Search Results

ScanDiags Ortho L-Spine MR-Q software is an image-processing and measurement software tool that provides quantitative spine measurements from previously-acquired DICOM lumbar spine Magnetic Resonance (MR) images for users' review, analysis, and interpretation. It provides the following functionality to assist users in visualizing, and documenting area and distance measurements of relevant anatomical structures (vertebral body, intervertebral disc, neuroforamina, thecal sac) of the lumbar spine:

Feature Segmentation;

Feature measurement; and

Export of measurement results to a PDF report containing measurement results and overlay images for user's review, revise and approval.

ScanDiags Ortho L-Spine MR-Q software does not produce or recommend any type of medical diagnosis or treatment. Instead, it simply helps users to more easily identify and classify features in lumbar MR images and compile a report. The user is responsible for reviewing and verifying the software-generated measurements and approving draft report content using their medical judgment and discretion.

The device is intended to be used only by hospitals and other medical institutions.

Only DICOM images of MRI acquired from lumbar spine exams of patients aged 22 and above are considered to be valid input. ScanDiags Ortho L-Spine MR-Q software does not support DICOM images of patients that are pregnant, undergo MRI scan with contrast media, have post-operational complications or infections.

ScanDiags Ortho L-Spine MR-Q software is a software as a medical device (SaMD) intended for visualization, and quantification of lumbar spine anatomical structures including vertebral bodies, intervertebral discs, neuroforamina, thecal sacs from a set of standard lumbar spine MRI images in DICOM (Digital Imaging and Communications in Medicine) format. The semi-automatic segmentation of these structures forms the bases for the distance and area measurement outputs. The software has features for log-in, viewing, revising, and saving measurement results in addition to generating PDF reports. The PDF report includes images, measurements.

ScanDiags Ortho L-Spine MR-Q software includes a viewing application (ScanDiags DICOM Viewer) to visualize, review, and apply corrections to the measurement results shown as overlay on the original lumbar spine MRI images. Pre-existing MR images of the lumber spine are uploaded into the software for analysis. The semi-automatic segmentations are based on deep convolutional neural networks (DCNN) which are developed by applying well-established supervised deep learning methods on unstructured MRI scans (DICOM image format). ScanDiags Ortho L-Spine MR-Q combines deep learning, image analysis, as well as regression-based machine methods. The segmentations and distance measurements are user modifiable. Results are reviewed and approved by the radiologist's user before a PDF report is generated. Once approved, the result PDF report is sent to the clinician's PACS system. The PACS system stores the PDF report within the corresponding patient MRI study.

ScanDiags Ortho L-Spine MR-Q does not interface directly with any MR or data collection equipment; instead, the software uploads data files previously generated by such equipment. Its functionality is independent of the vendor type of the acquisition equipment. The analysis results are available on the ScanDiags DICOM Viewer screen and can be edited, saved, and approved. The approved measurement results are sent back to the PACS system as a Measurement Result PDF Report. The software does not perform any functions that could not be accomplished by a trained user with manual tracing methods; the purpose of the software is to save time and automate the tedious manual task of segmentation and distance measurement.

ScanDiags Ortho L-Spine MR-Q software is an adjunct tool and is not intended to replace a radiologist's review of the MRI study, nor is it intended to replace his or her clinical judgment, and it does not detect, diagnose or identify any abnormalities. Radiologists must not use the generated output as a primary interpretation.

Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) submission information:

Acceptance Criteria and Reported Device Performance

The device, ScanDiags Ortho L-Spine MR-Q, uses machine learning (Deep Convolutional Neural Networks) for semi-automatic segmentation and quantitative measurements of lumbar spine anatomical structures from MRI images. Its performance was evaluated against ground truth established by expert radiologists.

Table of Acceptance Criteria and Reported Device Performance:

Note: The document states "The device successfully passed the primary ICC acceptance criteria," "The device successfully passes the secondary DICE acceptance criteria," and "The device successfully passes the co-secondary MAE acceptance criteria," implying that specific thresholds were met, though the exact numerical criteria are not explicitly stated in this summary.

