METHODS: We proposed a new feature extraction method by replacing fully-connected layer with global average pooling (GAP) layer. A comparative analysis was conducted to compare the efficacy of 16 different convolutional neural network (CNN) feature extractors and three machine learning classifiers.
RESULTS: Experimental results revealed the potential of CNN feature extractors in conducting multitask diagnosis. Optimal model consisted of VGG16-GAP feature extractor and KNN classifier. This model not only outperformed the other tested models, it also outperformed the state-of-art methods with higher balanced accuracy, higher Cohen's kappa, higher F1, and lower mean squared error (MSE) in seven OA features prediction.
CONCLUSIONS: The proposed model demonstrates pain prediction on plain radiographs, as well as eight OA-related bony features. Future work should focus on exploring additional potential radiological manifestations of OA and their relation to therapeutic interventions.
METHODS: Twenty-three collegiate male soccer players (22.1±1.7 years) were tested in four conditions; vertical jump (VJ), vertical jump immediately after slow running (VJSR), vertical jump immediately after sprinting (VJFR) and double horizontal jump immediately after sprinting (HJFR). The kinematics and kinetics data were measured using Vicon motion analyzer (100Hz) and two Kistler force platforms (1000Hz), respectively.
RESULTS: For knee flexion joint angle, (p = 0.014, η = 0.15) and knee valgus moment (p = 0.001, η = 0.71) differences between condition in the landing phase were found. For knee valgus joint angle, a main effect between legs in the jumping phase was found (p = 0.006, η = 0.31), which suggests bilateral deficit existed between the right and left lower limbs.
CONCLUSION: In brief, the important findings were greater knee valgus moment and less knee flexion joint angle proceeding sprint (HJFR & VJFR) rather than no sprint condition (VJ) present an increased risk for knee injuries. These results seem to suggest that running and sudden subsequent jumping-landing activity experienced during playing soccer may negatively change the knee valgus moment. Thus, sprinting preceding a jump task may increase knee risk factors such as moment and knee flexion joint angle.
METHODS: A total of 137 patients with 212 consecutive knees who underwent TKAs with or without functional stepwise MNP of superficial medial collateral ligament was recruited in this prospective cohort. Eighty-one patients with 129 knees who performed serial stress radiographs were enrolled in the final assessment. Superficial medial collateral ligament was punctured selectively (anteriorly or posteriorly or both) and sequentially depending on the site and degree of tightness. Mediolateral stability was assessed using serial stress radiographs and comparison was performed between the MNP and the non-MNP groups at postoperative 6 months and 1 year. Clinical outcomes were also evaluated between 2 groups.
RESULTS: Fifty-five TKAs required additional stepwise MNP (anterior needling 19, posterior needling 3, both anterior and posterior needling 33). Preoperative hip-knee-ankle angle and the difference in varus-valgus stress angle showed significant difference between the MNP and the non-MNP groups, respectively (P = .009, P = .037). However, there was no significant difference when comparing the varus-valgus stress angle between the MNP and the non-MNP groups during serial assessment. Clinical outcomes including range of motion also showed no significant differences between the 2 groups.
CONCLUSION: Functional medial ligament balancing with stepwise MNP can provide sufficient medial release with safety in TKA with varus aligned knee without clinical deterioration or complication such as instability.
LEVEL OF EVIDENCE: Level II, Prospective cohort study.