Proceedings of the 4th International Conference on Electronics, Biomedical Engineering, and Health Informatics

Medical data information plays an important role in supporting the confidentiality of diagnosis data, treatment, patient health history, examination results, and laboratory examination results to assist doctors in making accurate diagnoses and choosing appropriate treatment; now, this data has been integrated into the internet. The information system in this application is needed to be more efficient and effective in supporting medical data security. To integrate a confidential medical data privacy system, the information system is equipped with an encryption security system on the NIK, Diagnosis, Amnesia, and Therapy tables to guarantee the privacy of the data sent. The prototype multi-socket network-based communication system, PHP with the Laravel Framework, uses the Advanced Encryption Standard (AES) encryption algorithm using the same Key and Initialization Vector (IV) for encryption and decryption. It was designed to support the medical data security system using the Laravel NET website programming language framework and MySQL database. The system has been built based on two access rights, doctor, and super admin, where doctors can enter patient diagnoses. AES encryption will be applied to all data when downloaded by a doctor. The super admin can input access rights as a doctor but cannot see the patient because it has been encrypted with the AES algorithm through medical records diagnosis. Because when the data is sent, this system is tested using black box testing with 34 black box tests to ensure the confidentiality of confidential patient data. The system with the AES 256 algorithm can be useful for patient data security systems in hospitals.

Virtual Reality (VR) and the Kinect sensor, as promising tools in biomedical research, offers diverse applications in medical education, rehabilitation, diagnostics, and health research. The problem statement highlights the demand for innovative solutions and introduces VR and Kinect as potential transformative technologies. This review analyzes the importance of these technologies, their contributions, and future potential. It stands out by evaluating various Kinect-based systems in medical settings. By highlighting distinct features, advancements, and limitations, it provides guidance for future research. Relevant literature was gathered from databases such as Google Scholar, IEEE Xplore, and PubMed. The results showcase a wide range of applications, including patient autonomy, stroke rehabilitation, diagnostics, and monitoring. Despite challenges in accurate movement tracking, integration into clinical settings, and limited generalizability of findings due to small sample sizes, VR and Kinect show potential for revolutionizing healthcare delivery and improving patient outcomes. Their adaptability, affordability, and immersive nature of these technologies offer promising avenues for personalized interventions, remote healthcare, training, and enhanced patient engagement. As these technologies evolve, continued research and development are crucial to optimize their impact in shaping the future of healthcare.

Since practically all activities were impeded by the COVID-19 epidemic, they were conducted as autonomously as possible at home, or what is known as Work from Home (WFH). Taekwondo activities are among those that cannot be performed as WFH. The COVID-19 epidemic disrupted regular Taekwondo training, necessitating autonomous practice at home. However, without a trainer's presence, technical errors in Taekwondo kicks could occur. The research presents an automated system utilizing an IMU sensor and SVM for Taekwondo kick classification, empowering athletes to improve their movements independently. Type of kicks that can be classified are Eolgol Ap Chagi, Momtong Ap Chagi, Eolgol Dollyo Chagi, Momtong Dollyo Chagi, and Dwi Chagi. Because it is a basic kick that must be mastered by Taekwondo Athletes. When using tools, taekwondo athletes can move more quickly, thanks to direct implementation on compact devices. Consequently, a simple machine-learning model with the fewest input characteristics is required. On the Arduino Nano 33 BLE, the LSM9DS IMU sensor was used to collect the data. The dataset goes through a cleansing procedure before being labelled and trained, which comes before pre-processing. Three options that may have been employed are SVM, RBF, and DT. In this study, the micromlgen library will be used. Consequently, this work employs a Support Vector Machine (SVM) methodology. Mean, median, max, min, and variance are the five features used in the pre-processing technique. The median and variance properties are used to get an accuracy of 99.35%. The experimental findings demonstrate that the SVM algorithm successfully categorizes the different kinds of taekwondo kicks. The developed technology serves as a valuable tool for Taekwondo athletes, providing a means to enhance their skills through self-guided practice during situations like the COVID-19 pandemic.

EEG is a powerful and popular technique for measuring brain activity, which reflects the condition of a person’s brain. A person’s brain health can be determined by monitoring brain activity used EEG technique. It has been demonstrated that EEG signals can be used as a diagnostic tool in evaluating individuals with alcoholism. Using the proper diagnostic method for EEG signals, the individual evaluating under alcoholism has been demonstrated. EEG signals record the brain’s electrical activity, measured from the scalp. The measurements obtained from the EEG are used to confirm or rule out conditions such as alcoholism. Alcohol consumption is associated with specific patterns of brain electrical activity in adults, and the brain activity of individuals with alcoholism differs from non-alcoholics in several ways. This study proposes a feature extraction method for multichannel EEG signals using texture analysis for alcoholic and non-alcoholic classification. Multichannel EEG signal is treated as an image and processed using the texture analysis method. Then it is classified using an artificial neural network. The highest accuracy is achieved using the GLDM feature extraction method at a distance of 3 and an angle of 0°, resulting in an accuracy rate of 93.9%. The method used proved to be higher than previous studies using similar methods. The proposed method is expected to be used for other multichannel biomedical signal processing.

There is currently no dedicated network for IoT needs on the Siliwangi University campus, IoT device data communication still relies on the wifi network for internet network needs, which means that the quality of data transmission will be compromised if the network is overloaded with internet users. Siliwangi University needs a suitable, practical, and good network so that it can be used as a data communication line by the installed IoT devices. The viability of setting up an interconnection network on the Siliwangi University campus needs to be examined. In this study, the effectiveness of the CC1101 network used by Siliwangi University to monitor air quality via the Internet of Things was evaluated. The system consists of two components: nodes and gateways. The nodes use Arduino Nano microcontroller boards to transmit air quality data to the gateway, while the gateway uses a nodeMCU ESP8266 microcontroller board to transmit the data to the internet. According to the test findings for the CC1101 using the 433 MHz frequency, the most data that can be transferred is 64 bytes, and it can transmit data up to a distance of 68 m under line-of-sight conditions and 40 m under non-line-of-sight conditions. The data collected shows that there is an average data transmission delay of 859.698 ms and a packet loss of 3.12%.

Solar irradiance is paramount when assessing solar panels’ efficiency and overall effectiveness. The intensity and consistency of solar irradiance directly impact the performance and output of solar panels. The study aims to address the problem of understanding the impact of solar irradiance on the efficiency and performance of solar-powered cars. Specifically, the relationship between solar irradiance levels and the voltage output of solar panels installed on a car, focusing on the EPIC Solar Farm pathway in Kemaman, Terengganu. This research involved modeling solar irradiance and its impact on the performance of a solar-powered car. By analyzing the data measured solar irradiance, the research could correlate it with the voltage output of the solar panels installed on the car. The findings revealed a clear relationship between solar irradiance levels and the performance of the solar-powered car. The study highlighted the significance of accurate solar irradiance predictions for optimizing the performance of solar-powered cars. Car manufacturers and designers can enhance solar-powered vehicles’ overall efficiency and range by considering solar irradiance levels at specific locations, such as the EPIC Solar Farm pathway. These results contribute valuable insights into the field of solar-powered transportation, showcasing the importance of solar irradiance modeling in developing more effective and sustainable solar-powered vehicles.

