Biyolojik sinyallere dayalı robotik el kontrol sistemleri hareket analizinin doğruluğunu, sağlamlığını ve doğallığını iyileştirmeyi amaçlamaktadır. Makalede parmak hareketlerinin sınıflandırılmasını ve öznitelik değişkenlerini kullanarak hareket tahmini ile robotik el kontrolü konu edinilmektedir. Çoklu DOF parmak kontrolü için makine öğrenme modellerinin karşılaştırması ve hibrit model veri toplama sistemi kullanılarak oluşturulan veri setinin eğitilmesi ile modellerin performans değerlerinin yükseltilmesi sağlanmaktadır. Geliştirilen modeller, belirli ampute birey kullanıcılarının etkili bir şekilde robotik el kullanabilmesi için özel olarak tasarlanmış çoklu DOF kontrolünde kullanılmaktadır. Gelişmiş yöntemlerle makine öğrenme modellerinin tahmin çıktıları kullanılarak servo motorların hareketleri sağlanmaktadır. Doğruluk oranları sırasıyla %99,15 ve %97,47 olan kontrol modelleri için MLP ve SVM makine öğrenmesi algoritmaları kullanılmaktadır. Veri setine haricen uygulanan her bir hareketin vektörel filtrelenmiş hareket verisi, geliştirilen modelerin çalışması ile robotik el hareketleri sağlanmaktadır. Her bir parmaktan metal dişli servo motor ile el bileğine doğru kapanma ve el bileğinden dışa doğru açılma gibi 0-128 derece hareket aralığında parmak hareketleri geliştirilen interpolasyon kontrol yöntemi ile sağlanmaktadır.

Biyomedikal mühendisliği ve robotik teknolojideki gelişmeler, protezden rehabilitasyona kadar çeşitli uygulamalarda geleneksel yöntemlerle kullanılan el-parmak hareketlerinin algılanması yoluyla insan-makine etkileşiminin önemini artırmaktadır. Bu çalışmada, kamera görüntüleri ve sEMG sensör verilerinin eş zamanlı olarak kaydedilmesini sağlanarak denetimli makine öğrenme algoritmalarının yüksek doğrulukta eğitim ve test performansı oluşturabilmesi için gerekli öznitelik veri setinin oluşturulabilmesi ve filtrelenerek doğruluk performansının arttırılması amaçlanmaktadır. Benzer hibrit çalışmaların literatür taramaları, %80,22 ile %99,58 arasında değişen doğruluk oranları bildirmektedir. Bu çalışmada, veri önişleme ve öznitelik çıkarımından sonra, Destek Vektör Makineleri (SVM), k-en yakın Komşuluk (KNN) ve Yapay Sinir Ağlarının (ANN) performansı Matlab’ın sınıflandırma öğreticisi uygulaması kullanılarak değerlendirilmiş ve sırasıyla %92,70, %92,26 ve %90,41 test doğrulama sonuçları elde edilmiştir. Veri ön işleme sırasında oluşturulan verisetine, geliştirilen filtreleme işlemi uygulandığında sırasıyla %99,11, %98,55 ve %97,87 doğrulama performans değerleri elde edilmektedir. Bu makale, her iki yöntemin de güçlü yönlerinden yararlanarak sEMG sensörlerini ve kamera görüntü verilerini birleştiren bütünleşik bir yaklaşım sunmaktadır. Önerilen hibrit yöntem, el-parmak hareketlerini sınıflandırmak için makine öğrenimi algoritmalarının doğruluğunu artırarak, hassas hareket algılama ve kontrolü gerektiren uygulamalar için sağlam bir çözüm sunmaktadır. Bu araştırma, insan-makine etkileşimi teknolojilerinde yeni bir yaklaşımı literatüre kazandırmaktadır.

Memory assessment is a critical component of cognitive research and clinical practice, providing insights into cognitive well-being and performance. While traditional neuropsychological tests remain standard, advancements in virtual reality (VR) technology have offered innovative methods for assessing memory. This rapid review examines 9 studies to explore the use of VR-based memory palace techniques and memory assessments. The findings reveal a significant alignment between VR tasks and memory enhancement, while virtual reality also captures relationships with executive functions and overall cognitive performance. By incorporating various ecological contexts, such as residential or commercial environments, virtual reality enhances the environmental validity of memory assessments. However, challenges such as limited accessibility and variability in both VR hardware and software may hinder broader adoption. These limitations, along with the rapidly advancing nature of VR technology, underscore the need for further research to optimize and expand the role of virtual reality in memory assessment.

