Search

This concise analysis examines the application of AI, particularly chatbots and natural language processing (NLP), to enhance maternal-infant healthcare in Spain and Brazil, where maternal mortality rates are significantly different, as well as birth rate and reproductive health planning or birth control. The application will be preliminary tested in complex situations regarding these two use cases. It highlights the integration of disruptive technologies into prenatal care, emphasizing the role of AI in facilitating personalized healthcare and self-management through data acquisition on health parameters, such as weight gain, blood pressure, incontinence, mental health, nutrition and physical activity. The study reflects on the impact of digital tools in optimizing patient engagement and early diagnosis of alarm symptoms, showcasing the potential of m-Health applications in transforming care practices and contributing to attaining Global Health Objectives for 2030. [ABSTRACT FROM AUTHOR]

Traditional methods for assessing lower body strength and balance, particularly in older individuals, lack objective and comprehensive analysis. The Up-Down Hop Test is widely used, yet leveraging mobile device sensors, specifically accelerometers and gyroscopes, can enhance the effectiveness of the Up-Down Hop Test by providing an objective evaluation of lower body strength, balance, and neuromuscular control. This study involved attaching mobile devices to participants during the Up-Down Hop Test. Data from the sensors were analyzed to assess patterns of body imbalance and fatigue, focusing on older participants. Integrating mobile sensor data offered insights into the participants' physical performance, revealing specific patterns of imbalance and fatigue. Although traditional assessment methods can observe these patterns, they lack measurability and auditability when evaluating the physical performance of older adults. The experimental results, obtained from analyzing data from mobile device sensors during the Up-Down Hop Test, revealed patterns of imbalance. These findings can be utilized in health monitoring and targeted intervention scenarios, specifically those aimed at enhancing physical performance and preventing injuries. [ABSTRACT FROM AUTHOR]

Skin lesions refer to changes or abnormalities on the skin, such as moles, freckles, cysts, warts, and more. While not all skin lesions are cancerous, some types can indicate or develop into skin cancer. Skin cancer has been on the rise in recent years, causing many deaths. However, manual examination by dermatologists is expensive and time-consuming. Research in this field has focused on developing an automated system to detect skin cancer and reduce the associated mortality rate. However, the performance of such systems needs improvement. In this study, we propose a Synthetic Minority Over-sampling Technique combined with Edited Nearest Neighbors (SMOTEENN) to augment the data, resulting in a more balanced dataset. Then, we used a Convolutional Neural Network (CNN) model on the benchmark HAM10000 dataset. The dataset comprises seven skin diseases, including Actinic Keratoses, Basal cell carcinoma, Benign keratosis, Dermatofibroma, Melanocytic nevi, Melanoma, and Vascular skin lesions. With a balanced dataset, the proposed model achieves an accuracy, precision, recall, and f1-score of 99.35%, 99.00%, 99.00%, and 99.00%, respectively. The balanced dataset performs 22% better than an unbalanced dataset. [ABSTRACT FROM AUTHOR]

Healthcare is a global pillar, with a surge in the adoption of information technology, particularly in hospital information systems (HIS). However, global protocols are needed to meet the growing demand for data interchange, practical implementations for sharing healthcare data among facilities, and a pressing need for processing and storage infrastructure to handle the escalating volume of healthcare data. This study proposes a solution for efficient data transmission using electronic health records (EHR) and Platform-as-a-Service (PaaS) to leverage cloud computing resources. This framework's architecture boasts robustness and adaptability, providing all registered software programs access to data interchange services. Through a comprehensive examination of the framework's structure, the essay also explores the most effective data-sharing methods. It identifies the healthcare system's optimal EHR data model. According to multiple healthcare experts, the operational building of this framework is expected to catalyze the growth of healthcare institutions both nationally and within specific industries. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a type of dementia that leads to memory loss and impairment, which afects patients' lives badly. It is not curable yet, but its progression can be slowed down if detected at earlier stages. In this research study, we propose a transfer learning-based convolutional neural network (CNN) model to classify magnetic resonance imaging (MRI) into one of four stages of Alzheimer's disease. One of the major limitations of the deep learning-based classification model is the non-availability of healthcare datasets related to AD. The widely used Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset has a major class imbalance issue. We propose a generative adversarial network (GAN) based data augmentation technique to overcome the data imbalance. This promotes the investigation of applying GANs to generate synthetic samples for minority classes in Alzheimer's disease datasets to enhance classification performance. The results show the progression in the overall classification process of AD. [ABSTRACT FROM AUTHOR]