The transportation sector’s heavy reliance on fossil fuels significantly contributes to air pollution and climate change. There is a growing interest in using solar-powered cars to address these environmental challenges and promote sustainable transportation. However, there is a need to comprehensively explore the performance of solar-powered cars to understand their efficiency and effectiveness. The primary objective of this research is to comprehensively explore the performance of solar-powered cars along the EPIC Solar Farm pathway in Kemaman, Terengganu. The development comprised an array of sensors, communication modules, and onboard processing systems that were integrated into the cars to facilitate the real-time collection, interpretation, and transmission of crucial data. This dataset encompasses vital information relating to battery consumption, environmental factors, and navigation elements. The findings of this study underscore a distinct relationship between solar energy and battery energy. Specifically, as the conversion of solar radiation into solar energy intensifies, the stored energy within the battery increases proportionally. Moreover, the study places significant emphasis on the influence of battery usage on overall performance, particularly during operation along the EPIC Solar Farm pathway. Notably, the study investigates the impact of charging and discharging capacity and battery usage to understand the buggy’s efficiency and effectiveness comprehensively. In summary, the results of this study highlight the substantial benefits associated with the adoption of solar energy in transportation, particularly within the context of solar farm environments. The findings underscore the positive implications for air quality improvement, climate change mitigation, and promoting sustainable practices within the transportation sector.

Induction motors are widely used to convert electrical power into mechanical power because they are simple, rugged, robust, efficient, and suitable for applications in harsh environments. Previous studies have been conducted on magnetic energy harvesting from manufacturing machinery, such as an induction motor, the electromagnetic transducer developed employed the clamped current-transformer technique. Until now, no study has been conducted on mathematical modelling for harvesting magnetic energy using a Clampless Current-Transformer method, specifically from an induction motor. This article proposes the mathematical modelling of electromagnetic transduction from an induction motor using Ansys Engineering Simulation Software. It models and simulates the core design with each material before using Clampless Current-Transformer method electromagnetic transducer to capture the radiation of magnetic field energy in various core designs. The simulation contrasts Stainless and Carbon Steel for the core material consisting of different core shapes; cylindrical and square rods with 300 and 500 turns. The proposed electromagnetic transduction provides better flexibility due to its clampless feature from the incoming supply to the induction motor; instead, it can be placed anywhere near the most substantial magnetic field radiation. The modelling demonstrates that long cylindrical core Carbon Steel produces higher output power for 500 turns. The proposed modelling with the Clampless Current-Transformer approach can efficiently harvest magnetic field energy. It could have practical implications for energy harvesting from induction motors, enabling the transducer to be conveniently near the strongest magnetic field without clamping to the motor supply.

Cataract is an ophthalmic condition where the lens of the eye becomes cloudy, leading to symptoms such as blurry vision, increased sensitivity to light, nearsightedness, and blindness. The World Health Organization (WHO) states that 40% of blindness cases in the world are caused by cataracts. The rate of blindness caused by cataracts in developed countries is 5% while for countries and/or remote areas it is as high as 50%. This leads to a decrease in the productivity level of the society. Therefore, early detection plays an important role in this regard as it can help patients recognize the cataract at an early stage and take action according to the level of cataract experienced. The purpose of this study is to create an automated model to detect and classify cataracts into four conditions, normal, immature, mature, and hypermature using machine learning algorithms. In order to obtain an optimal model to detecting and classifying cataract, this study compares the performance of two machine learning algorithms, Convolutional Neural Network with the proposed layer and K-Nearest Neighbor. In this study, the dataset used was 2000 fundus images which is divided into 1600 training data images and 400 test data images taken from Cicendo Hospital, Garut, West Java, Indonesia. In the previous study, the cataract classification and detection system were carried out using a 3-Layer Convolutional Neural Network by classifying two conditions, normal and cataract with an accuracy value of 95%. Model performance is showed by confusion matrix analysis which includes accuracy, precision, recall, and F1-Score. The best performance is obtained when using 5-Layer Convolutional Neural Network with 98% system accuracy with hyperparameter 100 of epoch, 64 of batch size, and 0.001 of learning rate. Meanwhile, the system accuracy obtained by the K-NN method is 97% with Euclidean Distance, k = 3 and 0° of angle orientation. In this study it can be concluded that the classification system and cataract detection through fundus images obtain good results with the Convolutional Neural Network algortima with an accuracy value of 98%.

Respiratory rate is one of the essential components in measuring vital signs that convey information about health conditions and diagnoses to patients measured in breaths per minute (BrPM). Breaths per minute refer to the number of breaths taken by a person in one minute. The current clinical respiratory rate calculation method, i.e., a standard calculation by counting the number of chest movements, can cause patient discomfort, bias in counting, and waste the nurse’s time. This study aims to develop a tool for measuring respiratory rate by recording a one-minute photoplethysmography (PPG) signal non-invasively that is more comfortable and easier to use. The PPG signal obtained is first processed to remove unnecessary signal frequencies. The developed algorithm counts the number of peaks in the PPG signal to calculate the respiratory rate. It employs Zero-Phase Filtering to remove noise and motion artifacts (MA) on the signal to calculate the number of peaks in one minute. The device is designed using the MAX30102 sensor, XIAO ESP32-C3 controller, and TFT OLED display to show the measurement value. This tool uses a 3D case and Velcro tape on the sensor to reduce the possibility of a Motion Artifact appearing in the signal. The respiratory rate prediction test results obtained more than 94% accuracy with a maximum error of 3 BrPM in five subjects aged 15–48 years who were in good health. Based on the results, measuring respiratory rate from the PPG signal with a wearable device design can be a quite effective non-invasive solution and contribute to the medical world for measuring vital signs.

