Sharif Digital Repository

In this paper, we have trained several deep convolutional networks with introduced training techniques for classifying X-ray images into three classes: normal, pneumonia, and COVID-19, based on two open-source datasets. Our data contains 180 X-ray images that belong to persons infected with COVID-19, and we attempted to apply methods to achieve the best possible results. In this research, we introduce some training techniques that help the network learn better when we have an unbalanced dataset (fewer cases of COVID-19 along with more cases from other classes). We also propose a neural network that is a concatenation of the Xception and ResNet50V2 networks. This network achieved the best... 

Background The ratio of total lumbar rotation over pelvic rotation (lumbopelvic rhythm) during trunk sagittal movement is essential to evaluate spinal loads and discriminate between low back pain and asymptomatic population. Methods Angular rotations of the pelvis and lumbar spine as well as their sagittal rhythm during forward flexion and backward extension in upright standing of eight asymptomatic males are measured using an inertial tracking device. The effect of variations in the lumbopelvic ratio during trunk flexion on spinal loads is quantified using a detailed musculoskeletal model. Findings The mean of peak voluntary flexion rotations of the thorax, pelvis, and lumbar was 121 (SD... 

Background: Work-related risk factors of low back disorders have been identified to be external moments, awkward postures, and asymmetrical dynamic lifting amongst others. The distinct role of asymmetry of load versus posture is hard to discern from the literature. Hence, the aim of this study is to measure isometric trunk exertions at upright standing posture at different exertion level and degree of asymmetry to further delineate the effects of exertion level and asymmetry on neuromuscular capability response. Methods: Fifteen healthy volunteers randomly performed trunk exertions at three levels (30%, 60%, and 100% of maximum voluntary exertion and five different angles (0°, 45°, 90°,... 

A computational method for simulation of 3-D movement of the trunk under the control of 48 anatomically oriented muscle actions was developed. Neural excitation of muscles was set based on inverse dynamics approach along with the stability-based optimization. The effect of muscle spindle reflex response on the trunk movement stability was evaluated upon the application of a perturbation moment. The method was used to simulate the trunk movement from the upright standing to 60° of flexion. Incorporation of the stability condition as an additional constraint in the optimization resulted in an increase in antagonistic activities demonstrating that the antagonistic co-activation acts to increase... 

Cardiovascular diseases are very common and are one of the main reasons of death. Being among the major types of these diseases, correct and in-time diagnosis of coronary artery disease (CAD) is very important. Angiography is the most accurate CAD diagnosis method; however, it has many side effects and is costly. Existing studies have used several features in collecting data from patients, while applying different data mining algorithms to achieve methods with high accuracy and less side effects and costs. In this paper, a dataset called Z-Alizadeh Sani with 303 patients and 54 features, is introduced which utilizes several effective features. Also, a feature creation method is proposed to... 

An essential input to the musculoskeletal (MS) trunk models that estimate muscle and spine forces is kinematics of the thorax, pelvis, and lumbar vertebrae. While thorax and pelvis kinematics are usually measured via skin motion capture devices (with inherent errors on the proper identification of the underlying bony landmarks and the relative skin-sensor-bone movements), those of the intervening lumbar vertebrae are commonly approximated at fixed proportions based on the thorax-pelvis kinematics. This study proposes an image-based kinematics measurement approach to drive subject-specific (musculature, geometry, mass, and center of masses) MS models. Kinematics of the thorax, pelvis, and... 

Whole-body vibration (WBV) has been identified as one of the serious risk factors leading to spinal disorders, particularly in professional drivers. Although the influential factors in this area have been investigated epidemiologically, finite element (FE) modeling can efficiently help us better understand the problem. In this study, a modified HYBRID III dummy FE model which was enhanced by detailed viscoelastic discs in the lumbar region was utilized to simulate the effect of WBV on lumbar spine loads. Spinal responses to the vertical sinusoidal vibrations of a generic seat were obtained and spinal injury risk factors were calculated. Effects of variation of excitation frequencies, three... 

Purpose: Cardiac multimodal image fusion can offer an image with various types of information in a single image. Many coronary stenosis, which are anatomically clear, are not functionally significant. The treatment of such kind of stenosis can cause irreversible effects on the patient. Thus, choosing the best treatment planning depend on anatomical and functional information is very beneficial. Methods: An algorithm for the fusion of coronary computed tomography angiography (CCTA) as an anatomical and transthoracic echocardiography (TTE) as a functional modality is presented. CCTA and TTE are temporally registered using manifold learning. A pattern search optimization algorithm, using... 

Freezing of gait (FOG) is a debilitating symptom in patients with Parkinson's disease (PD), which may be associated with motor control impairments in tasks other than gait. This study aimed to examine whether symmetric and asymmetric bimanual coordination is impaired in PD with FOG (PD +FOG) patients and whether dual-task and drug phases may affect bimanual coordination in these patients. Twenty PD +FOG patients, 20 PD patients without FOG (PD –FOG) performed symmetric and asymmetric functional bimanual tasks (reach to and pick up a box and open a drawer to press a pushbutton inside it, respectively) under single-task and dual-task conditions. PD patients were evaluated during on- and... 

Two artificial neural networks (ANNs) are constructed, trained, and tested to map inputs of a complex trunk finite element (FE) model to its outputs for spinal loads and muscle forces. Five input variables (thorax flexion angle, load magnitude, its anterior and lateral positions, load handling technique, i.e., one- or two-handed static lifting) and four model outputs (L4-L5 and L5-S1 disc compression and anterior-posterior shear forces) for spinal loads and 76 model outputs (forces in individual trunk muscles) are considered. Moreover, full quadratic regression equations mapping input-outputs of the model developed here for muscle forces and previously for spine loads are used to compare the... 

Background: Activities involving axial trunk rotations/moments are common and are considered as risk factors for low back disorders. Previous biomechanical models have failed to accurately estimate the trunk maximal axial torque exertion. Moreover, the trunk stability under maximal torque exertions has not been investigated. Methods: A nonlinear thoracolumbar finite element model along with the Kinematics-driven approach is used to study biomechanics of maximal axial torque generation during upright standing posture. Detailed anatomy of trunk muscles with six distinct fascicles for each abdominal oblique muscle on each side is considered. While simulating an in vivo study of maximal axial... 

Maximal strength measurements of the trunk have been used to evaluate the maximum functional capacity of muscles and the potential mechanical overload or overuse of the lumbar spine tissues in order to estimate the risk of developing musculoskeletal injuries. A new triaxial isometric trunk strength measurement system was designed and developed in the present study, and its reliability and performance was investigated. The system consisted of three main revolute joints, equipped with torque sensors, which intersect at L5-S1 and adjustment facilities to fit the body anthropometry and to accommodate both symmetric and asymmetric postures in both seated and standing positions. The dynamics of...