Algorithm Developed to Segment the Meniscus in MRIs
Semi-automated technique produces results consistent with those of manual segmentations
MONDAY, Feb. 1 (HealthDay News) -- Semi-automated segmentation of magnetic resonance images can accurately assess the lateral meniscus in patients with and without knee osteoarthritis, according to a study published in Osteoarthritis and Cartilage.
Michele S. Swanson, Ph.D., of the Ohio State University in Columbus, and colleagues developed an algorithm to semi-automatically segment the meniscus, and evaluated it on 10 baseline magnetic resonance images obtained from subjects with no evidence of knee osteoarthritis and 14 images from subjects with established knee osteoarthritis.
The researchers found that the semi-automated segmentation method produced accurate and consistent segmentations with results similar to those found by trained observers. They calculated an average similarity index of more than 0.80 for subjects without knee osteoarthritis, and 0.75, 0.67, and 0.64 for subjects with established knee osteoarthritis with Osteoarthritis Research Society International joint space narrowing scores of zero, one, and two, respectively.
"Future efforts will expand the algorithm to include the medial meniscus in order to provide a more comprehensive examination of the knee," the authors conclude. "Given the more complicated structure of the lateral meniscus, segmentation should also be feasible for the medial meniscus using similar methodology."