Obstructive Sleep Apnea (OSA), or cessation of breathing during sleep, is caused by a blocked upper airway. OSA is one of the causes of sudden death caused by heart failure during sleep. Given the impact of OSA, early detection is therefore necessary. This article presents an innovation in OSA detection using first-order statistical features of an electrocardiogram (ECG) and machine learning. The first-order statistical features were obtained from the RR interval of the ECG single-lead. Various statistical features were evaluated for OSA detection. Moreover, machine learning techniques such as Linear Discriminant Analysis (LDA), Artificial Neural Networks (ANN), K-Nearest Neighbors (K-NN), and Support Vector Machines (SVM) were investigated for OSA detection. The accuracy, sensitivity, and accuracy values of each classification are the performance outcomes of machine learning. The proposed OSA detection method was validated using clinical data from patients diagnosed with sleep apnea. When using KNN for OSA detection, the results showed an accuracy of 83.91%, sensitivity of 92.31%, and specificity of 67.19%. Using SVM, the OSA detection demonstrated an accuracy of 83.75%, sensitivity of 92.97%, and specificity of 65.80%. The ANN method for OSA detection yielded an accuracy of 80.83%, sensitivity of 92.67%, and specificity of 59.25%. Lastly, OSA detection using LDA exhibited an accuracy of 79.27%, sensitivity of 82.27%, and specificity of 73.29%. From the results, it was concluded that the KNN classification, with all its features, provided the best performance for OSA detection, achieving the highest accuracy of 83.91%.

Humans play more of a role as operators who carry out control functions, so cognitive abilities, especially those related to perception and decision-making, become very important. Cognitive performance can be seen in a person’s ability to complete a cognitive activity. Electroencephalography is a tool for measuring cognitive performance through human brain activity wave signals. The main objective of this article is to systematically review Power Spectrum Density (PSD) feature extraction methods for cognitive performance measurement based on EEG. The methods used in this study are the PRISMA (Preferred Reporting Items for Systematic Review and Meta-Analysis) method and Bibliometric analysis using VOSViewer. Data sources totaled 50 articles obtained from Scopus, Science Direct, and IOP Science for the 2013–2023 periods. The results of this article were obtained by observing keywords, density, article trends, feature extraction methods, and the application of EEG. The results showed that out of 50 articles that had been reviewed, 19 used PSD to measure cognitive performance, the Alpha frequency band is the most commonly used in measuring cognitive performance. The increasing use of PSD methods for the measurement of cognitive aspects (fatigue, workload, performance, mental) shows that future research directions can still be developed.

This research addresses the problem of improving image classification accuracy, given the importance of classification accuracy in applications such as disease diagnosis and object recognition. This research aims to explore various deep-learning architectures and ensemble methods to improve classification accuracy effectively. We use a comprehensive approach for methodology, evaluating training data from various architectures using 17 transfer learning models. Our method incorporates models such as EfficientNetB0, AlexNet, and MobileNetV2, utilizing their unique strengths. We also designed a two-stage model that progressively combines architectures between low-accuracy and two high-accuracy models to create a more accurate classifier. The dataset contains 5932 images of rice leaf diseases distributed into four classes: Bacterial Leaf Blight, Blast, Brown Spot, and Tungro. The outstanding results showed a substantial increase in accuracy from 0.35 to 0.97 in the ensemble model. This significant improvement underscores the potential of integrating different architectures to utilize complementary features, ultimately improving classification accuracy. This research provides insights into image classification and offers practical solutions to improve accuracy in various domains. As for the implications, this study shows the promise of a blended approach to deep learning architecture in significantly improving image classification performance in various domains.

Addressing the challenge of unbalanced data sets in convolutional neural network (CNN) models for image recognition, this study aims to investigate the impact of data augmentation and normalization. The problem lies in the limited generalization of the model due to class-based data differences, hence the need for several techniques in data preprocessing such as data augmentation and normalization. The main contribution of this research is a comprehensive analysis of the effectiveness of data augmentation and normalization techniques in improving model performance. The research utilized the AlexNet and VGG16 architectures and conducted extensive experiments on data sets with varying degrees of imbalance. Data augmentation generates additional examples, while normalization reduces convergence issues. The results show that training the AlexNet model without augmentation results in a low accuracy of 0.24, underscoring the challenges posed by skewed data distributions. In contrast, augmented data substantially improves performance, with AlexNet achieving an accuracy of 0.91 and VGG16 achieving 0.84. In addition, normalized data also made a positive contribution, showing an accuracy of 0.74 for AlexNet and 0.67 for VGG16. In conclusion, data augmentation and normalization techniques are essential in reducing the effects of data imbalance, thereby improving the generalizability of the models. The improved accuracy of the data using normalization techniques indicates the ability of the model to read the data after normalization. This study underscores the importance of preprocessing strategies in optimizing model performance and advancing the field of deep learning in image recognition.

The instantaneous frequency measurement is the primary focus of different variably-dimensions signal processing applications. It addresses the non-stationarity of signals globally and the spread of signal frequency locally across time. The present work evaluates the effect of a mindfulness intervention on cognitive workload through different frequency step sizes utilizing one dimensional Gabor function and discrete wavelet transform (DWT) on Electroencephalogram (EEG) signals. Time–frequency tiling is carried out for four different cognitive workload levels of a visual task which shows variation in frequencies along the timeline. The statistical analysis utilizes the mean of small segments (100 ms) from long-duration (7 s) EEG data. The significant difference is found in post-meditation data than pre-meditation data for 90% subjects (18 Subjects out of 20) using a paired t-test with p-value < 0.001 for frontal and occipital electrodes from left and right hemisphere. The right hemisphere shows higher modulation of α activity than the left hemisphere. The results using the Gabor function show good performance by meeting the generalized uncertainty principle. The results with DWT show ‘what is where’ on time as well as frequency scales clearly with orthogonal basis. Moreover, the cognitive load of four levels is discriminable for an individual with pre-meditation data of twenty subjects, which is clarified with behavioral measures and using statistical analysis on objective measures (p-value < 0.05). The empirical evidence based on EEG signal analysis clarifies that the proposed method is used as an initialization step to quantify cognitive workload based on its emergent frequency at a given time scale for developmental research. This initial investigation holds relevance in determining the optimal duration for implementing meditative programs among students.

SpO2 monitoring needs to be done routinely and can be done anywhere to prevent delays in handling. However, during the COVID-19 pandemic, sterile and limited patient handling is needed to prevent the spread of this disease, so monitoring of patients cannot be carried out optimally and can cause delays in treatment. One of the efforts that can be made to improve service to patients is by monitoring SpO2 which is carried out based on IoT, one of which is through an HTML Web Page. The purpose of this study is to analyze SpO2 data sent and received by IoT media. The contribution of this research is knowing the response of SpO2 data sent and received through IoT media for the development of a SpO2 monitoring system. The procedure for achieving this goal is the MAX30100 sensor whose output will be processed and displayed on an HTML Web Page. This study obtained an average delay time of 219.3 s. In this study, it was concluded that the transmission of SpO2 data on HTML Web Pages was said to be good. This research is expected to be useful for the development of a SpO2 monitoring system so that it can improve services for patients, especially those who need sterile treatment.