Study Details Proving Acceptance Criteria

1. Sample Size and Data Provenance:

   • Test Set Sample Size: 100 individual patient MRI studies.
   • Data Provenance: Retrospective, multicenter study. Data collected from two hospital groups in the United States: one in Missouri (18 patients from a rural hospital group) and one in North Carolina (82 patients from urban and rural hospital groups). Images were acquired from MRI systems from GE (40), Siemens Healthineers (42), and Philips (18).

2. Number of Experts and Qualifications:

   • Number of Experts: 3.
   • Qualifications: US board-certified MSK (Musculoskeletal) radiologists. (Specific years of experience are not mentioned, but board certification implies a certain level of expertise).

3. Adjudication Method for Ground Truth:

   • For Anatomic Structure Segmentation: Pixel-based majority opinion between the three radiologists.
   • For Area and Distance Measurements: Averaging the measurements of all three readers.

4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

   • No, a MRMC comparative effectiveness study was not done. The study focuses on the standalone performance of the algorithm against expert-derived ground truth. The device is described as an "adjunct tool" that requires human review and verification, but the provided performance data is for the algorithm itself, not human-AI collaboration.

5. Standalone Performance Study:

   • Yes, a standalone (algorithm only without human-in-the-loop performance) study was performed. The "Machine Learning Performance Evaluation Summary" clearly outlines the results of the algorithm's performance (ICC, DSC, MAE) against the established ground truth.

6. Type of Ground Truth Used:

   • Expert Consensus: The ground truth was established by the consensus (majority opinion for segmentation, average for measurements) of three US board-certified MSK radiologists.

7. Training Set Sample Size:

   • The document states that images and cases used for verification and validation testing were "separate and carefully segregated from training datasets." However, the sample size for the training set is not provided in the excerpt. It mentions that the Deep Convolutional Neural Networks (DCNN) were developed by "applying well-established supervised deep learning methods on unstructured MRI scans (DICOM image format)."

8. How Ground Truth for Training Set was Established:

   • The document implies that the DCNN utilized "supervised deep learning methods." While it doesn't explicitly detail the ground truth establishment for the training set, it can be inferred that it involved labeled data, likely expert annotations, similar to how the test set ground truth was established, given the nature of supervised learning for medical imaging segmentation and measurement. However, the specific process (e.g., number of annotators, their qualifications, adjudication method) for the training set is not described in this provided text.

Ask a specific question about this device

The MR Q and the MR SUPINE are intended to perform:

• Posterior capsulotomy -
• Iridotomy -

The MR Q SLT in YAG mode is intended to perform:

• Posterior capsulotomy -
• -Iridotomy

The MR Q SLT in SLT mode is intended to perform:

• Selective laser trabeculoplasty

Not Found

I am sorry, but the provided text does not contain the detailed information necessary to answer your request about the acceptance criteria, study details, and specific performance metrics of the device.

The document is a 510(k) clearance letter from the FDA, which confirms that the device (Meridian AG's MR Q, MR Q SUPINE, MR Q SLT) is substantially equivalent to legally marketed predicate devices for its stated indications for use. It outlines regulatory requirements and general information but does not include the results of clinical studies, acceptance criteria, or detailed performance data for the device.

Specifically, the document does not provide:

• A table of acceptance criteria and reported device performance.
• Sample sizes for test sets, data provenance, number of experts for ground truth, or adjudication methods for test sets.
• Information on Multi-Reader Multi-Case (MRMC) studies or the effect size of AI assistance.
• Results from standalone algorithm performance studies.
• Details on the type of ground truth used.
• Sample sizes for training sets or how ground truth for training data was established.

The "Indications for Use" section (page 3) describes what the device is intended to do (e.g., Posterior capsulotomy, Iridotomy, Selective laser trabeculoplasty), but not the performance metrics or studies used to demonstrate those capabilities.

Ask a specific question about this device