Webcam-based physiological Signals Screen in Real-Time for Stress and Interpretable Machine Learning began a non-invasive solution to strain checking owing to a webcam. With the increasing number of stress-related diseases, the method proposed here is a worthwhile one for early intervention and mental health management. The real-time stress level can be successfully measured with the highest degree of accuracy by getting data, including facial expressions, heart rate, and respiration rates, and analyzing it through complicated machine learning models. Research has presented that the classification rate can reach hight rate, making it an ideal technique for discovering stress through deep learning frameworks. These tests involve actions that can be planned ahead of time and become adaptable to the environments, such as virtual consultations through the Internet. Thus, this will boost the mental health in the following areas. However, problems involving privacy violations, ethical issues, and technical hitches are testing grounds that must be solved to ensure responsible installation. Future studies should focus on forming new procedures, raising people’s interest, and finding a way to link technological improvements with moral principles. The rapid and timely identification of stress in different work and healthcare settings and state-of-the-art telemedicine systems guarantee enormous advantages in this area. This paper aims to introduce an interpretable machine learning algorithm using webcam-based physiological signals, which can recognize the existence of stress in real time and be a non-invasive and immediate system to address mental illnesses.

A dataset has been created to support advancements in brain-computer interface (BCI) research, particularly focusing on P300 speller systems and electroencephalography (EEG) signal analysis. This dataset provides detailed EEG recordings from 30 healthy participants during offline analysis, online character recognition, and word-writing tasks. A 16-channel Brain Products V-Amp device was utilized, and data were collected via a 7×7 visual stimulus matrix designed to evoke reliable P300 responses, with stimuli presented in randomized sequences. The dataset comprises raw EEG signals, binary labels, and stimulus timing information, structured to facilitate the development of innovative BCI algorithms and real-time applications. This open-access resource enables novel approaches to EEG signal classification and supports the design of adaptive P300 speller interfaces, offering a foundation for advancing assistive technologies and neuroscience research.

Advances in robotics and biomedical engineering have expanded the possibilities of Human-Computer Interaction (HCI) in the last few years. The identification of hand movements is the accurate and real-time signal acquisition of hand movements through the use of image-based systems and surface electromyography sensors. This study uses multithreading to record motion signals from the forearm muscles in conjunction with a surface electromyography (sEMG) sensor and a camera image. The finger movement information labels were tabulated and analyzed along with the simultaneous acquisition of surface electromyography signals and these gestures through the camera. After the acquisition, signal processing techniques were applied to the sEMG signal markered from the camera. Therefore, once the interface is established, data sets suitable for machine learning can be generated.

The test system was created by adding flex and force sensors to this manikin to evaluate the results of heart massage applied on an artificial adult manikin, which has a spring that can give the chest stiffness of an adult human. This test system monitors and analyzes cardiopulmonary resuscitation (CPR) applications performed by the automated CPR device or manually by the person. The signals received from the sensor are transferred to the computer via the serial port with the Arduino Uno card and displayed in real-time in MATLAB graphical user interface (GUI). This GUI, designed with Matlab 2021a software, analyzes the sensor signals resulting from CPR. It gives a graph of the repeats of compression per minute, the depth of each compression, and the compressions rate variable performed by the user or the automatic CPR device during the CPR application. This created test system can evaluate the accuracy of both the automatic CPR device performed in this study and the manual CPR application. The test system designed in this respect can be used in the training and evaluation of cardiac massage applications, which is included in the first aid courses in secondary education, associate degree, undergraduate, and vocational education courses.