Cardiovascular disease one of the leading global causes of death, this need to be monitored for heart conditions which must done regularly. Electrocardiograph (ECG) method currently being developed facilitate patient monitoring, and it falls under the Internet of Things as we know IoT advancements for ECG devices. This study for analyze ECG signal and BPM also called Beats per Minutes values sent and received through IoT media to aid in the diagnostic process. The study’s contribution lies in understanding the format of ECG signals and BPM values transmitted and received via IoT media. The AD8232 sensor is used in the process, and its output is processed and displayed on an HTML web page. The research shows an average data loss of 0.3652%. Based on this study, the transmission of ECG signal data and BPM values on HTML web pages is deemed effective. The findings of this study are expected to be utilized on further research and may be beneficial for developing an ECG monitoring system to improve patient care, especially for those requiring sterile treatment.

Non-destructive testing (NDT) comprises various analysis techniques used in science and industry to evaluate material properties without causing harm. NDT finds applications across diverse industries, constantly evolving with new methods. It’s pivotal for assessing remaining wall thickness in objects prone to corrosion or erosion, like vessel and piping. This study supporting the relativity of elevated temperature and thickness measurement in industry. While most measurements occur up to 60 °C with standard procedures, exceptions exist. Industries like refining and chemicals assess component thickness at 60 to 550 °C, where cooling isn’t feasible. Temperature accelerates corrosion, demanding more frequent measurements. This study focuses on material thickness and hardness before and after heating. Stainless steel, alloy, and brass are compared, heating from 100 to 500 °C. Results reveal alloy steel thickness increases by 14.1%, followed by brass at 13.8% and stainless steel at 5.65%. In hardness, stainless steel rises by 23.8%, alloy by 23.58%, and brass by 17.12%. Thermal expansion, tied to bond energy and melting points, influences post-heating changes. Material thickness shifts due to expansion/contraction, while hardness alters due to microstructural shifts. Account for these changes in precise assessments. This effect was taken into account as a parameter relating with performance of thickness measurement.

Supervising an infant while managing household tasks poses a significant challenge for parents, often leading to safety concerns due to limited time and attention for constant monitoring. Conventional sensor-based monitoring systems have limitations in accurately detecting an infant’s condition and position. This research proposes a deep learning algorithm method as an alternative to sensors. The deep learning algorithm will be integrated with a camera to assess the infant’s position and condition using the Convolutional Neural Network (CNN) method. To enhance performance, a development board integrated with a Video Graphic Array (VGA) is utilized to accelerate the processing time of the deep learning method, and the CNN model is simplified using the MobileNetv2 architecture to reduce model weight. Testing with various scenarios, including an infant in potentially hazardous positions (i.e., prone, covered by an obstacle, or standing near a fence) and an infant in safe positions, achieved an accuracy of 94%. This indicates that the model can effectively identify the infant’s position and determine their well-being. Additionally, the Frame Per Second (FPS) results of 20 demonstrate real-time infant supervision capabilities, allowing for timely intervention in case of any potential risks. Based on the experimental findings, this method holds promise for applications in the medical field and for parents who may struggle with continuous infant supervision, enabling them to fulfill other responsibilities without compromising infant safety.

Recently, there has been a trend in the field of surgery towards the development of smart medical devices. Sensors of various functionalities are embedded in medical-surgical devices, which makes it possible to read data during surgical interventions and, based on the data obtained, track critical situations and analyze the actions of the surgeon. One of the recent trends is the introduction of multicomponent force-torque sensors in surgical instruments. However, when processing data from such sensors, a problem arises with the visualization of data in a form convenient for human perception. The purpose of this article is to describe an approach to presenting data in the form of violin diagrams for analyzing the data of forces and moments occurring on medical instruments with built-in multi-component force sensors. The main criteria necessary for obtaining the maximum information content of visualization are highlighted, based on which the synthesis of the visualization method is carried out. The proposed method is also tested for various surgical effects on patient tissues. In particular, the results of testing the visualization method are described by experimenting using a multifunctional device for measuring forces and moments on back phantoms, followed by data visualization. The possibility of using the data visualization method for various methods of surgical interventions, as well as the possibility of comparing numerical parameters, is evaluated.

After the Covid-19 pandemic hit Indonesia. Many people are starting to be aware of their own health. However, they do not understand about correct and safe workout movements. Therefore, we want to classify workout movements with a machine learning model using the Convolutional Neural Network algorithm and compare multiple hyperparameter combinations to achieve optimal outcomes, allowing developers to deploy it on websites and applications. By evaluating all hyperparameters, this research aims to optimize model performance in the domain of workout movement classification. In this study there are 22 classes containing workout movements that focus on weightlifting exercises. The dataset was obtained from a downloaded YouTube video that the author subsequently divided into multiple frames. The dataset was splitted into 90% training and 10% validation. To achieve optimal results, the authors examine the impact of batch size, learning rate, and optimizer on the accuracy parameter of the model. The dataset will be trained using InceptionV3 CNN model architecture. The result shows adaptive optimizers like Adam, RMSprop, and AdamW outperformed basic optimizers like SGD and AdaGrad, suggesting that dynamically adjusting learning rates based on gradient information is more effective. A learning rate of 0.001 paired with a high batch size yielded superior results compared to other hyperparameter combinations when training the InceptionV3 model on this dataset. We hope application developers can utilize the study’s findings to improve the performance of fitness apps that involve workout movement classification.

The urgency of handling Covid-19 is very necessary because of its rapid spread. Self-isolation (Isoman) is a way of handling asymptomatic to mild patients. Hence, the need for mobile application technology arises to facilitate remote communication between patients and doctors. This research focuses on the development of a mobile application with a patient module that enables COVID-19 patients to effectively manage their self-isolation period. In this application, features are provided for self-reporting conditions, consultations with doctors, self-isolation monitoring and its duration, Community Health Center (Puskesmas) Hotline, tips, and tricks for undergoing self-isolation. Hence, the contribution of this study was to create a user-centric tool that empowers patients to adhere to self-isolation guidelines and addresses the communication challenges posed by physical distancing. The design of this application, utilizing the Agile Design Science Research Method (ADSRM), encompasses design and development, demonstration, evaluation, and communication. Evaluation of the application at the healthcare center is conducted using a User Experience Questionnaire (UEQ) and User Acceptance to be validated by the healthcare center. This apps used for the patient in Sawah Besar Community Health Center. The results of this development state that the health center approves of the application to help make it easier for Sawah Besar health center patients to undergo Isoman.

