Prediction of salivary cortisol level by electroencephalography features.

Change in cortisol affects brain EEG signals. So, the identification of the significant EEG features which are sensitized to cortisol concentration was the aim of the present study. From 468 participated healthy subjects, the salivary samples were taken to test the cortisol concentration and EEG signal recording was done simultaneously. Then, the subjects were categorized into three classes based on the salivary cortisol concentration (<5, 5-15 and >15 nmol/l). Some linear and nonlinear features extracted and finally, in order to investigate the relationship between cortisol level and EEG features, the following steps were taken on features in sequence: Genetic Algorithm, Neighboring Component Analysis, polyfit, artificial neural network and support vector machine classification. Two classifications were considered as following: state 1 categorized the subjects into three groups (three classes) and the second state put them into two groups (group 1: class 1 and 3, group 2: class 2). The best classification was done using ANN in the second state with the accuracy=94.1% while it was 92.7% in the first state. EEG features carefully predicted the cortisol level. This result is applicable to design the intelligence brain computer machines to control stress and brain performance.

Development of Decision Support System to Predict Neurofeedback Response in ADHD: an Artificial Neural Network Approach.

INTRODUCTION: Clinical decision support system (CDSS) is an analytical tool that converts raw data into useful information to help clinicians make better decisions for patients. AIM: The purpose of this study was to investigate the efficacy of neurofeedback (NF), in Attention Deficit Hyperactivity Disorder (ADHD) by the development of CDSS based on artificial neural network (ANN). METHODS: This study analyzed 122 patients with ADHD who underwent NF in the Parand-Human Potential Empowerment Institute in Tehran. The patients were divided into two groups according to the effects of NF: effective and non-effective groups. The patients' record information was mined by data mining techniques to identify effective features. Based on unsaturated condition of data and imbalanced classes between the patient groups (patients with successful NF response and those without it), the SMOTE technique was applied on dataset. Using MATLAB 2014a, a modular program was designed to test both multiple architectures of neural networks and their performance. Selected architecture of the neural networks was then applied in the procedure. RESULTS: Eleven features from 28 features of the initial dataset were selected as effective features. Using the SMOTE technique, number of the samples rose to around 300 samples. Based on the multiple neural networks architecture testing, a network by 11-20-16-2 neurons was selected (specify>00.91%, sensivity=100%) and applied in the software. CONCLUSION: The ANN used in this study has led to good results in sensivity, specificity, and AUC. The ANN and other intelligent techniques can be used as supportive tools for decision making by healthcare providers.

Wavelet-enhanced convolutional neural network: a new idea in a deep learning paradigm.

PURPOSE: Manual brain tumor segmentation is a challenging task that requires the use of machine learning techniques. One of the machine learning techniques that has been given much attention is the convolutional neural network (CNN). The performance of the CNN can be enhanced by combining other data analysis tools such as wavelet transform. MATERIALS AND METHODS: In this study, one of the famous implementations of CNN, a fully convolutional network (FCN), was used in brain tumor segmentation and its architecture was enhanced by wavelet transform. In this combination, a wavelet transform was used as a complementary and enhancing tool for CNN in brain tumor segmentation. RESULTS: Comparing the performance of basic FCN architecture against the wavelet-enhanced form revealed a remarkable superiority of enhanced architecture in brain tumor segmentation tasks. CONCLUSION: Using mathematical functions and enhancing tools such as wavelet transform and other mathematical functions can improve the performance of CNN in any image processing task such as segmentation and classification.

The design and implementation of the software tracking cervical and lumbar vertebrae in spinal fluoroscopy images.

AIM: Manual analysis of neck kinematics is usually associated with measurement errors and it requires the use of software capabilities. Considering laboratory usage, software has been developed to solve the associated problems. MATERIALS & METHODS: Fluoroscopic images taken from 78 women were used to design and evaluate the performance of the software. The software was implemented using C# language, according to the case-based reasoning technique. RESULTS: The viewpoints of experts suggest accuracy of the software in tracking and calculations, which meets their information requirements. CONCLUSION: Using the software could help physiotherapists to accomplish their work in decreased time and with improved accuracy.

Analysis and Modeling of Threatening Factors of Workforce's Health in Large-Scale Workplaces: Comparison of Four-Fitting Methods to select optimum technique.

INTRODUCTION: Workforce is one of the pillars of development in any country. Therefore, the workforce's health is very important, and analyzing its threatening factors is one of the fundamental steps for health planning. This study was the first part of a comprehensive study aimed at comparing the fitting methods to analyze and model the factors threatening health in occupational injuries. METHODS: In this study, 980 human occupational injuries in 10 Iranian large-scale workplaces within 10 years (2005-2014) were analyzed and modeled based on the four fitting methods: linear regression, regression analysis, generalized linear model, and artificial neural networks (ANN) using IBM SPSS Modeler 14.2. RESULTS: Accident Severity Rate (ASR) of occupational injuries was 557.47 ± 397.87. The results showed that the mean of age and work experience of injured workers were 27.82 ± 5.23 and 4.39 ± 3.65 years, respectively. Analysis of health-threatening factors showed that some factors, including age, quality of provided H&S training, number of workers, hazard identification (HAZID), and periodic risk assessment, and periodic H&S training were important factors that affected ASR. In addition, the results of comparison of the four fitting methods showed that the correlation coefficient of ANN (R = 0.968) and the relative error (R.E) of ANN (R.E = 0.063) were the highest and lowest, respectively, among other fitting methods. CONCLUSION: The findings of the present study indicated that, despite the suitability and effectiveness of all fitting methods in analyzing severity of occupational injuries, ANN is the best fitting method for modeling of the threatening factors of a workforce's health. Furthermore, all fitting methods, especially ANN, should be considered more in analyzing and modeling of occupational injuries and health-threatening factors as well as planning to provide and improve the workforce's health.

Use of Artificial Neural Networks (ANNs) for the Analysis and Modeling of Factors That Affect Occupational Injuries in Large Construction Industries.

INTRODUCTION: Occupational injuries as a workforce's health problem are very important in large-scale workplaces. Analysis and modeling the health-threatening factors are good ways to promote the workforce's health and a fundamental step in developing health programs. The purpose of this study was ANN modeling of the severity of occupational injuries to determine the health-threatening factors and to introduce a model to predict the severity of occupational injuries. METHODS: This analytical chain study was conducted in 10 large construction industries during a 10-year period (2005-2014). Nine hundred sixty occupational injuries were analyzed and modeled based on feature weighting by the rough set theory and artificial neural networks (ANNs). Two analytical software programs, i.e., RSES and MATLAB 2014 were used in the study. RESULTS: The severity of occupational injuries was calculated as 557.47 ± 397.87 days. The findings of both models showed that the injuries' severity as a health problem resulted in various factors, including individual, organizational, health and safety (H&S) training, and risk management factors, which could be considered as causal and predictive factors of accident severity rate (ASR). CONCLUSION: The results indicated that ANNs were a reliable tool that can be used to analyze and model the severity of occupational injuries as one of the important health problems in large-scale workplaces. Additionally, the combination of rough set and ANNs is a good and proper chain approach to modeling the factors that threaten the health of workforces and other H&S problems.