Respiration is a vital process for all living organisms. In the diagnosis and the detection of many health problems, patient's respiration rate, breath inhalation, and breath exhalation conditions are primarily taken into consid-eration by doctors, clinicians, and healthcare staff. In this study, an interactive application is designed to collect audio signals, present visual information about them, create a novel 21253x20 audio signal dataset for the detection of breath inhalation and breath exhalation that can be performed through nose and mouth, and classify audio signals based on machine learning (ML) models as breath inhalation and breath exhalation. Audio signals are received from both volunteers' hearts (method 1) and trachea (method 2). ML models as decision tree (DT), Naive Bayes (NB), support vector machines (SVM), k-nearest neighbor (KNN), gradient boosted trees (GBT), random forest (RF), and artificial neural network model (ANN) are used on the created dataset to classify the received audio signals from nose and mouth into two different conditions. The highest sensitivity, specificity, accuracy, and Matthews correlation coefficient (MCC) for the classification of breath inhalation and breath exhalation are respectively obtained as 91.82%, 87.20%, 89.51%, and 0.79 by method 2 based on majority voting of KNN, RF, and SVM. This paper mainly focuses on usage of audio signals and ML models as a novel approach to classify respiratory conditions based on breath inhalation and breath exhalation via an interactive application. This paper uncovers that audio signals received from method 2 are more effective and eligible to extract information than audio signals received from method 1.

This study uses machine learning to perform the hearing test (audiometry) processes autonomously with EEG signals. Sounds with different amplitudes and wavelengths given to the person tested in standard hearing tests are assigned randomly with the interface designed with MATLAB GUI. The person stated that he heard the random size sounds he listened to with headphones but did not take action if he did not hear them. Simultaneously, EEG (electro-encephalography) signals were followed, and the waves created in the brain by the sounds that the person attended and did not hear were recorded. EEG data generated at the end of the test were pre-processed, and then feature extraction was performed. The heard and unheard information received from the MATLAB interface was combined with the EEG signals, and it was determined which sounds the person heard and which they did not hear. During the waiting period between the sounds given via the interface, no sound was given to the person. Therefore, these times are marked as not heard in EEG signals. In this study, brain signals were measured with Brain Products Vamp 16 EEG device, and then EEG raw data were created using the Brain Vision Recorder program and MATLAB. After the data set was created from the signal data produced by the heard and unheard sounds in the brain, machine learning processes were carried out with the PYTHON programming language. The raw data created with MATLAB was taken with the Python programming language, and after the pre-processing steps were completed, machine learning methods were applied to the classification algorithms. Each raw EEG data has been detected by the Count Vectorizer method. The importance of each EEG signal in all EEG data has been calculated using the TF-IDF (Term Frequency-Inverse Document Frequency) method. The obtained dataset has been classified according to whether people can hear the sound. Naïve Bayes, Light Gradient Strengthening Machine (LGBM), support vector machine (SVM), decision tree, k-NN, logistic regression, and random forest classifier algorithms have been applied in the analysis. The algorithms selected in our study were preferred because they showed superior performance in ML and succeeded in analyzing EEG signals. Selected classification algorithms also have features of being used online. Naïve Bayes, Light Gradient Strengthening Machine (LGBM), support vector machine (SVM), decision tree, k-NN, logistic regression, and random forest classifier algorithms were used. In the analysis of EEG signals, Light Gradient Strengthening Machine (LGBM) was obtained as the best method. It was determined that the most successful algorithm in prediction was the prediction of the LGBM classification algorithm, with a success rate of 84%. This study has revealed that hearing tests can also be performed using brain waves detected by an EEG device. Although a completely independent hearing test can be created, an audiologist or doctor may be needed to evaluate the results.

The adaptive filter is a variation of the digital filtering technique. This form of filter, which does not resemble the classical filtering technique, consists of three basic elements. These elements are collector element, weighting (multiplication) element and a digital filter structure. A system, which has these elements, can make the necessary change in the filter characteristics depending on the environmental media by changing the filter coefficients. Breathing corresponds to the movement of the thorax and lungs and to volume and pressure changes that occur successive in these organs. Respiratory rate (RR) means the respiratory frequency per minute. Since the RR is used to detect and monitor the serious diseases, designing a respiratory rate detection system by means of using adaptive filtering is considered one of the important issues. In this study, a system that detects respiratory rate has been implemented. In this system, there are adaptive filtering, speech boundaries detection algorithm in the sound signal, two stethoscopes whose internal part is placed a microphone and a MATLAB GUI interface design. One of the microphones is placed to upper part of the trachea and the other is placed to upper part of the heart. The sound signals coming from the stethoscope on the heart are used as a noise source and a preprocessing is carried out by means of making free from this noise the sound signals received from the microphone placed on the trachea. After this pre-processing, a clean breathing sound signal is reached by means of making free the heart sounds from the stethoscope placed on the trachea at the adaptive filter output. Inhalation and exhalation time intervals can be determined by running the speech boundaries detection algorithm on this clean breathing sound signal. The respiratory rate is obtained by using these determined time intervals.

