Regional Signal Recognition of Body Sounds

One of the most important signals in the field of biomedicine is audio signals. Sound signals obtained from the body give us information about the general condition of the body. However, the detection of different sounds when recording audio signals belonging to the body or listening to them by doctors makes it difficult to diagnose the disease from these signals. In addition to isolating these sounds from the external environment, it is also necessary to separate their sounds from different parts of the body during the analysis. Separation of heart, lung and abdominal sounds will facilitate digital analysis, in particular. In this study, a dataset was created from the lungs, heart and abdominal sounds. MFCC (Mel Frekans Cepstrum Coefficient) coefficient data were obtained. The obtained coefficients were trained in the CNN (Convolution Neural Network) model. The purpose of this study is to classify audio signals. With this classification, a control system can be created. In this way, erroneous recordings that may occur when recording physicians' body voices will be prevented. When looking at the results, the educational success is about 98% and the test success is about 85%.

Ovarian Cancer Prediction Using PCA, K-PCA, ICA and Random Forest

Ovarian cancer, which is the most common in women and occurs mostly in the post-menopausal period, develops with the uncontrolled proliferation of the cells in the ovaries and the formation of tumors. Early diagnosis is very difficult and in most cases, it is a type of cancer that is in advanced stages when first diagnosed. While it tends to be treated successfully in the early stages where it is confined to the ovary, it is more difficult to treat in the advanced stages and is often fatal. For this reason, it has been focused on studies that predict whether people have ovarian cancer. In our study, we designed a RF-based ovarian cancer prediction model using a data set consisting of 49 features including blood routine tests, general chemistry tests and tumor marker data of 349 real patients. Since the data set containing too many dimensions will increase the time and resources that need to be spent, we reduced the dimension of the data with PCA, K-PCA and ICA methods and examined its effect on the result and time saving. The best result was obtained with a score of 0.895 F1 by using the new smaller-sized data obtained by the PCA method, in which the dimension was reduced from 49 to 6, in the RF method, and the training of the model took 18.191 seconds. This result was both better as a success and more economical in terms of time spent during model training compared to the prediction made over larger data with 49 features, where no dimension reduction method was used. The study has shown that in predictions made with machine learning models over large-scale medical data, dimension reduction methods will provide advantages in terms of time and resources by improving the prediction results.

Artificial Intelligence Applications in Nutrition and Dietetics

Artificial intelligence (AI) is a branch of computer science whose purpose is to imitate thought processes, learning abilities, and knowledge management. The increasing number of applications in experimental and clinical medicine is striking. An artificial intelligence application in the field of nutrition and dietetics is a fairly new and important field. Different apps related to nutrition are offered to the use of individuals. The importance of individual nutrition has also triggered the increase in artificial intelligence apps. It is thought that different apps such as food preferences and dietary intake can play an important role in health promotion. Researchers may have some difficulties such as remembering the frequency or amount of intake in assessment of dietary intake. Some applications used in the assessment of food consumption contribute to overcoming these difficulties. Besides, these apps facilitate the work of researchers and provide more reliable results than traditional methods. The apps to be used in the field of nutrition and dietetics should be developed by considering the disadvantages. It is thought that artificial intelligence applications will contribute to both the improvement of health and the assessment and monitoring of nutritional status.

Diabetes Prediction Using Machine Learning Techniques

Technologies which are used in the healthcare industry are changing rapidly because the technology is evolving to improve people's lifestyles constantly. For instance, different technological devices are used for the diagnosis and treatment of diseases. It has been revealed that diagnosis of disease can be made by computer systems with developing technology.Machine learning algorithms are frequently used tools because of their high performance in the field of health as well as many field. The aim of this study is to investigate different machine learning classification algorithms that can be used in the diagnosis of diabetes and to make comparative analyzes according to the metrics in the literature. In the study, seven classification algorithms were used in the literature. These algorithms are Logistic Regression, K-Nearest Neighbor, Multilayer Perceptron, Random Forest, Decision Trees, Support Vector Machine and Naive Bayes. Firstly, classification performance of algorithms are compared. These comparisons are based on accuracy, sensitivity, precision, and F1-score. The results obtained showed that support vector machine algorithm had the highest accuracy with 78.65%.

EEG based Spatial Attention Shifts Detection using Time-Frequency features on Empirical Wavelet Transform

The human nervous system has over 100b nerve cells, of which the majority are located in the brain. Electrical alterations, Electroencephalogram (EEG), occur through the interaction of the nerves. EEG is utilized to evaluate event-related potentials, imaginary motor tasks, neurological disorders, spatial attention shifts, and more. In this study, We experimented with 29-channel EEG recordings from 18 healthy individuals. Each recording was decomposed using Empirical Wavelet Transform, a time-frequency domain analysis technique at the feature extraction stage. The statistical features of the modulations were calculated to feed the conventional machine learning algorithms. The proposal model achieved the best spatial attention shifts detection accuracy using the Decision Tree algorithm with a rate of 89.24%.