The exponential growth of hospital information systems (HIS) has led to the accumulation of vast amounts of medical data, necessitating effective analysis methods to enhance the quality and efficiency of medical services. Machine learning has emerged as a valuable technology for the automated and accurate analysis of medical data, offering potential applications in disease diagnosis and treatment. This study aims to contribute to the advancement of classification methods and address data imbalance issues in the context of hematological data. Specifically, we propose an efficient algorithm for disease classification utilizing hemogram blood test samples, employing the random forest algorithm in conjunction with the synthetic minority oversampling technique. Experimental results using real hematological data from a local hospital demonstrate the superiority of the proposed method, achieving an impressive accuracy rate of up to 97.75% and an Area Under the Curve value of up to 98.65%. The findings underscore the value of leveraging machine learning techniques in diagnoses and treatment in clinical practice, especially when integrated into HIS systems.

Data mining is a process of analyzing a set of existing data in the database so that information is obtained that is used in the next stage. Classification is a technique by forming models from unclassified data, to be used to classify new data. Classification is included in the type of supervised learning, meaning that training data is needed to build a classification model. Each classification has a choice of algorithms, which are often used are the naive Bayes algorithm, k-nearest neighbor, decision tree and support vector machine. In this study, a comparison was made to a case study of decision making in skin cases. The research method used in this study is a training method using data training and testing mode using data testing with a comparison of training and testing 70%: 30%, 80%: 20%, 90%: 10% by naive bayes algorthm, K-nearest neighbor, decision tree. Evaluate the training and testing carried out on the naive Bayes classification algorithm, nearest neighbor, decision tree in this study, namely how much accuracy the algorithm produces. The results of the comparison of the nearest neighbor, naive Bayes and decision tree classification algorithms used in the case study of clothing pattern selection decision making state that the decision tree classification algorithm is the classification algorithm that has the highest level of accuracy, reaching 97.5% in testing.

Identifying an athlete’s speed and quickness talent is a crucial stage in the world of sports that can determine an athlete’s success. However, in general, the athlete’s speed and quickness measurements are still done manually with a stopwatch, and then the results are written down in a book or paper. This study proposes a Quickness Aspect Talent Identification (QATI) system that uses Internet of Things (IoT) technology to identify athlete talent for quickness. This system is designed to obtain athlete quickness data by monitoring their movements through sensors placed at predetermined points and connected to a database via the Internet. The QATI system can help identify athletes that traditional talent identification methods might miss. The QATI system comprises four nodes, one controller, and a website-based information system. The node serves as an athlete’s goal point, which can be placed at four points. The controller will set the on–off combination on the node with an adjustable duration or run with a random algorithm. NodeMCU sends time measurement results to a Wifi or Bluetooth network server. The measurement results from each sensor on the node will appear on the website that has been created. The QATI application is proven to perform several configurations such as training mode, delay between devices, duration of on, the distance the device detects, and viewing indicators of the number of connected sensors. The QATI system is an innovative approach to talent identification that utilizes IoT technology to identify athletes for quickness.

The development of mobile technology has penetrated into the health sector. With the development of this technology, information about health can improve the quality of health independently. The integration of information technology into the healthcare sector, namely the advancement of mobile-based health services, has significantly revolutionised the accessibility of healthcare across various regions in Indonesia. Halodoc is the leading digital health company in Indonesia and has substantially changed the axis of health services in Indonesia by providing health information that is easy to understand, credible, and accessible to everyone. This research will analyze Halodoc’s service quality based on customer reviews from Google Play Store, which will be analyzed using sentiment analysis. Specifically, this research uses VADER (Valence Aware Dictionary and Sentiment Reasoner) and Roberta Pretrained Model to analyze the sentiment of user reviews on Halodoc mobile application. The results of the sentiment analysis can provide insight into users’ overall satisfaction with the platform’s service quality. Results show that 84% users leave reviews with Neutral Sentiment Analysis, 13% with positive comments, and the smallest number of Sentiment Analysis is negative with 3%, in this application. We create correlation matrix to see correlation coefficients between sets of variables and the wordcloud. This research contributes to the growing body of research on sentiment analysis in the healthcare industry and can inform the development of strategies to improve the quality of online healthcare services.

The field of the Internet of Medical Things (IoMT) is experiencing significant growth as it involves the integration of medical devices and systems with the internet and various digital technologies. This study provides a bibliometric analysis of scholarly articles pertaining to the Internet of Medical Things (IoMT) during the time frame of 2018 to 2022. The analysis is conducted using data sourced from the Scopus database. A total of 1165 documents were identified and analyzed to gain insights into the current state of research in this field. The analysis includes trends in publication output, top authors and institutions, and key research topics. The results of the analysis indicate a steady increase in publication output on IoMT over the past five years, with the highest number of publications in 2022. The top authors in this field were Mohanty, S.P., Kougianos, E., and Pirbhulal, S., while the top institutions were King Saud University, Chinese Academy of Sciences, and University of North Texas. Co-citation analysis revealed that wearable devices, telemedicine, and data analytics were among the key research topics in IoMT. This bibliometric analysis offers useful insights into the present research landscape of the Internet of Medical Things (IoMT) spanning the years 2018 to 2022. The findings can be beneficial for researchers, practitioners, policymakers, and funding agencies in understanding the trends and priorities in this field. The top authors and institutions identified in this analysis can serve as valuable resources for collaboration and staying updated on the latest developments in IoMT.

A reverse vending machine is a device that uses garbage, such as used plastic bottles, to produce rewards when it is fed into the system one unit at a time. The loadcell sensor module HX711 and light-dependent resistor (LDR) sensor, as well as the ultrasonic sensor HCSR04, are used in the system’s design. The system also uses the Arduino Mega 2560 microcontroller. This study aims to design an embedded system to assist in the identification of plastic bottle waste based on volume, transparency, and weight, to observe the test findings, and to contrast the test results with earlier research on the plastic bottleneck detection system. The study approach involves integrating the many parts of the plastic bottle garbage detecting systems into a single system by building system integration. The Plastic Bottle Garbage Detection System developed and installed in the Reverse Vending Machine (RVM) provided detection results of up to 95.33% of the testing results of 33 sample plastic bottler garbs in various sizes tested ranging from 160 to 1500 ml with a different number of fruits per sample. On each sample tested, testing was done up to 30 times to see how many detections there were. As a reference and benchmark for the development of plastic bottle waste detection systems using straightforward components with better results in the future, this improved detection result is anticipated to be a fundamental focus for the field of sensing systems and electronics education.

Challenges related to corrosion significantly impact the durability and safety of mild steel structures when exposed to aggressive environments like seawater. This research explores the potential of Cocos nucifera leaf extract (CNLE) as a natural inhibitor to enhance the resistance of mild steel against corrosion in seawater. The study employs potentiodynamic polarization and linear polarization resistance techniques to assess the extract’s inhibitory properties. Additionally, it conducts an analysis of inhibition efficiency and adsorption isotherms using the Langmuir model to shed light on the adsorption mechanism. The inhibition efficiency shows an increase with higher extract concentrations, eventually reaching a maximum value. The adsorption isotherm analysis validates the applicability of the Langmuir model, indicating monolayer adsorption and a strong interaction between the inhibitor and mild steel. These findings contribute to the development of environmentally friendly and cost-effective methods for enhancing the longevity and reliability of mild steel materials used in seawater applications.

