Search

Auscultation is a treatment method frequently used by physicians in terms of giving general information about the body. By listening to the sounds of breathing, the physician can have a general knowledge of the patient's body. In addition, respiratory distress occurs in many diseases such as COPD and asthma. The development of biomedical device technology makes it faster and easier to diagnose and treat. In this study, an electronic equipment has been prepared for the auscultation process. It has 4 channels in the hardware, and two of the channels listen to the heart. Of the remaining 2 channels, one performs listening to the left and the other the right lung. A filter is designed for the heart, which decays between 20-500 Hz. Since breathing sounds can be heard at a wider frequency, a filter has been prepared that passes between 100-1000 Hz.

With the development of technology, cameras are used more widely. It is possible to evaluate the widespread use of cameras in various subjects in daily life. Especially face recognition systems are one of the most important areas of use of cameras. Facial recognition systems can be used in many areas such as cyber security, entertainment, security applications of daily used devices, and faster and easier transactions in financial areas. Although a lot of progress has been made in this regard, face recognition systems are still used widely enough because it is thought that they have weaknesses in terms of security. Many scientists are working on facial recognition. In this study, it is aimed to detect the faces of people determined from videos or live camera images in the best and safest way. Yolov4 object detection algorithm, a ready-made algorithm, was used for the detection of human faces on images. The faces of the people in the images were detected by training the data set we created in the Yolov4 algorithm. An accuracy of 99.1 has been achieved for detecting people's faces on images. The data set we created with pictures of certain people is trained in the CNN algorithm. The faces of the people detected on the images were classified on the model trained with the CNN algorithm for the identification of the people, and the accuracy value was examined for the detection of the identified people on the video recordings or live images from the cameras.

Heart sounds are important data that reflect the state of the heart. It is possible to prevent larger problems that may occur with early diagnosis of abnormalities in heart sounds. Therefore, in this study, the detection of abnormalities in heart sounds has been studied. In order to detect abnormalities in heart sounds, the heartbeat-sounds data set obtained free of charge from the kaggle.com website was examined. Mel frequency cepstral coefficients (MFCCs) were used in the selection of the characteristics of the sounds. Parameters such as the number of filters to be applied for MFCCs, the number of attributes to be extracted are examined separately with different values. The classification performance of heart sounds with feature matrices extracted in different parameters of MFCCs with K-nearest neighbor algorithm was investigated. The classification performance of different feature extractions was compared and the best case was tried to be determined. Two different records that make up the data set were examined separately as normal and abnormal. Then, the new data set obtained by combining the two records was examined as normal and abnormal.

Diagnosis of disease with respiratory data is very important today as it was in the past. These diagnoses, which are mostly based on human experience, have begun to leave their place to machines with the development of technology. Especially with the emergence of the COVID-19 epidemic, studies on the ability of artificial intelligence to diagnose diseases by using respiratory data have increased. Sharing open-source data has paved the way for studies on this subject.Artificial intelligence makes important contributions in many fields. In the field of health, significant accuracy results have been obtained in studies on respiratory sounds. In this article, a literat ure review on respiratory sounds and artificial intelligence achievements was made. 34 articles -that were selected from IEEE, Elsevier, Pubmed, and ScienceDirect digital databases and published after 2010- were used for comparisons. As keywords, "breathing sounds and", "respiratory sound classification", together with "artificial intelligence" and "machine learning" were chosen. In this study, artificial intelligence methods used in 34 publications selected by literature review were compared in terms of the performances obtained in the training.

A platform has been created that allows playing chess with Turkish voice commands. The aim of this study is to enable individuals with limited movement abilities as a result of congenital reasons or a certain disease or accident to play chess and perform a social activity without the help of another person, and to be rehabilitated at the same time. It consists of three parts: Chess module, Human-computer interaction module and Artificial Intelligence module. 29 words have been determined to provide movement in the game on the platform. Voice recordings from 151 people, 86 men and 65 women, were used. Feature selection was made on 43790 voice recordings by using mel frequency cepstral coefficients (MFCC) and gammatone cepstral coefficients (GTCC) methods. The results obtained were classified using the traditional CNN model. The data obtained after using MFCC and GTCC methods were used as inputs in the CNN model. In addition, the data obtained by the two methods were combined and trained in the model. Depending on the methods used in the created model, 83% to 85.9% results were obtained. It was determined that the results obtained using the MFCC method were more successful.