Manufacturing Multicolor LED-Based Phototherapy Device with a Novel 3D Design

Jaundice is a condition that results from an increase in bilirubin level in the blood. Its prevalence in newborns is around 60-70%. When this temporary jaundice becomes pathological and left untreated, significant damages may occur such as brain damage, vision loss, lung and kidney dysfunction, and even death. One of the methods used for the treatment of jaundice is phototherapy. In this study, a design has been made with 3 foldable LED panels to increase the target area. In addition, high-voltage LEDs with blue-green white wavelengths were used. Thus, it was aimed to minimize the risks of nausea and dizziness caused by intense blue light. An automatic system has been achieved by using temperature and light intensity sensors. The system will warn the user at temperatures and light intensity that are harmful to the baby.

Histopathological Evaluation of Tissue Specific Responses of Subacute Clothianidin Administration in Oncorhynchus mykiss

Clothianidin, one of the latest members of neonicotinoids, is a systemic insecticide of the neonicotinoid group that affects the central nervous system by acting as a nicotinic acetylcholine receptor agonist. Although it is stated that it has no dangerous potential for aquatic organisms, accumulation in water basins is important in terms of environmental toxicity. In this study, the histopathological changes caused by clothianidin applied in subacute application (7 days) form and in environmental doses (3, 15 and 30 µg/L) in the brain, kidney, muscle and gill tissue of juvenile Oncorhynchus mykiss were determined. Parallel to the administration of increasing doses of clothianidin, an increase in the severity of pathological lesions is observed in the brain, muscle, kidney and gill tissue. In particular, it shows that as a result of the accumulation of pesticides in aquatic organisms, lesions may develop as tissue-specific responses, thus leading to tissue dysfunction.

Interpolation-Based Smart Video Stabilization

Video stabilization is the process of eliminating unwanted camera movements and shaking in a recorded video. Recently, learning-based video stabilization methods have become very popular. Supervised learning-based approaches need labeled data. For the video stabilization problem, recording both stable and unstable versions of the same video is quite troublesome and requires special hardware. In order to overcome this situation, learning-based interpolation methods that do not need such data have been proposed. In this paper, we review recent learning-based interpolation methods for video stabilization and discuss the shortcomings and potential improvements of them.

Forecasting Model to Predict the Spreading of the COVID-19 Outbreak in Turkey

This study aimed to forecast the future of the COVID-19 outbreak parameters such as spreading, case fatality, and case recovery values based on the publicly available epidemiological data for Turkey. We first performed different forecasting methods including Facebook's Prophet, ARIMA and Decision Tree. Based on the metrics of MAPE and MAE, Facebook's Prophet has the most effective forecasting model. Then, using Facebook's Prophet, we generated a forecast model for the evolution of the outbreak in Turkey fifteen-days-ahead. Based on the reported confirmed cases, the simulations suggest that the total number of infected people could reach 4328083 (with lower and upper bounds of 3854261 and 4888611, respectively) by April 23, 2021. Simulation forecast shows that death toll could reach 35656 with lower and upper bounds of 34806 and 36246, respectively. Besides, our findings suggest that although more than 86.38% growth in recovered cases might be possible, the future active cases will also significantly increase compared to the current active cases. This time series analysis indicates an increase trend of the COVID-19 outbreak in Turkey in the near future. Altogether, the present study highlights the importance of an efficient data-driven forecast model analysis for the simulation of the pandemic transmission and hence for further implementation of essential interventions for COVID-19 outbreak.

Evaluation of Wigner-Ville Distribution Features to Estimate Steady-State Visual Evoked Potentials' Stimulation Frequency

Brain Computer Interface (BCI) is a system that enables people to communicate with the outside world and control various electronic devices by interpreting only brain activity (motor movement imagination, emotional state, any focused visual or auditory stimulus, etc.). The visual stimulation based recording is one of the most popular methods among various electroencephalography (EEG) recording methods. Steady-state visual-evoked potentials (SSVEPs) where visual objects are blinking at a fixed frequency play an important role due to their high signal-to-noise ratio and higher information transfer rate in BCI applications. However, the design of multiple (more than 3) commands systems in SSVEPs based BCI systems is limited. The different approaches are recommended to overcome these problems. In this study, an approach based on machine learning is proposed to determine stimulating frequency in SSVEP signals. The data set (AVI SSVEP Dataset) is obtained through open access from the internet for simulations. The dataset includes EEG signals that was recorded when subjects looked at a flickering frequency at seven different frequencies (6-6.5-7-7.5-8.2-9.3-10Hz). In the machine learning-based approach Wigner-Ville Distribution (WVD) is used and features are extracted using Time-Frequency (TF) representations of EEG signals. These features are classified by Decision Tree, Linear Discriminant Analysis (LDA), k-Nearest Neighbor (k-NN), Support Vector Machine (SVM), Naive Bayes, Ensemble Learning classifiers. Simulation results demonstrate that the proposed approach achieved promising accuracy rates for 7 command SSVEP systems. As a consequence, the maximum accuracy is achieved in the Ensemble Learning classifier with 47.60%.