Ultrasound elastography can quantify stiffness distribution of tissue lesions and complements conventional B-mode ultrasound for breast cancer screening. Recently, the development of computer-aided diagnosis has improved the reliability of the system, whilst the inception of machine learning, such as deep learning, has further extended its power by facilitating automated segmentation and tumour classification. The objective of this review was to summarize application of the machine learning model to ultrasound elastography systems for breast tumour classification. Review databases included PubMed, Web of Science, CINAHL, and EMBASE. Thirteen (n = 13) articles were eligible for review. Shear-wave elastography was investigated in six articles, whereas seven studies focused on strain elastography (5 freehand and 2 Acoustic Radiation Force). Traditional computer vision workflow was common in strain elastography with separated image segmentation, feature extraction, and classifier functions using different algorithm-based methods, neural networks or support vector machines (SVM). Shear-wave elastography often adopts the deep learning model, convolutional neural network (CNN), that integrates functional tasks. All of the reviewed articles achieved sensitivity ³ 80%, while only half of them attained acceptable specificity ³ 95%. Deep learning models did not necessarily perform better than traditional computer vision workflow. Nevertheless, there were inconsistencies and insufficiencies in reporting and calculation, such as the testing dataset, cross-validation, and methods to avoid overfitting. Most of the studies did not report loss or hyperparameters. Future studies may consider using the deep network with an attention layer to locate the targeted object automatically and online training to facilitate efficient re-training for sequential data.

Purpose: Evaluation of the internal biomechanics of the foot-and-ankle complex is challenging for the prescription of orthosis particularly for midfoot arthritis patients in which the joint condition is crucial. Methods: Using computational modeling and design optimization techniques, the objective of this study was to compare the biomechanical functions among different combinations of design factors using computer simulation. A finite element foot model was reconstructed from a midfoot arthritis patient Orthotic designs with 3 levels for each of the 3 design factors (arch height, lateral wedge angle, and insole stiffness) contributed to 9 configurations using a fractional factorial design were tested. Results: An increase in peak plantar stress of the midfoot was facilitated by a medium arch height and wedge angle, and stiffest insole material, notwithstanding the combination neither reduced the peak plantar stress of other foot regions nor was consistent with the combination that minimized the stress of the articular cartilage. Conclusions: Insole with high arch (H = 30 mm), low stiffness (E = 1.0 MPa), and medium wedge angle (A = 5 degrees) could minimize the stress of the cartilage at the arthritic joint (primal)/ outcome) and could be beneficial to the patients. Also, insole stiffness predominantly influenced cartilage stress. However, secondary outcomes including the stress of the navicular and medial cuneiform and the regional plantar stress did not produce the same solution. Future studies can consider a patient-specific loading profile to further the investigation on the stabilizing effect and the attenuation of load transfer induced by the insole.

Elastography complements traditional medical imaging modalities by mapping tissue stiffness to identify tumors in the endocrine system, and machine learning models can further improve diagnostic accuracy and reliability. Our objective in this review was to summarize the applications and performance of machine-learning-based elastography on the classification of endocrine tumors. Two authors independently searched electronic databases, including PubMed, Scopus, Web of Science, IEEEXpress, CINAHL, and EMBASE. Eleven (n = 11) articles were eligible for the review, of which eight (n = 8) focused on thyroid tumors and three (n = 3) considered pancreatic tumors. In all thyroid studies, the researchers used shear-wave ultrasound elastography, whereas the pancreas researchers applied strain elastography with endoscopy. Traditional machine learning approaches or the deep feature extractors were used to extract the predetermined features, followed by classifiers. The applied deep learning approaches included the convolutional neural network (CNN) and multilayer perceptron (MLP). Some researchers considered the mixed or sequential training of B-mode and elastographic ultrasound data or fusing data from different image segmentation techniques in machine learning models. All reviewed methods achieved an accuracy of ≥80%, but only three were ≥90% accurate. The most accurate thyroid classification (94.70%) was achieved by applying sequential training CNN; the most accurate pancreas classification (98.26%) was achieved using a CNN–long short-term memory (LSTM) model integrating elastography with B-mode and Doppler images.

While hallux valgus (HV) surgeries are useful for correcting skeletal alignment problems, their effects on plantar load, which reflects forefoot functions, are less understood. The objective of this study is to conduct a systematic review and meta-analysis on the plantar load change after HV surgeries. A systematic search of Web of Science, Scopus, PubMed, CENTRAL, EMBASE, and CINAHL was performed. Studies that assessed the pre- and post-operative plantar pressure of HV patients undergoing surgeries and reported load-related parameters over the hallux, medial metatarsal, and/or central metatarsal regions were included. Studies were appraised by using the modified NIH quality assessment tool for before-after study. Studies suitable for meta-analysis were pooled with the random-effects model, using the standardized mean difference of the before-after parameters as an effect measure. Twenty-six studies containing 857 HV patients and 973 feet were included for the systematic review. Meta-analysis was conducted on 20 of them, and most studies did not favor HV surgeries. Overall, HV surgeries reduced the plantar load over the hallux region (SMD −0.71, 95% CI, −1.15 to −0.26), indicating that forefoot function worsened after surgeries. For the other five outcomes, the overall estimates were not statistically significant, indicating that surgeries did not improve them either. There was substantial heterogeneity among the studies, which in most cases could not be resolved by pre-planned subgroup analyses by surgical classification, year of publication, median age of patients, and length of follow-up. Sensitivity analysis removing lower-quality studies showed that the load integrals (impulse) over the central metatarsal region significantly increased (SMD 0.27, 95% CI, 0 to 0.53), indicating that surgeries increased the risk of transfer metatarsalgia. There is no solid evidence that HV surgeries could improve forefoot functions from a biomechanical point perspective. Currently available evidence even suggests that surgeries might reduce the plantar load over the hallux and adversely affect push-off function. The reasons behind and the effectiveness of alternative surgical methods warrant further investigation.

To explore how the thickness of the femoral lateral wall influences the effectiveness of internal fixation systems used to treat intertrochanteric fractures. CT images of the pelvis and femur of a male adult were used to construct an intertrochanteric fracture model (AO/OTA 31-A2) with various thicknesses of the femoral lateral wall (FLW). Four finite element (FE) models were created with the lateral femoral walls being 10 mm, 20 mm, 30 mm, and 40 mm thick. The fracture models were fixed with a dynamic hip screw (DHS), a proximal femoral nail anti-rotation (PFNA), and a proximal femoral locking compression plate (P-FLCP). A simulated vertical load was applied to the femoral head. The stress and displacement of the implant and femur in each model were recorded for comparison. The FE analysis of the intertrochanteric fracture models showed that the PFNA system could provide better stability than the DHS and P-FLCP with the same thickness of FLW. The FLW provided buttress support to the femoral head and neck when using a DHS and PFNA, and the buttress strength was proportional to the thickness of FLW. The maximum stress in the DHS model was recorded on the DHS plate which accommodated the lag screw. For the PFNA model, the maximum stress appeared at the connection between the nail and blade. In the P-FLCP model, the maximum stresses were highly concentrated at the connection between the cephalic nails and the proximal plate. The thickness of the femoral lateral wall should be considered an important factor when selecting a suitable internal fixation system for intertrochanteric fractures. Based on the FE analysis, intramedullary fixation, such as PFNA, experiences lower stress levels and a moderate displacement in comparison to DHS and P-FCLP when used to treat intertrochanteric fractures.