Iron production in the iron and steel industry is a process that starts with the melting of scrap in electric arc furnaces or iron ore in basic oxygen furnaces. The proportions of the alloys in the liquid steel obtained from the liquid steel obtained by melting scrap are of great importance in order to produce the desired quality iron. In steel production, it is necessary to reduce the carbon rate to the desired level, to reduce the proportions of manganese, silicon and other chemicals to the values prescribed in the prescription, and to remove sulfur from liquid steel as much as possible. Therefore, alloys are added (FeSiMnPOTP, AltelPOTP, GrnKrbnPOTP, FeMnOrtCPOTP, KirecPOTP, FeSiPOTP, AlPOTP, FlşptPOTP etc.). Each alloy added has a chemical that acts. For example; If it is desired to change the aluminum ratio of liquid steel, AltelPOTP alloy is added. In the analysis results, it is observed that the aluminum ratios have changed. The liquid steel transferred to the ladle furnace is analyzed at certain intervals and the addition of chemical alloys continues until the required ratios are obtained. Chemical alloys added to liquid steel should not be less or more than they should be, in terms of both material and quality standards. Because the mentioned alloys are serious cost items when purchased in dollars and spread over a long term. For this reason, the rates should be adjusted very accurately. All these metallurgical processes are complex, multivariate systems. Looking at the examinations made, it is seen that while the alloys to be added to the liquid steel in the ladle furnace are rehearsed for an average of 4 times in a casting, this process is repeated at least 2 and at most 6 times. Taking samples from the liquid steel in the ladle furnace, sending the sample for chemical analysis, obtaining the result of chemical analysis and repeating these processes if the desired quality standards are not obtained, the average time is 45 minutes. These periods cause serious waste of time. For this reason, the time of the next casting has to be started later than the planned time. This causes delay in the subsequent processes (pouring liquid steel into molds in continuous casting, forming in the rolling mill, passing through quality tests, etc.). Today, with the advancement of technology, the use of artificial intelligence in the iron and steel industry will be a mandatory approach to minimize the number of proofs and minimize the loss of material and temporal labor.

In academic studies, there are many factors that change depending on the changes in the parameters of the process, such as the processing time, the required processing power, as well as the success. In the methods used for classification, recognition, and detection, the changes in the data received as input may affect the result, as well as the variables specific to the methods used. Convolutional neural networks, whose use is increasing day by day in processes such as classification and recognition using images, learn and classify the characteristics of data sets in different image sizes, including color, gray or black and white images, with filters and functions on the layers in the model. Many different parameters such as layers in the created model and filters and functions in these layers can be changed. As a result of these changes, the most suitable number of layers, the optimum values for the parameters and functions in these layers are determined for the data set used. There are studies focused on optimizing many different structures, such as reproducing the images in the used data set or determining the best by testing different parameters in the classification method. In this study, while the changes were made in the leaf images with a fixed background in the determined leaf data set, the model used in leaf classification with convolutional neural network was kept constant. It is aimed to examine the pictures used for 3 different image sizes, the gray picture or color picture difference and the changes caused by the background color.

Natural language processing is a branch of computer science that combines artificial intelligence with linguistics. It aims to analyze a language element such as writing or speaking with software and convert it into information. Considering that each language has its own grammatical rules and vocabulary diversity, the complexity of the studies in this field is somewhat understandable. For instance, Turkish is a very interesting language in many ways. Examples of this are agglutinative word structure, consonant/vowel harmony, a large number of productive derivational morphemes (practically infinite vocabulary), derivation and syntactic relations, a complex emphasis on vocabulary and phonological rules. In this study, the interesting features of Turkish in terms of natural language processing are mentioned. In addition, summary info about natural language processing techniques, systems and various sources developed for Turkish are given., 8 pages, Natural and Engineering Sciences

Goal: Chronic obstructive pulmonary disease (COPD) is one of the deadliest diseases in the world. Because COPD is an incurable disease and requires considerable time to be diagnosed even by an experienced specialist, it becomes important to provide analysis abnormalities in simple ways. The aim of the study is to compare multiple machine-learning algorithms for the early diagnosis of COPD using multichannel lung sounds. Methods: Deep learning (DL) is an efficient machine-learning algorithm, which comprises unsupervised training to reduce optimization and supervised training by a feature-based distribution of classification parameters. This study focuses on analyzing multichannel lung sounds using statistical features of frequency modulations that are extracted using the Hilbert–Huang transform. Results: Deep-learning algorithm was used in the classification stage of the proposed model to separate the patients with COPD and healthy subjects. The proposed DL model with the Hilbert–Huang transform based statistical features was successful in achieving high classification performance rates of 93.67%, 91%, and 96.33% for accuracy, sensitivity, and specificity, respectively. Conclusion: The proposed computerized analysis of the multichannel lung sounds using DL algorithms provides a standardized assessment with high classification performance. Significance: Our study is a pioneer study that directly focuses on the lung sounds to separate COPD and non-COPD patients. Analyzing 12-channel lung sounds gives the advantages of assessing the entire lung obstructions.

Patient monitoring systems (PMS) is a method used for rating and controlling of the treatments and disorders. While PMS offers as wearable technologies for adults; it is used for adapting to incubator modules to ensure continuous and instant monitoring for newborn babies at the pediatric level. The purpose of this study is to make a pediatric PMS with the ability to be integrated to all type of classical incubators, and monitoring the conditions of the newborn babies with various sensors. It is designed as an Arduino based system that instantly records the variables such as temperature, humidity, air quality to the server using Wi-Fi connection and instantly providing the monitoring ability for clinicians and nurses. Accelerometer is used to control the baby's awake and asleep status, and patientbased safe identification and assessments are made using RFID. As a result, a system has been designed to integrate all the classical incubators and ensure that a large number of incubators can be monitored from a single center.