Background: P300 spellers are brain-computer interfaces (BCIs) that display desired characters, once at a time, on a screen through the detection of P300 event-related potentials (ERPs) generated in response to flashing visual stimuli using classification methods to determine desired outcomes. Individual words can also be displayed, rather than the letter-by-letter display typical of P300 spellers. Method: An innovative interface using a 7 x 7 letters visual stimulus matrix was designed, as the Easy Screen P300 Speller. In addition to alphabetic characters, 20 shortcut elements (E1-E20) can be used to display words directly on the screen. After first one or more letters of a desired word are determined, 20 words are formed in the word list. If the selection of a shortcut element is detected, the word corresponding to that element is displayed on the screen. Result and discussion: An innovative P300 speller BCI was tested for 19 men and 11 women. Offline, online, and word typing studies were performed using the designed interface. During online analysis, on average, each subject focused on 28.37 of 30 characters. Each subject was asked to display 10 words pre-selected according to the subjects' wishes using the Easy Screen P300 Speller. The same word could be displayed in an average of 1.31 min using the Easy Screen P300 Speller compared with 4.53 min using a conventional P300 speller. This paper uncovers upper results in terms of character detection accuracy and Output Characters per Minute (OCM) value across word-typing interfaces than state-of-the-art.

Amaç: Python programlama dili tabanlı, mini jel elektrofrez sistemi görüntü işleme arayüzü yardımıyla jel elektroforez bant görüntülerinin iyileştirilmesini sağlayan algoritma fonksiyonu oluşturarak deneysel bir çalışmanın yapılması amaçlanmıştır. Gömülü sistem tabanlı mini jel elektroforez sistemi ile bütünleşik olarak kullanılabilen program arayüzü ile, kullanıcıya filtreleme seçenekleri sunulmakta, BP (Base pair-Baz çifti) sayıları, bantların özellikleri, piksel konumları, RF (Relative font) değerleri görüntü üzerine yazdırılabilmektedir. Böylece, klasik yöntem olarak kullanıcının kendisi tarafından cetvel ile yapılan mesafe ölçümlerinde oluşan hatalar, teknolojik sistemler üzerinden oluşturulan algoritmalar ile gerçeğe en yakın değerde ölçülen değerlerin bulunmasıyla en aza indirilebilmektedir. Yöntemler: Bu sistem ile jel görüntülerinin UV (ultraviyole) altında analizi, tasarlanan arayüz yazılımı ile yapılmıştır. Görüntüler kamera aracılığıyla arayüz yazılımına aktarılır, şerit ve BP sayılarının en doğru sonuçla uygulayan görüntü işleme fonksiyonu uygulanmıştır. Yazılım, Raspberry pi 3 B + 'da OpenCV kütüphaneli Python programlama dili kullanılarak mini jel elektroforez sistemine entegre edilerek kullanılabilmektedir. Bulgular: Bu çalışmada, gömülü sistem tarafından kontrol edilen mini jel elektroforez sisteminden elde edilen jel görüntüleri, görüntülerdeki bantların BP sayılarını tahmin etmek için kullanılmıştır. Jel görüntülerini sistemdeki kamera üzerinden veya dosyadan içe aktararak, arayüz yazılım algoritması, el ile yapılan ölçümlerdeki BP sayılarının tahminini ve ortalama hata oranlarını % 30'dan % 0,55 -% 0,8655 aralığına düşürülmesini sağlamıştır. Sonuç: Jel görüntülerinde şeritlerin ve BP sayılarının en düşük hata oranı ile hesaplanabilmesini sağlayan arayüz yazılımı, üstel fonksiyonların kuvvet fonksiyonu olarak da tanımlanabilmesi ilkesinden yola çıkarak iki terimli üstel fonksiyonun iki terimli kuvvet fonksiyonu olarak görüntü işleme algoritmasına uygulanmıştır. Uygulama sonucunda ortalama hata oranının en düşük değerde ve R2=0,9999533 değerinde bulunması ile uygulanan yeni fonksiyonun amacına uygun çalıştığı ortaya konmuştur.