This concise analysis examines the application of AI, particularly chatbots and natural language processing (NLP), to enhance maternal-infant healthcare in Spain and Brazil, where maternal mortality rates are significantly different, as well as birth rate and reproductive health planning or birth control. The application will be preliminary tested in complex situations regarding these two use cases. It highlights the integration of disruptive technologies into prenatal care, emphasizing the role of AI in facilitating personalized healthcare and self-management through data acquisition on health parameters, such as weight gain, blood pressure, incontinence, mental health, nutrition and physical activity. The study reflects on the impact of digital tools in optimizing patient engagement and early diagnosis of alarm symptoms, showcasing the potential of m-Health applications in transforming care practices and contributing to attaining Global Health Objectives for 2030. [ABSTRACT FROM AUTHOR]

Traditional methods for assessing lower body strength and balance, particularly in older individuals, lack objective and comprehensive analysis. The Up-Down Hop Test is widely used, yet leveraging mobile device sensors, specifically accelerometers and gyroscopes, can enhance the effectiveness of the Up-Down Hop Test by providing an objective evaluation of lower body strength, balance, and neuromuscular control. This study involved attaching mobile devices to participants during the Up-Down Hop Test. Data from the sensors were analyzed to assess patterns of body imbalance and fatigue, focusing on older participants. Integrating mobile sensor data offered insights into the participants' physical performance, revealing specific patterns of imbalance and fatigue. Although traditional assessment methods can observe these patterns, they lack measurability and auditability when evaluating the physical performance of older adults. The experimental results, obtained from analyzing data from mobile device sensors during the Up-Down Hop Test, revealed patterns of imbalance. These findings can be utilized in health monitoring and targeted intervention scenarios, specifically those aimed at enhancing physical performance and preventing injuries. [ABSTRACT FROM AUTHOR]

Skin lesions refer to changes or abnormalities on the skin, such as moles, freckles, cysts, warts, and more. While not all skin lesions are cancerous, some types can indicate or develop into skin cancer. Skin cancer has been on the rise in recent years, causing many deaths. However, manual examination by dermatologists is expensive and time-consuming. Research in this field has focused on developing an automated system to detect skin cancer and reduce the associated mortality rate. However, the performance of such systems needs improvement. In this study, we propose a Synthetic Minority Over-sampling Technique combined with Edited Nearest Neighbors (SMOTEENN) to augment the data, resulting in a more balanced dataset. Then, we used a Convolutional Neural Network (CNN) model on the benchmark HAM10000 dataset. The dataset comprises seven skin diseases, including Actinic Keratoses, Basal cell carcinoma, Benign keratosis, Dermatofibroma, Melanocytic nevi, Melanoma, and Vascular skin lesions. With a balanced dataset, the proposed model achieves an accuracy, precision, recall, and f1-score of 99.35%, 99.00%, 99.00%, and 99.00%, respectively. The balanced dataset performs 22% better than an unbalanced dataset. [ABSTRACT FROM AUTHOR]

Healthcare is a global pillar, with a surge in the adoption of information technology, particularly in hospital information systems (HIS). However, global protocols are needed to meet the growing demand for data interchange, practical implementations for sharing healthcare data among facilities, and a pressing need for processing and storage infrastructure to handle the escalating volume of healthcare data. This study proposes a solution for efficient data transmission using electronic health records (EHR) and Platform-as-a-Service (PaaS) to leverage cloud computing resources. This framework's architecture boasts robustness and adaptability, providing all registered software programs access to data interchange services. Through a comprehensive examination of the framework's structure, the essay also explores the most effective data-sharing methods. It identifies the healthcare system's optimal EHR data model. According to multiple healthcare experts, the operational building of this framework is expected to catalyze the growth of healthcare institutions both nationally and within specific industries. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is a type of dementia that leads to memory loss and impairment, which afects patients' lives badly. It is not curable yet, but its progression can be slowed down if detected at earlier stages. In this research study, we propose a transfer learning-based convolutional neural network (CNN) model to classify magnetic resonance imaging (MRI) into one of four stages of Alzheimer's disease. One of the major limitations of the deep learning-based classification model is the non-availability of healthcare datasets related to AD. The widely used Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset has a major class imbalance issue. We propose a generative adversarial network (GAN) based data augmentation technique to overcome the data imbalance. This promotes the investigation of applying GANs to generate synthetic samples for minority classes in Alzheimer's disease datasets to enhance classification performance. The results show the progression in the overall classification process of AD. [ABSTRACT FROM AUTHOR]