This systematic review explores the rising use of virtual clinics, including telemedicine technologies like virtual visits and virtual clinics, in maternity care. These technologies enable continuous monitoring of expectant mothers, offering valuable and satisfactory services, especially in the post-COVID-19 pandemic era. The study aims to comprehensively analyze the applications, effectiveness, benefits, challenges, and overall impact of virtual clinics in gynecological and pregnancy care. A systematic review search encompassing the period from 2019 to 2023 was performed on Scopus and PubMed databases. Additional manual searches on reference lists were conducted for comprehensive coverage. Eight articles were included, comprising various study designs, such as experimental, cohort, and cross-sectional studies. Pregnant women, parents, and healthcare practitioners were the target users. Virtual clinics were utilized for virtual visits, consultations, monitoring, follow-up, and home care, resulting in highly satisfactory patient experiences. However, challenges, including poor internet connection, were identified. The reviewed studies reveal promising outcomes in patient and provider satisfaction. Virtual clinics have the potential to revolutionize maternity care by providing accessible, efficient, and personalized healthcare services. Despite technical challenges, the integration of telemedicine solutions, particularly virtual clinics, is expected to play a growingly significant role in obstetric care. These insights are valuable for healthcare providers, policymakers, and researchers, leading to improved maternal care outcomes through the adoption of virtual clinics.

Evaluation through learning assessments is a fundamental factor in determining student’s success in achieving specified competencies. During the essay evaluation, the lecturer needs to check each assignment individually. However, when dealing with long essay answers, extra attention is necessary to extract the key points effectively. As a result, this process takes a lot of time and may potentially lead to carelessness or boredom during the assignment-checking process. The way to overcome this problem is by using Extractive summarization. The Extractive Summarization method summarizes by extracting key points from the student assignments and creating a summary without making any changes to the original text. Currently, Extractive Summarization widely uses deep learning technology, such as Bidirectional Encoder Representations from Transformers (BERT). BERT can effectively recognize contextual information within sentences. Output from BERT is sentence embeddings that can serve as valuable features for clustering. The summary result generated by applying k-means clustering to group sentences that have similarities and then selecting one primary sentence from each cluster to represent the cluster. This research proposed an approach for selecting sentence candidates for each cluster using TF-IDF weighting. The proposed method achieved the best ROUGE score on ROUGE-1 recall with 0.73003. We compare our results with the previous study’s BERT k-means approach, which selects sentence candidates from the closest sentences to the centroid for summary selection. The experimental results show that the proposed method achieves slightly better ROUGE scores than the previous study. Furthermore, in terms of execution time comparison, the proposed method has a shorter execution time.

In response to the burgeoning interest in tourist destination sentiment analysis, this study focuses on Indonesia’s renowned Raja Ampat. Our primary objective is to employ Latent Dirichlet Allocation (LDA), a topic modeling technique, to delve into sentiments expressed in a corpus of 5,227 TripAdvisor documents concerning Raja Ampat. Leveraging the Vader library for data preprocessing, our analysis reveals a dominant trend of positivity, with 80% of sentiments classified as positive, while 15.8% are neutral, and 4.2% are negative. Through systematic experimentation with Bag of Words (BoW) and Term Frequency-Inverse Document Frequency (TF-IDF) feature representations, two major topics emerge: ‘Environmental Tourism’ and ‘Snorkeling & Services.‘ Remarkably, the LDA model attains coherence scores of 0.6005 (BoW) and 0.4148 (TF-IDF), underscoring the robustness of our analysis. These findings not only contribute valuable insights into the prevailing sentiments and dominant topics associated with Raja Ampat tourism but also offer practical implications for the industry. The tourism sector in Raja Ampat can leverage these insights to enhance services and experiences, ultimately facilitating sustainable tourism development in this ecologically vital region. This research thus serves as a significant step toward promoting informed decision-making and improving tourist experiences in Raja Ampat and similar destinations, fostering sustainable growth and preservation of natural beauty.

The possibility of radiographers being exposed to harmful scattered radiation in X-ray examinations with X-ray mobile unit is higher because the locations are not possibly designed with protective shielding. The longer distance of the exposure position can minimize the exposure to the radiographer, but is limited to the length of the wire. This study aims to develop a wireless exposure switch with Bluetooth and microcontroller and to verify its performance for safer X-ray mobile examinations. Wireless exposure switch was developed based on Bluetooth version 4 and microcontroller Arduino Nano. Performance tests were performed in the laboratory, consisting of functional test to each component, connection test from 1–15 m, response time from the transmitter to the receiver, and its ability to produce X-ray image. Data were analyzed descriptively to find out the performance. The developed wireless exposure switch can trigger an exposure up to 15 m, 6 times longer than the conventional wire exposure switch. The average response times are 0.061 s for electrons forming and 0.059 s for X-ray forming. It is compatible with an installed X-ray mobile system to produce X-ray images. The Bluetooth and microcontroller-enabled exposure switch are functionally tested and applicable to the X-ray mobile unit. It is prospective to be applied for minimizing radiation exposure for radiographers.

This paper discusses a comprehensive review related to Battery Swap Station (BSS) technology in Indonesia and proposes the prototype of an IoT-based battery swap station management system in Indonesia. The spread of BSS in Indonesia can be reviewed in terms of several things that support the development of BSS in Indonesia, such as the type of batteries and technology used on BSS. Some types of batteries can be distinguished according to the brands that exist in Indonesia, including Gogoro, Viar, Smoot, Gesits, and Volta. Battery technology can be reviewed from the connector or plug on the BSS that connects the battery, Battery Management Systems (BMS), and the app to the user. By 2022, there are 369 BSSs scattered across Indonesia, which will continue to be pushed by the Indonesian government for the use of more massive battery-based electric motor vehicles. This leads to the heterogeneity issue of several different battery voltages in the Indonesian market that the suggested prototype can address.

Business intelligence in healthcare continues to evolve and is essential for increasing healthcare system efficiency, improving patient data management, developing entrepreneurship, and providing deeper insights for healthcare professionals to improve clinical services and outcomes. This study attempts to visually map the sector and examine current business intelligence trends in the healthcare sector. The study analyzes 547 secondary data publications from the Scopus database that were published between 2002 and 2022 using a bibliometric knowledge structure and level of analysis. According to the findings of the research conducted at Universidade do Minho, Portugal, and the United States were the most productive research institutions and countries related to business intelligence in healthcare. This research suggests classifying business intelligence in healthcare studies themes, such as business analytics, information systems, health service, internet of things, and analytics, into BIHIA research themes.

