Clinical Study):
  • Sample Size: Not explicitly stated in this document ("Results of the Bone Suppression clinical study demonstrated..."). It refers to K133442 for more details.
  • Data Provenance: Clinical study, likely prospective. The country of origin is not specified.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

• Linear Tomography Diagnostic Quality (Bench Testing):

  • Number of Experts: A single expert ("evaluated by a board-certified radiologist").
  • Qualifications: "board-certified radiologist."

• For other clinical studies (Detectors, LLI, Bone Suppression): The number and qualifications of experts are not detailed in this summary; they would be in the referenced 510(k) documents (K130464, K120062, K130567, K133442).

4. Adjudication Method for the Test Set

• Linear Tomography Diagnostic Quality: No formal adjudication method is described beyond a single board-certified radiologist's evaluation.
• For other clinical studies (Detectors, LLI, Bone Suppression): Adjudication methods are not detailed in this summary; they would be in the referenced 510(k) documents.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

• No explicit MRMC comparative effectiveness study involving AI assistance for human readers is described in this document. The clinical studies mentioned for Bone Suppression and DR Long Length Imaging software compare the investigational software/images to a predicate or standard-of-care, but they don't explicitly state an "AI vs without AI assistance" MRMC study for improved human reader performance with an effect size.
  • The Bone Suppression study mentions a "companion image that, when presented to the physician along with the standard-of-care image, is rated substantially equivalent or improved." This implies a reader study where human performance (or perception of image utility) is assessed with and without the software-generated companion image, but it does not specify an MRMC design or quantify an "effect size" in terms of improved diagnostic accuracy for human readers.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

• Yes, standalone performance was evaluated for certain aspects:
  • Linear Tomography Accuracy: The initial bench testing using a "tool phantom" evaluated the mechanical accuracy of the device's linear tomography function in a standalone manner (without human diagnostic interpretation), confirming "accuracy of the tube head movement."
  • Image Quality Metrics: The statement that DRX-1C and DRX 2530C detectors "provide equal or superior image quality with respect to noise and spatial resolution at equivalent doses" suggests that standalone technical image quality metrics were assessed, and these would be algorithm-only or device-only measurements.

7. The Type of Ground Truth Used

• Linear Tomography Accuracy: The ground truth for mechanical accuracy was based on the "expected" results from a "tool phantom," implying known physical measurements or standards.
• Linear Tomography Diagnostic Quality: The ground truth was based on expert consensus/opinion by a "board-certified radiologist" regarding "acceptable diagnostic quality" of anthropomorphic phantom images.
• Detector Image Quality, DR Long Length Imaging Software, Bone Suppression Software: For these, the ground truth was generally based on comparison to a predicate device or standard-of-care, with evaluation typically performed by qualified experts (e.g., physicians for diagnostic capability). While not explicitly stated as "ground truth," the predicate/standard-of-care serves as the reference for equivalence or improvement claims.

8. The Sample Size for the Training Set

• The document does not explicitly state the sample size for any training sets.
• The software features (Bone Suppression, Pneumothorax Visualization) likely involved machine learning/AI models that would require training data. However, this information is not provided in the 510(k) summary. References to K133442 (Bone Suppression) might contain this information.

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

• The document does not provide details on how ground truth for any training sets was established. Since training set details are absent, the method for establishing their ground truth is also not mentioned.

Ask a specific question about this device

Revolution XR/d Digital Radiographic Imaging System is indicated for use in generating Tomographic images. Revolution XR/d Digital Radiographic Imaging System is not intended for mammographic applications.

The Revolution XR/d with Tomosynthesis is a radiographic x-ray system intended for examinations using digital Tomosynthesis acquisition techniques. Tomosynthesis is a hardware & software option to the Revolution XR/d Digital Radiography System. The Revolution XR/d with Tomosynthesis consists of a elevating radiographic table with integrated digital detector, a radiographic wall stand with integrated digital detector, and an x-ray tube hanger, Overhead Tube Crane or Floor Mounted Tube Stand.

The provided documents, K051967, contain limited information about the acceptance criteria and study details. Based on the "Summary of Safety and Effectiveness Prepared in accordance with 21 CFR Part 807.92(c)," here's what can be extracted:

1. Table of Acceptance Criteria and Reported Device Performance

Explanation of Implied Criteria: The document explicitly states the goal of the clinical tests was to show the tomosynthesis diagnostic capability is "as good as" the predicate device and that key features are "consistent." While no specific quantitative metrics or thresholds are provided as acceptance criteria, these qualitative statements serve as the basis for the reported performance.

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

• Sample Size: The document does not specify the sample size used for the clinical tests.
• Data Provenance: The document does not provide information on the country of origin of the data or whether the study was retrospective or prospective. It only states, "Clinical Tests: Clinical tests under the authority of an IRB..."

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

• The document does not provide any information regarding the number of experts used or their qualifications for establishing ground truth.

4. Adjudication Method for the Test Set

• The document does not mention any adjudication method used for the test set.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

• The document does not indicate that a multi-reader, multi-case (MRMC) comparative effectiveness study was performed or provide any effect size information. The clinical tests simply aimed to compare the diagnostic capability to a predicate device.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

• This device is a "Digital Radiographic Imaging System with Tomosynthesis," implying it's an imaging hardware and software system, not solely an algorithm. The clinical tests assessed its "diagnostic capability," which inherently involves the output of the system that would then be interpreted by a human. Therefore, a standalone algorithm-only performance assessment in the modern sense (e.g., AI output without human interpretation) is not explicitly described or implied in this context, given the 2005 date of the submission. The "Tomosynthesis" part is described as an "acquisition technique" and a "software option."

7. The Type of Ground Truth Used

• The document does not explicitly state the type of ground truth used (e.g., expert consensus, pathology, outcomes data). The comparison is made against the "Tomo-link linear tomographic device," suggesting the predicate device's diagnostic performance served as the benchmark.

8. The Sample Size for the Training Set

• The document refers to a "Revolution XR/d with Tomosynthesis" which is described as a "hardware & software option." Given the 2005 date, it is highly unlikely that "training set" in the context of modern machine learning was applicable in the same way. The document does not mention any training set or data used for algorithm development or machine learning.

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

• As no "training set" for an AI algorithm is mentioned or implied, there is no information on how its ground truth would have been established.

Summary of the Study:

The provided 510(k) summary for the Revolution XR/d Digital Radiographic Imaging System with Tomosynthesis (K051967) indicates that clinical tests were conducted under the authority of an Institutional Review Board (IRB). The primary finding of these tests was that the "Tomosynthesis diagnostic capability is as good as the Tomo-link linear tomographic device" (K944967), which was the predicate device. Additionally, it was concluded that the device's "key features are consistent with traditional clinical procedures."

It's important to note that this submission dates back to 2005. The level of detail regarding clinical study design, acceptance criteria, and AI-specific performance metrics has evolved significantly in regulatory submissions since then. The information provided in this document is much less detailed than what would typically be required for a modern AI/ML-enabled device submission.

Ask a specific question about this device