(35 days)
The REGIUS Unitea software is intended for installation on an off-the-shelf PC meeting or exceeding minimum specifications. The REGIUS Unitea software primarily facilitates processing and presentation of medical images on display monitors suitable for the medical task being performed. The REGIUS Unitea software can process and display medical images from the following modality types: Plain X-ray Radiography, X-ray Computed Tomography, Magnetic Resonance imaging, Ultrasound, Nuclear Medicine and other DICOM compliant modalities. The REGIUS Unitea must not be used for primary image diagnosis in mammography.
REGIUS Unitea is a software which is intended for installation on an off-the-shelf PC meeting or exceeding minimum specifications. REGIUS unitea software controls and manages the cassette type CR (Computed Radiography) such as REGIUS MODEL 170,190 and 110 that is connected via the network. REGIUS Unitea receives and displays images from other DICOM compliant modalities connected via the network and whole use digital media such as DVD/CD-R, DSC and USB memory cards connected as disk drives.
REGIUS Unitea software has the following set of features.
- Feature to automatically obtain patient demographic information (Name, Age, Sex, Date of Birth, etc) from Hospital Information Systems.
- Feature to specify the reading condition (Sampling Sensitivity of the sensor and so on) of the connected CR device.
- Receive and store image data from REGIUS MODEL 170, 190 and 100 CR, other DICOM compliant modalities and digital media (such as DVD/CD-R, DSC and USB memory cards).
- Display image data received from the REGIUS MODEL 170, 190 and 100 CR, other DICOM compliant modalities and digital media (such as DVD/CD-R, DSC and USB memory cards).
- Feature to apply image processing to images received from REGIUS MODEL 170. 190 and 100 CR.
This 510(k) summary (K071436) describes a medical image processing workstation called REGIUS Unitea. The submission primarily focuses on establishing substantial equivalence to a predicate device (REGIUS CS-2000 and CS-3000, K051523) and details the device's features and intended use.
Crucially, this 510(k) summary does NOT contain information about specific acceptance criteria or an analytical study proving the device meets those criteria, as it relates to performance metrics like accuracy, sensitivity, or specificity for a specific medical task.
Instead, the submission for REGIUS Unitea, being a "Picture archiving and communications system" (PACS) product (21 CFR 892.2050), falls under a general controls regulatory pathway. For such devices, the primary "acceptance criteria" revolve around demonstrating that the device functions as intended, handles various image modalities, complies with DICOM standards, and does not introduce new safety or efficacy concerns compared to a legally marketed predicate device.
The study described is largely a comparative analysis for substantial equivalence, not a performance study measuring clinical accuracy or effectiveness in a typical sense for an AI/CAD product.
Given the information provided, here's a breakdown of the requested points:
1. Table of Acceptance Criteria and Reported Device Performance
Based on the provided K071436 summary, there are no explicit quantitative acceptance criteria or reported device performance metrics (e.g., sensitivity, specificity, AUC) for a specific diagnostic task. The acceptance is primarily based on:
| Acceptance Criteria Category | Reported Device Performance/Compliance |
|---|---|
| Functional Equivalence | "REGIUS Unitea is substantially equivalent to our REGIUS CS-2000 and CS-3000, 510(k) number: K051523. Comparison of the principal characteristics of these devices is shown in the Section 3." (Section 3 is not fully provided here, but it would detail feature-by-feature comparison). The new device offers similar image processing, display, and management capabilities to the predicate. |
| Safety and Efficacy (New Issues) | "REGIUS Unitea introduces no new safety and efficacy issues other than those already identified with the predicate device. The results of a hazard analysis, combined with the appropriate preventive measure taken indicate that the device is of minor level of concern as per the May 11, 2005 issue of the "Guidance for the Content of Premark Submissions for Software Contained in Medical Devices"." |
| Intended Use Compliance | Intended for installation on off-the-shelf PC, processing and presentation of medical images on display monitors suitable for the medical task. Supports various modalities (Plain X-ray Radiography, CT, MRI, Ultrasound, Nuclear Medicine, other DICOM compliant). Explicitly not for primary image diagnosis in mammography. This matches the scope of a PACS device. |
| DICOM Compliance | Implicitly stated throughout in relation to receiving/displaying images from "DICOM compliant modalities" and outputting to "other DICOM devices." |
| Connectivity/Compatibility | Controls/manages cassette-type CR (REGIUS MODEL 170, 190, 110), receives/displays images from other DICOM modalities, utilizes digital media (DVD/CD-R, DSC, USB). |
| Image Processing Features | The device successfully implements a range of image processing features as described (Contrast/Density adjustments, F-processing, E-processing, H-processing, I-processing, Masking, Rotating/Flipping, Re-sampling/Resizing, Stitching, Grid Suppression, Digital marker/Annotation). The "acceptance" here is that these features are present and function as specified, presumably verified through internal testing. |
2. Sample Size Used for the Test Set and Data Provenance
Given this is a PACS device submission for substantial equivalence, there is no mention of a specific "test set" in the context of diagnostic performance (e.g., a set of patient cases to evaluate diagnostic accuracy).
The evaluation typically involves:
- Verification and Validation (V&V) testing: This would involve testing the software's functionalities (e.g., image loading, display, processing, storage, network communication) using various types of DICOM images and system configurations.
- Hazard Analysis: To identify and mitigate risks.
The K071436 does not specify the sample size of images or the origin of data used for these V&V activities. Since it's a software device interacting with CR systems, the "data" would consist of DICOM images (potentially synthetic, proprietary, or from various sites) to confirm proper handling. It's likely a mix of internal, proprietary data, and potentially public DICOM conformance test images. The provenance is not specified (e.g., country of origin, retrospective/prospective).
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
Not applicable. This submission doesn't describe a study requiring diagnostic "ground truth" established by medical experts. A PACS workstation stores, retrieves, processes, and displays images; it doesn't make diagnostic interpretations itself requiring ground truth for AI performance evaluation. The "ground truth" for its functions would be whether it accurately performs the specified operations (e.g., does it correctly adjust contrast, does it store the image without corruption, does it display it correctly according to the DICOM header). These are typically verified by engineers and quality assurance personnel, not medical experts establishing diagnostic ground truth.
4. Adjudication Method for the Test Set
Not applicable. As no diagnostic performance test set is described, there is no adjudication method mentioned.
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 device is a PACS workstation, not an AI or CAD system intended to assist readers in diagnostic interpretation. Therefore, no MRMC study or assessment of human reader improvement with AI assistance was performed or reported.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
Not applicable. The REGIUS Unitea is an image management and processing system, entirely human-in-the-loop, as it's a tool for medical professionals to view and manipulate images. It does not perform any standalone diagnostic analysis or algorithmic interpretation where "algorithm only" performance would be relevant.
7. The Type of Ground Truth Used
Not applicable in the diagnostic sense. The "ground truth" for a PACS system would pertain to its functional performance and adherence to standards:
- Functional correctness: Does the software perform its operations (e.g., image processing, storage, display) as specified? (Verified by testing against software requirements and design specifications).
- DICOM conformance: Does it correctly interpret and generate DICOM objects? (Verified using DICOM conformance tools and test images).
- Image integrity: Are images stored and displayed without degradation or loss of information? (Verified by comparing processed/stored images with their originals, often using objective quality metrics or visual inspection).
8. The Sample Size for the Training Set
Not applicable. This device does not use machine learning or AI models that require a "training set." It is a software application developed using traditional programming paradigms.
9. How the Ground Truth for the Training Set Was Established
Not applicable. As there is no training set for an AI model, there is no corresponding ground truth establishment process.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).