Due to false negative results of the real-time Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) test, the complemental practices such as computed tomography (CT) and X-ray in combination with RT-PCR are discussed to achieve a more accurate diagnosis of COVID-19 in clinical practice. Since radiology includes visual understanding as well as decision making under limited conditions such as uncertainty, urgency, patient burden, and hospital facilities, mistakes are inevitable. Therefore, there is an immediate requirement to carry out further investigation and develop new accurate detection and identification methods to provide automatically quantitative evaluation of COVID-19. In this paper, we propose a new computer-aided diagnosis application for COVID-19 detection using deep learning techniques. A new technique, which receives symmetric X-ray data as the input, is presented in this study by combining Convolutional Neural Networks (CNN) with Ant Lion Optimization Algorithm (ALO) and Multiclass Naïve Bayes Classifier (NB). Moreover, several other classifiers such as Softmax, Support Vector Machines (SVM), K-Nearest Neighbors (KNN) and Decision Tree (DT) are combined with CNN. The promising results of these classifiers are evaluated and presented for accuracy, precision, and F1-score metrics. NB classifier with Ant Lion Optimization Algorithm and CNN produced the best results with 98.31% accuracy, 100% precision and 98.25% F1-score and with the lowest execution time.

No abstract

IntroductionSpasticity is a common sensory-motor control disorder characterized by increased velocity-dependent stretch reflex responses resulting from upper motor neuron (UMN) lesions (1). Spasticity is frequently observed in cases such as spinal cord injury, multiple sclerosis, traumatic brain damage, cerebral palsy, and stroke, which can be accompanied by cerebral and spinal pathology that is dispersed or regional (2). The pathophysiology of spasticity is complicated and it is rather difficult to understand the underlying mechanism since it is necessary to know the pathophysiological mechanism that would differentiate the UMN syndrome from other symptoms (3,4). Therefore, spasticity is a disorder that is both insufficiently defined and measured (5). There are three major approaches, clinical, neurophysiological, and biomechanical, for assessing spasticity.The Ashworth Scale (AS) and the modified Ashworth Scale (MAS) are the most commonly used clinical measures of spasticity. The validity, reliability, and sensitivity of the clinical scales, which are subject to interpretation and generally intensified upon passive movement resistance, are debatable (6-8). As a consequence, many studies have been conducted using biomechanical and electrophysiological measurements with the objective of evaluating spasticity, and the obtained results have been used to make correlations with the clinical scales (8-14). Surface electromyography (EMG) has been commonly used in electrophysiological measurements of spasticity (14-16). However, parameters extracted from surface EMG have been generally obtained from analysis by using only the time domain in previous studies, and results were generally based on correlations with the clinical scales rather than using stand-alone surface EMG as well as considering causality. Due to the fact that surface EMG signals are nonstationary and random signals, some pathological data might not be discriminated in the time axis. Therefore, it is possible to utilize the short-time Fourier transform (STFT) and the wavelet transform (WT) methods from the time and the frequency components of the signals found in the pathological symptoms (17,18).Background/aim: Spasticity is generally defined as a sensory-motor control disorder. However, there is no pathophysiological mechanism or appropriate measurement and evaluation standards that can explain all aspects of a possible spasticity occurrence. The objective of this study is to develop a fuzzy logic classifier (FLC) diagnosis system, in which a quantitative evaluation is performed by surface electromyography (EMG), and investigate underlying pathophysiological mechanisms of spasticity.Materials and methods: Surface EMG signals recorded from the tibialis anterior and medial gastrocnemius muscles of hemiplegic patients with spasticity and a healthy control group were analyzed in standing, resting, dorsal flexion, and plantar flexion positions. The signals were processed with different methods: by using their amplitudes in the time domain, by applying short...

No abstract

Background:Diagnosis of carpal tunnel syndrome is based on clinical symptoms, examination findings, and electrodiagnostic studies. For carpal tunnel syndrome, the most useful of these are nerve conduction studies. However, nerve conduction studie can result in ambiguous or false-negative results, particularly for mild carpal tunnel syndrome. Increasing the number of nerve conduction studie tests improves accuracy but also increases time, cost, and discomfort. To improve accuracy without additional testing, the terminal latency index and residual latency are additional calculations that can be performed using the minimum number of tests. Recently, the median sensory-ulnar motor latency difference was devised as another way to improve diagnostic accuracy for mild carpal tunnel syndrome.Aims:The median sensory-ulnar motor latency difference, terminal latency index, and residual latency were compared for diagnostic accuracy according to severity of carpal tunnel syndrome.Study Design:Diagnostic accuracy study.Methods:A total of 657 subjects were retrospectively enrolled. The carpal tunnel syndrome group consisted of 546 subjects with carpal tunnel syndrome according to nerve conduction studie (all severities). The control group consisted of 121 subjects with no hand symptoms and normal nerve conduction studie. All statistical analyses were performed using SAS v9.4. Means were compared using one-way ANOVA with the Bonferroni adjustment. Sensitivity, specificity, positive predictive value, and negative predictive value were compared, including receiver operating characteristic curve analysis.Results:For mild carpal tunnel syndrome, the median sensory-ulnar motor latency difference showed higher specificity and positive predictive value rates (0.967 and 0.957, respectively) than terminal latency index (0.603 and 0.769, respectively) and residual latency (0.818 and 0.858, respectively). The area under the receiver operating characteristic was highest for the median sensory-ulnar motor latency difference (0.889), followed by the residual latency (0.829), and lastly the terminal latency index (0.762). Differences were statistically significant (median sensory-ulnar motor latency difference being the most accurate). For moderate carpal tunnel syndrome, sensitivity and specificity rates of residual latency (0.989 and 1.000) and terminal latency index (0.983 and 0.975) were higher than those for median sensory-ulnar motor latency difference (0.866 and 0.958). Differences in area under the receiver operating characteristic curve were not significantly significant, but median sensory-ulnar motor latency difference sensitivity was lower. For severe carpal tunnel syndrome, residual latency yielded 1.000 sensitivity, specificity, positive predictive value, negative predictive value and area beneath the receiver operating characteristic curve. Differences in area under the receiver operating characteristic curve were not significantly different.Conclusion:The median sensory-ulnar motor latency difference is the best calculated para...