It is aimed to carry out an experimental study by creating an algorithm function that enables the improvement of gel electrophoresis band images with the help of the Python programming language-based, mini gel electrophoresis system image processing interface. With this system, the analysis of gel images under UV (Ultraviolet) light was made with the designed interface software. The images were transferred to the interface software via the camera and filters were applied to highlight the lanes and bands. The two-term power function, which estimates BP (Base Pair) numbers with the most accurate result, was used. The software can be used on Raspberry Pi 3B+ by being integrated into the mini gel electrophoresis system via Python programming language with OpenCV library. In this study, gel images obtained from the mini gel electrophoresis system controlled by the embedded system were used to estimate the BP numbers of the bands in the images. By importing the gel images from the camera or from the file in the system, the interface software algorithm enabled the estimation of BP numbers in the manual measurements and reducing the average error rates to the range of 0.55% - 0.86%. In the interface software, which enables the calculation of lanes and BP numbers in gel images with the lowest error rate, two-term power function has been applied to the image processing algorithm instead of the two-term exponential function, based on the principle that exponential functions can also be defined as power functions. It has been revealed that the BP values calculated with this two-term power function work in accordance with its purpose with the value of R-2=0.9999533, which shows the closeness of the actual BP values.

Brain computer interfaces (BCI) is a tool that can make user requests to computerized systems by directly processing brain signals.In order to perform the procedures to be performed, brain signals must be classified.For this purpose, many classification algorithms have been tried with machine learning.The purpose of this study is to talk about both the type of brain signals used in the brain computer interface and the machine learning techniques used in the classification of these signals.In addition, summary information about the classification methods used in brain computer interface control applications in recent years are given in a table.

In this study, a human-computer interface was created in C# so that individuals with physical mobility disabilities such as ALS can express their wishes. In this system created, pupil movements were analyzed and the patient's wishes were expressed both visually and audibly. In the system created for the tracking of the pupil, the face of the patient, which was detected by the camera, was detected autonomously by the system. An adaptive IR LED light source has been designed to illuminate the eye area of the user. Pupil motion detection was performed with the developed image processing algorithms. According to the movements of the detected pupil, commands were created on the user interface to express the wishes of the patient by using the location information of the patient. An application study was carried out by creating the prototype of the controlled patient bed with a 3D printer. At the end of this study, pupil motion detection was carried out using a camera without any contact with the user. With the algorithm created for pupil motion detection, it is ensured that the patient can express his wishes without the need for any movement other than eye movement. With this study, a uniquely developed algorithm that can be used in pupil tracking systems of individuals with physical movement disabilities such as ALS has been acquired.

Many eyeglass users have to change their lenses due to the continual decline in vision. Sometimes it can be uncomfortable for some people to use multiple eyeglasses of varying degrees to perform different activities. Therefore, researchers in the field of optics are developing solutions to solve this problem. This work aims to develop an adjustable eyeglass which are able to changes refraction degrees of their lenses. This type of eyeglasses helps patients with myopia and similar vision issues. In this study, a medical eyeglass was designed based on controlling the distance between two lenses by injecting a liquid silicone oil for changing the degree of the lenses. The amount of liquid between the lenses is controlled by an android application. Several tests and experiments have been performed to ensure the effectiveness and readiness of the eyeglass. The tests were performed with the help of a lensometer, and acceptable results are obtained as accurate. In the scientific literature, it is common to use a manual adjustment mechanism for the injection of a specified amount of a fluid. In this work, a different injection mechanism that automatically controlled by a mobile phone was used. By this way, the eyeglasses adjustment process became easier for individuals who wear eyeglasses. Personalized values in settings can be stored in the memory of the mobile application to make easy to change between different configurations.