Assessing respiratory rate is pivotal to monitoring an individual's health status. This research endeavors to meticulously evaluate the efficacy of three distinct mechanical sensors: piezoelectric sensors, force-sensitive resistor sensors, and Flex sensors, in the precise measurement of respiratory rates. The experimental design involved the strategic placement of these sensors onto a mannequin's chest cavity, enabling the conversion of chest pressure into tension. Subsequently, a sophisticated data processing system, driven by Arduino technology, was employed, with the results elegantly displayed through a Delphi interface. The outcomes of this extensive experimentation yielded compelling results, with the force-sensitive resistor sensor emerging as the standout performer, boasting a remarkably low error value of 0.2781 ± 0.4481. This exceptional precision was particularly notable when assessing respiratory rates at setting 20. Such findings underscore the pragmatic application of the force-sensitive resistor sensor within healthcare equipment systems, promising user-friendly operation and efficient retrieval of invaluable patient data. This study not only adds depth to our understanding of the suitability of mechanical sensors for respiratory rate monitoring but also accentuates the transformative potential of the force-sensitive resistor sensor in elevating measurement accuracy to unprecedented levels. Furthermore, it lays the foundation for future research endeavors, with promising directions including sensor refinement and the implementation of cutting-edge data processing techniques, aimed at ushering in a new era of excellence in respiratory rate assessment for healthcare applications.

Obesity is a health problem in the twenty-first century, and can trigger several dangerous diseases. Several studies have been conducted in predicting obesity rates. This study aims to propose an obesity prediction approach using a Self Organizing Map by considering habit parameters and clinical variables as input from the system. The variables considered are gender, age, height, weight, family history with overweight, FAVC, FCVC, NCP, CAEC, whether smoking or not, CH2O, SCC, FAF, TUE, CALC, and transportation information used every day. In this study, the level of obesity was divided into 7 cluster levels, namely insufficient weight, normal weight, overweight level I, overweight level II, obesity type I, obesity type II, and obesity type III. The stages of this research are data preprocessing, initialization process, learning process, and testing process. The learning process is carried out using a Self Organizing Map. System input consists of 16 parameters with 1 obesity level system output. The learning and testing process has been carried out and the results of the first testing process were 81.4%, the precision value was 0.825, while the recall value was 0.803. The second testing process obtained a system accuracy value of 74.3%, the precision value was 0.753, while the recall value was 0.743. However, this system has been able to classify according to the cluster well. The contribution of this research is that it can be a decision support system in predicting obesity levels, and based on the variables considered users can improve their living habits to avoid obesity and live healthier in the future.

The motor impairments that stroke patients experience as a primary contributor to death and disability also impact their quality of life. Due to its characteristics as complex muscle signals, the Electromyograph (EMG) signal may be one of the markers in rehabilitation and a substitute that may be used to judge the success of rehabilitation. This study aims to analyze the complexity of EMG signals originating from the flexion of the flexor carpi radialis muscle with various weights indicating the rehabilitation state. It is important to monitor the progress of rehabilitation, and it can be done by using complexity analysis. The complexity of the EMG is examined in this work utilizing entropy measurement, namely Approximate Entropy (ApEn), Sample Entropy (SampEn), and Fuzzy Approximate Entropy (fApEn). We concentrated on the flexor carpi radialis muscle using flexion–extension exercises and varying the weight load. This study’s results indicate differences in the complexity of the EMG signal in the flexor carpi radialis when performing flexion–extension movements. Three differences showed a correlation based on the value of ApEN, SampEn, and fApEn. Complexity Analysis using fApEn showed the best correlation at R = 0.9966, which can be used to evaluate the rehabilitation process. The increase in weight was followed by the increase of the fApEn value, indicating that the value can be used to evaluate the rehabilitation process. The use of complexity analysis of EMG signal from flexion movement in flexor carpi radialis with various weights can be used to indicate the progress in the rehabilitation process.

Manageable symptoms can be a critical and important thing for people with Parkinson’s disease (PWP) in order to maintaining their quality of life. PWP early detection and monitoring is one of way to understand whether the medication dosage and physical therapy manage to maintain the stage of symptoms. The cheapest monitoring method can be done is based on voice signal. PD detection method based on voice signal shows promising future to be implemented into real world through an online and mobile based medical related application. However, It is necessary to select the important voice features which contribute to highest detection accuracy so that the implementation of detection through the application can be done effectively without requiring complex mathematical code. This study aim to evaluate and analize the highest accuracy and suitable voice feature related to PWP early detection and monitoring using machine learning method. Voice data recorded from study participants consist of Hughes-based stages of Parkinson’s disease (PD) patients and healthy subjects. Participants recorded their voice said ‘aaaa..’ for 5 s then the voice data calculated into 22 voice features. Those features then examine using machine learning methods such as logistic regression, random forest, KNN and deep learning CNN and classified into four classes based on Hughes standard e.g. healthy, possible, probable and definite. The experimental result showed that the most suitable features are 11 features out of 22 features which examined using random forest and CNN method contributed to highest accuracy value of 95%. This most important features then can be implemented along with CNN method into an online and mobile based application for future study.

In orthopedic surgery, the bone drilling task is the main factor for the successful completion of the surgery. The bone drilling task depends on a high level of dexterity and experience of the orthopedist and surgeon. It is because the bone drilling resistance is relatively large and sometimes violent vibrations might cause difficulties in grasping the hand-piece. In a worse case, it might even break the slender drill. The objective of this paper is to gain an understanding of the frequency properties of bone in order to improve visualization and training. These properties can be detected using different imaging methods or techniques for processing signals. The experimental setup includes the robotic arm to provide an accurate thickness layer and consistent penetration of the drilling. Three-axis accelerometer equipped with National Instrument data acquisition (DAQ) was used in the experiment to acquire the vibration signal on different bone layers. This study proposes a successful approach to categorize bone drilling levels using a Convolutional Neural Network (CNN) with a customized architecture designed for this purpose. The CNN is utilized to classify the raw vibration signal into three distinct labels or layers, namely periosteum, first cortical, and spongy. The results of the study indicate that the CNN can accurately classify the three bone layers, with a higher degree of accuracy for the periosteum and first cortical layers, achieving over 98% accuracy, and a 100% accuracy for the spongy layer. This is due to the unique vibration signal of the spongy layer, which differs from the other two layers.

