INTRODUCTION: Magnetic resonance imaging is a useful technique to visualize soft tissues within the knee joint. Cartilage delineation in magnetic resonance (MR) images helps in understanding the disease progressions. Convolutional neural networks (CNNs) have shown promising results in computer vision tasks, and various encoder-decoder-based segmentation neural networks are introduced in the last few years. However, the performances of such networks are unknown in the context of cartilage delineation.
METHODS: This study trained and compared 10 encoder-decoder-based CNNs in performing cartilage delineation from knee MR images. The knee MR images are obtained from the Osteoarthritis Initiative (OAI). The benchmarking process is to compare various CNNs based on physical specifications and segmentation performances.
RESULTS: LadderNet has the least trainable parameters with the model size of 5 MB. UNetVanilla crowned the best performances by having 0.8369, 0.9108, and 0.9097 on JSC, DSC, and MCC.
CONCLUSION: UNetVanilla can be served as a benchmark for cartilage delineation in knee MR images, while LadderNet served as an alternative if there are hardware limitations during production.
METHODS: Osteoarthritis was induced at the right knee of sheep by complete resection of ACL and medial meniscus. Stem cells from sheep were induced to chondrogenic lineage. Test sheep received 5 mls single doses of 2 × 107 autologous PKH26-labelled ADSCs or BMSCs, while controls received basal medium. Functional recovery of the knees was evaluated via electromyography.
RESULTS: Induced ADSCs had 625, 255, 393, 908, 409, 157 and 1062 folds increases of collagen I, collagen II, aggrecan, SOX9, cartilage oligomeric protein, chondroadherin and fibromodullin compare to uninduced cells, while BMSCs had 702, 657, 321, 276, 337, 233 and 1163 respectively; p = .001. Immunocytochemistry was positive for these chondrogenic markers. 12 months post-treatment, controls scored 4 in most regions using ICRS, while the treated had 8; P = .001. Regenerated cartilages were positive to PKH26 and demonstrated the presence of condensing cartilages on haematoxylin and eosin; and Safranin O. OA degenerations caused significant amplitude shift from right to left hind limb. After treatments, controls persisted with significant decreases; while treated samples regained balance.
CONCLUSIONS: Both ADSCs and BMSCs had increased chondrogenic gene expressions using TGF-β3 and BMP-6. The treated knees had improved cartilage scores; PKH26 can provide elongated tracking, while EMG results revealed improved joint recoveries. These could be suitable therapies for osteoarthritis.