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.
Aim: To evaluate parameters of the proximal tibia geometry amongst the Igbos of South Eastern Nigeria and discuss the clinical implications in total knee replacement.
Methods: The proximal tibia parameters of 558 adult tibiae bones, 300 right-sided and 258 left-sided, collected from the osteological unit of the department of anatomy University of Nigeria Enugu campus were measured using a Venier calipers.
Results: The values of the determined parameters are as follows: anterior posterior dimension (APD): 5.50 cm, medial lateral dimension (MLD):7.53 ± 5.56 cm, proximal length (PL): 7.32 ± 0.67 cm, proximal width (PW) CM: 4.10 ± 0.30 cm, medial tibia posterior slope (MTPS):11.36 ± 4.15° and lateral tibia posterior slope (LTPS):5.65 ± 2.72°. The correlation test between these parameters of the proximal tibia shows a strong significant correlation between LPTS and MPTS (r =0.814, P < .001). The comparison of the means of the MPTS of both sides using independent samples t test shows a mean difference that is not significant (p =0.628). A comparison of the values with other populations shows significant mean difference for MLD with that of Chinese and American populations (p < .05). The mean difference between the APD of this study and the Chinese is significant (p =0.007). A comparison of MPTS and LPTS for Malaysian, Chinese, and white Americans shows a significance difference for all the three populations (p tibia surface coverage in total knee replacement amongst the Igbo population.
MATERIALS AND METHODS: A retrospective study of patients diagnosed as benign bone tumours according to the Enneking classification who underwent simple or extended curettage at Menoufia university-Orthopedic Oncology Division (with or without grafting or filling) during the surgical treatment (Jan 2015 to Feb 2020). A review of the medical records was done. Lesions' size (length, width and depth) was measured on plain radiographs using the image j program. When applicable, degrees of filling of the resultant cavity were classified into four categories, according to Modified Neer's classification. Functional evaluation using the musculoskeletal tumour society (MSTS) score was reviewed.
RESULTS: Overall, 88 patients diagnosed with a primary bone tumour and who received the surgical intervention were included in the study. The mean age of the patients was 22.61+13.497 (3-58) years. There were 48 males and 40 females (54 right and 34 left). The mean follow-up period was 28.09+16.13 months. The most common location was the distal femur in 15 patients, the proximal femur in 10 patients and the proximal tibia in 12 patients. The most common diagnosis was giant cell tumour in 20 patients, followed by UBC in 19 patients, ABC in 15 patients and enchondroma in 13 patients. Twenty-three patients had simple curettage, while 65 patients had extended curettage. Mean MSTS was 28.78±1.68. Fifty-five lesions were classified according to modified Neer's classification.Thirtty-two patients were classified as type 1 with complete healing,22 patient was classified as type 2 with partial healing, and only one was classified as a recurrent lesion. Seven patients (7.9%) developed local recurrences.
CONCLUSION: Filling the resulting cavity after the removal of the pathological tissues is usually necessary but not always required. This is determined by the type of lesion and the size of the resulting cavity following curettage. Individualised surgery is required; additional fixation should be considered.
MATERIALS AND METHODS: Results that are possible to be compared in more than two articles were presented as forest plots. A 95% confidence interval was calculated for each effect size, and we calculated the I 2 statistic, which presents the percentage of total variation attributable to the heterogeneity among studies. The random effects model was used to calculate the effect size.
RESULTS: Seven articles were included to the final analysis. Case groups were composed of HTO without concurrent procedures and control groups were composed of HTO with concurrent procedures such as marrow stimulation procedure, mesenchymal stem cell transplantation, and injection. The case group showed a higher hospital for special surgery score and mean difference was 4.10 [I 2 80.8%, 95% confidence interval (CI) - 9.02 to 4.82]. Mean difference of the mechanical femorotibial angle in five studies was 0.08° (I 2 0%, 95% CI - 0.26 to 0.43). However, improved arthroscopic, histologic, and MRI results were reported in the control group.
CONCLUSION: Our analysis support that concurrent procedures during HTO for medial compartment OA have little beneficial effect regarding clinical and radiological outcomes. However, they might have some beneficial effects in terms of arthroscopic, histologic, and MRI findings even though the quality of healed cartilage is not good as that of original cartilage. Therefore, until now, concurrent procedures for medial compartment OA have been considered optional. Nevertheless, no conclusions can be drawn for younger patients with focal cartilage defects and concomitant varus deformity. This question needs to be addressed separately.