Pneumonia is a significant respiratory condition that can cause severe health complications in adult patients if not diagnosed and treated promptly. This study focuses on enhancing the accuracy of pneumonia detection in adult patients using Convolutional Neural Networks (CNNs). The primary objective of this research is to improve upon the previous accuracy rate of 89% achieved by prior studies. A CNN architecture with five hidden layers was designed and implemented to pursue this goal. The dataset consisted of 5,000 medical images used for training, testing, and random checks to assess the model's performance comprehensively. In this research, we compared various compositions to achieve the highest precision. Changes in the number of image data with a normal data value of 1800 and pneumonia data of 3200 achieve the highest degree of precision, namely 0.9167. The accuracy of changes to the epoch value 20 is 0.9721. Using a size of 244 × 244 yields a precision of 0.9443. The proposed CNN architecture significantly advanced pneumonia detection accuracy through meticulous data preprocessing, augmentation, and training. The experimental results showed a notable improvement in accuracy, reaching a value of 91%. This enhancement can be attributed to the depth and complexity of the CNN architecture, along with the augmentation techniques employed. The achieved accuracy indicates the potential of CNNs in aiding medical professionals with more reliable and accurate pneumonia diagnoses in adult patients. The findings of this study underscore the efficacy of using CNNs in medical image analysis, specifically for pneumonia detection. Further exploration of larger datasets, more intricate architectures, and incorporation of additional diagnostic features could potentially lead to even higher accuracy rates, contributing to improved patient care and diagnostic outcomes.

Heart signals and heart rate are important aspects that need to be monitored in the care of premature babies because they describe the health condition of premature babies. This research aims to develop a baby incubator with accurate Electrocardiogram (ECG) monitoring that can facilitate monitoring of the baby's condition directly so as to get the right treatment. This research uses an Electrocardiograph circuit, temperature sensor, NTC skin sensor, loadcell sensor, and ultrasonic sensor. Arduino is used as a processor of all data and in ECG processing the Pan-Tompkins algorithm is used to process the QRS signal which will then detect the R peak and calculate the accurate heart rate value. The testing procedure of this module is carried out by taking data directly from 10 subjects and then the data obtained is processed to obtain accuracy regarding heart signals and heart rate. The main result of this research is the development of a baby incubator that can monitor heart signals in real time using the Pan-Tompkins algorithm. Measurements using a comparison tool showed an average heart rate measurement error of 0.2%. It was concluded that the incubator performance showed good results. It was concluded that the incubator performance showed good results. The results of this study can be used for doctors to monitor the condition of premature babies remotely using a smartphone or computer so that it is expected to accelerate the analysis and treatment of patients.

Traditional methods of creating prosthetics are often expensive, time-consuming, and lack personalization. However, 3D printing technology has emerged as a transformative force that can address these challenges. The purpose of this abstract is to analyze the research documents on this topic from Scopus, a database of peer-reviewed literature and scientific information. The search query was TITLE-ABS-KEY (prosthetic AND hand AND 3d AND printing) AND PUBYEAR > 2012 AND PUBYEAR < 2024, which returned 236 documents in total. The document analysis reveals that the research on prosthetic hands using 3D printing technology has increased over time, especially in 2020 and 2022. In conclusion, the research on prosthetic hands using 3D printing technology is a growing and diverse field that involves many researchers, institutions, countries/territories, and disciplines. The use of various filaments such as PLA, TPU, and silicone rubber has enabled the creation of low-cost, functional, and customizable prosthetic devices. However, there is still room for improvement and further research in this field to continue advancing this important technology.

This review focuses on various control systems and provides a thorough overview of developments in wearable hand exoskeletons for post-stroke rehabilitation. The Scopus database was used to compile a dataset of scholarly papers in order to comprehend the state of the study in this area at the moment. A total of 1482 papers from 84 different nations were found. Using a VOS viewer, the acquired data was analyzed to identify significant clusters and relationships in the literature. The study identifies four major clusters, each reflecting a different facet of control techniques in wearable hand exoskeletons: human, robotic exoskeletons, electromyography, and biomechanics. The analysis demonstrates an increasing interest in novel control strategies, including hybrid control systems, brain-computer interface (BCI) control, myoelectric control, and impedance control. These tactics present opportunities for improving post-stroke patients’ rehabilitation outcomes. This review article clarifies the quick developments and various control methods in wearable hand exoskeletons for post-stroke therapy. In order to provide efficient and individualized solutions, the analysis emphasizes the value of multidisciplinary collaboration among engineers, physicians, and researchers. For academics and practitioners hoping to contribute to the continued advancement in this crucial area of rehabilitation technology, the identified problems and future directions offer insightful information that is helpful.

Electrocardiogram (ECG)-Derived respiratory methods have become a non-invasive technique for monitoring and assessing respiratory functions. The purpose of this study was to evaluate the efficacy of various ECG-derived respiratory (EDR) methods based on signal decomposition techniques, including empirical mode decomposition (EMD), variational mode decomposition (VMD), empirical wavelet transforms (EWT), and kernel principal component analysis (KPCA). The contributions obtained Were able to provide information about respiratory function non-invasively. This research compares several approaches’ concepts, benefits, limitations, and comparative performance using the correlation and coherence coefficients as evaluation metrics. With correlation coefficients of 0.91 and 0.94, respectively, and coherence coefficients of 0.95 and 0.97, EWT and KPCA perform the best in accuracy and robustness. The paper indicates that EWT is the most effective and efficient decomposition approach for extracting respiratory signals from ECG data, as it can adapt to the dynamically varying features of ECG signals, decrease noise, and provide smooth and accurate EDR signals. The article suggests that EDR techniques based on decomposition techniques can provide a non-invasive and continuous monitoring strategy for respiratory parameters. Nonetheless, they encounter obstacles and constraints that must be addressed to improve their clinical usefulness.

Electromyography (EMG) signals have emerged as vital tools for prosthetic control, motor function assessment, and rehabilitation technology. Traditional methods, relying on manual feature extraction, struggle with high-dimensional EMG data. Deep learning, specifically Convolutional Neural Networks (CNNs), has shown promise in addressing these challenges. However, there is a need to explore CNNs’ effectiveness, optimal architecture, real-time processing, and data augmentation for EMG signal classification. This study aimed to introduce a CNN-based approach tailored to classify ten-class, two-channel EMG signals. Eight participants provided EMG data by performing ten distinct finger and hand movements. Data were collected using two-channel EMG sensors, amplified, sampled at 4000 Hz, and filtered. A CNN architecture with eight layers was designed, incorporating one-dimensional convolutional layers, max pooling, global average pooling, dropout layers, and a dense output layer. The CNN-based approach demonstrated promising results. It exhibited high accuracy in classifying EMG signals for most classes, with an overall accuracy of approximately 91%. It excelled in the thumb (THU) and thumb-ring (TH-R) classes. However, some classes, like thumb-middle (TH-M) and thumb-index (TH-I), showed room for improvement in terms of recall. This study introduced an innovative CNN-based approach to tackle the challenges of classifying ten-class, two-channel EMG signals. It addressed critical research gaps and significantly advanced EMG signal processing, particularly for prosthetic control and rehabilitation technology.