Room: Track 2
Purpose: Arthritis is the most prevalent chronic health condition in the United States and Canada, with the two most common forms being osteoarthritis (OA) and rheumatoid arthritis (RA). OA and RA cause cartilage degradation, synovium inflammation, and subchondral bone remodeling, resulting in severe pain, mobility limitations, and a decreased quality-of-life. We have developed and aim to validate a handheld mechanical 3D ultrasound (3DUS) device against the current standard of MRI for monitoring knee cartilage degradation and synovium inflammation at the patient's bedside.
Methods: Knee images of 25 healthy volunteers were acquired using our 3DUS scanner with accompanying 3.0T MRI scans. The trochlear cartilage and suprapatellar synovium were manually segmented from both modalities using 3D Slicer. A semi-automated surface-based registration algorithm was used to register the 3DUS to MRI segmentations for comparison. Intra-rater and inter-rater reliabilities for the manual 3DUS segmentations were assessed by calculating intraclass correlation coefficient (ICC) values from the segmentation volumes. Mean surface distances (MSD), Hausdorff distances (HD), and Dice similarity coefficients (DSC) were calculated between segmentations.
Results: Intra-rater ICC values were 0.99 and 0.98 for raters 1 and 2 respectively, and inter-rater reliability was 0.95 between raters. Segmentation comparisons between intra-rater, inter-rater, and 3DUS to MRI registrations resulted in a global mean MSD of 0.28 mm, with DSC of 0.93 and 0.88 (intra-rater), 0.87 (inter-rater), and 0.75 (3DUS to MRI registration). The global mean HD between the 3DUS and MRI segmentations was 2.98 mm.
Conclusion: We have developed a handheld mechanical 3DUS device that can be used to provide accurate and precise measurements of the knee cartilage and synovium. 3DUS is a reliable imaging modality for monitoring OA and RA at the patient's bedside, and has the potential to increase the quality-of-life of patients suffering from knee arthritis by reducing the need for MRI prior to treatment.
Funding Support, Disclosures, and Conflict of Interest: This project is supported by the Canadian Institutes of Health Research (CIHR) and the Natural Sciences and Engineering Research Council of Canada (NSERC).