(41 days)
The device is designed to perform radiographic x-ray examinations on pediatric and adult patients treatment areas.
Carestream Health, Inc. is submitting this Traditional 510(k) premarket notification for a modification to the cleared DRX-Revolution Nano Mobile X-ray System (K170755). The product will continue to be marketed as the DRX-Revolution Nano Mobile X-ray System. Consistent with the original system, the modified DRX-Revolution Nano Mobile X-ray System (also referred to throughout this document as the Nano system) is a diagnostic mobile x-ray system utilizing digital radiography (DR) technology. The system consists of a self contained x-ray generator, image receptor(s), imaging display and software for acquiring medical diagnostic images outside of a standard stationary x-ray room. The modified Nano system can also be used to expose Computed Radiography (CR) storage phosphor or film cassettes. Modifications have been made to the design of the system's x-ray tube and its operational characteristics for the purpose of improving the x-ray tube reliability and life span. This required modifications to the embedded software to accommodate these tube design changes.
This document describes the validation study for the DRX-Revolution Nano Mobile X-ray System, which involved modifications to the x-ray tube and software. The study's primary goal was to demonstrate that the modified device performs as well as or better than its predicate device with respect to diagnostic image quality.
1. Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Mean RadLex rating of images to be greater than 3.5 (the mid-value of the 'Diagnostic (3)' quality rating scale), indicating Diagnostic (3) quality or better. | The mean RadLex rating for the 60 responses was 3.6. |
| No significant difference in ratings across readers. | The ANOVA yielded a p-value > 0.05, failing to reject the null hypothesis "All means are equal" for the RadLex rating means of each Reader. This indicates no significant difference in ratings across Readers. |
| Diagnostic capability of the modified system is statistically equivalent to or better than that of the predicate device. | "Results of the Phantom Reader Study indicated that the diagnostic capability of the modified DRX-Revolution Nano Mobile X-ray System is statistically equivalent to or better than that of the predicate device." (It is explicitly stated that 100% of the image ratings were Diagnostic (3) quality, and approximately 78% of ratings fell between Diagnostic and Exemplary. The one-sample t-test showed statistical significance at the 95% level, implying quality ratings equal to or better than a Diagnostic (3) quality rating). |
| Non-clinical test results demonstrate that the device conforms to its specifications, meeting predefined acceptance criteria for workflow, performance, function, shipping, verification, validation, and reliability (including software and hardware). | "Non-clinical test results have demonstrated that the device conforms to its specifications. Predefined acceptance criteria were met and demonstrated that the device is as safe, as effective, and performs as well as or better than the predicate device." |
2. Sample size used for the test set and the data provenance
- Sample size for the test set: 60 RadLex ratings (from 3 radiologists, each rating a set of images, totaling 60 individual responses). The number of unique images or cases is not explicitly stated, but it can be inferred that there were at least 20 images/cases if each radiologist rated all the same images.
- Data provenance: Prospective, synthetic data using anthropomorphic phantoms. The phantoms included commercial products representing various human anatomical sections (adult chest "Lungman," adult pelvis, upper and lower limbs of a 5-year-old child, and a Gammex 610 neonatal chest phantom). Additional PMMA plates were used to simulate larger adult patients. The country of origin of the data is not specified beyond the manufacturers of the phantoms (Kyoto Kagaku for Lungman, CIRS Inc for child phantoms, Supertech for neonatal phantom).
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Number of experts: Three (3) board-certified radiologists.
- Qualifications of experts: "Three (3) board certified radiologists with extensive reading experience." Specific years of experience are not provided.
4. Adjudication method for the test set
- The radiologists provided independent ratings (single stimulus format) on a work list of images.
- There is no explicit mention of an adjudication method for disagreements between readers; rather, the statistical analysis (ANOVA) was used to determine if there was a significant difference between reader ratings. The finding was that "There is no significant difference in ratings across Readers," implying sufficient agreement for the purpose of the study.
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, a multi-reader, multi-case (MRMC) comparative effectiveness study comparing human readers with AI assistance versus without AI assistance was explicitly done. This study focused on the image quality of the device itself using human readers as evaluators, not on the impact of AI assistance on human reader performance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- No, an algorithm-only standalone performance study was not done. The study evaluated the diagnostic imaging performance of the system by using human radiologists to rate the images. The device itself is an X-ray system, not an AI algorithm for image interpretation.
7. The type of ground truth used
- The ground truth in this study was expert consensus rating by multiple radiologists, rather than an objective "true" diagnosis or pathology. The images were generated from anthropomorphic phantoms (synthetic data), so there was no "patient outcome" or "pathology" data to serve as ground truth in a clinical sense. The "ground truth" for the acceptance criteria was the radiologists' ratings on the RadLex scale.
8. The sample size for the training set
- Not applicable / not provided in the input. This study describes the validation of an X-ray system after modifications to its hardware and embedded software. It is not an AI algorithm study that typically involves a training set and a separate test set. The "training" for such a system would involve engineering and development, not data-driven machine learning as understood in AI contexts.
9. How the ground truth for the training set was established
- Not applicable / not provided in the input. As noted above, this isn't an AI study with labeled training data.
§ 892.1720 Mobile x-ray system.
(a)
Identification. A mobile x-ray system is a transportable device system intended to be used to generate and control x-ray for diagnostic procedures. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.(b)
Classification. Class II.