Sensor-equipped driver monitoring systems are a significant advancement in vehicle safety technology, continuously monitoring drivers' condition and behavior using various sensors. These systems identify exhaustion, inattention, or impairment indicators to improve road safety. They can measure the driver's focus and attentiveness by tracking their head position, eye movements, and facial expressions. If detected, the system can notify the driver via visual, aural, or tactile alerts, prompting the driver to refocus or take a break. Some sophisticated systems can even take over the car's steering in emergencies. The study focuses on integrating artificial intelligence (AI) and machine learning (ML) into Driver Monitoring Systems (DMS) to reduce accidents caused by driver exhaustion and distraction. The study discusses two primary DMS systems: Multi-Sensor Based Systems (S-HDx) and Vision-Based Systems (V-HDx). Vision-based systems examine drivers or specific driving characteristics to identify whether a driver is inattentive or sleepy. Multi-sensor-based systems identify drowsiness using physiological or non-visual factors. The study evaluated AI and ML algorithms for fatigue and distraction detection in drivers. Algorithms used in vision-based systems include single-shot detection, MobileNetV2, feature pyramid networks, and convolutional neural networks. Multi-sensor-based systems use algorithms such as SVM, CNN, XGBoost, and decision trees. Vision-based systems are recommended for DMS development due to their user-friendliness and non-intrusive nature. Future research could independently examine these techniques and algorithms to create a more successful DMS. [ABSTRACT FROM AUTHOR]

The need for creative solutions in real-time health monitoring has been highlighted by the rise in health-related incidents involving drivers of motor vehicles. It has led to the development of wearable technology that seamlessly integrates with the Internet of Medical Things (IoMT) to improve driver safety and healthcare responsiveness. The development of a revolutionary wearable technology system is presented in this study as an innovative approach to vehicle safety and healthcare. This system's real-time ability to track a driver's health is a significant development in guaranteeing driver safety and wellness. The study examines the hardware component's complex design and implementation, particularly concerning the printed circuit board (PCB) layout and electrical schematic. The gadget emphasizes wearability, robustness, affordability, and user-friendliness and is a shining example of valuable and effective medical technology. The research delves deeper into possible improvements for the system, like adding complex algorithms and a user-friendly interface. Enhancing user involvement and system intelligence hopes to maximize the system's potential for real-time health monitoring. The significance of this study in utilizing Internet of Medical Things (IoMT) technology is highlighted by its junction with multiple fields, including electronics, hardware engineering, human-computer interaction, and health informatics. This dissertation emphasizes the potential of wearable technology in bridging the gap between healthcare monitoring and vehicle safety by focusing on real-time health monitoring in the automotive context. [ABSTRACT FROM AUTHOR]

To bridge the digital barrier between those with disabilities and those without impairments, several applications can be operated by brain signals. It highlights the importance of employing brain-computer interfaces (BCI) and mentions a specific system called the "Brain-Controlled Computer System" that can control a computer cursor and perform mouse actions using brain signals.. The system can control the cursor and provide mouse left/right click with brain signals using a low-cost Brainlink Lite device. Based on this system, we employed machine learning and deep learning methodologies to develop a system that classifies brain signals for cursor movement. The methodology involved gathering user input through surveys, interviews, and comparison studies, as well as utilizing pre-developed datasets collected using a mind sensor. We initially trained a model on a previous dataset, achieving approximately 70% accuracy. However, when applied to the new dataset, the accuracy dropped to 49%. To improve the results, we experimented with more complex models, including additional layers and a dropout layer to prevent overfitting. Despite these enhancements, the accuracy only reached a maximum of 52%. The research team collected a dataset tailored to their specific application. They made a groundbreaking discovery by applying a convolutional neural network (CNN), typically used for image analysis, to process continuous brainwave data. Despite this non-standard use, their hypothesis proved correct, with the CNN model achieving an impressive 80% accuracy in classifying brain signals for controlling a cursor. [ABSTRACT FROM AUTHOR]

Federated learning (also known as collaborative learning) is a decentralized approach to training machine learning models. In 5G smart healthcare, federated machine learning (FML) can potentially improve patient care by offering improved diagnosis, prognosis, and therapy models. Nevertheless, a significant worry regarding FML is its lack of security. Within the context of 5G smart healthcare, this review paper looks at FML from a security point of view, discussing the benefits and risks of using FML in 5G smart healthcare and the possible solutions to these risks. The issues of privacy, adversarial attacks, communication security, and malevolent clients are brought up in the discussion on security challenges. Differential privacy, secure aggregation and training, adversarial training, secure communication, client authentication, and model pruning are some of the solutions that have been suggested. We will be able to protect the privacy of patient data in FML if we take the necessary steps to address these security problems. [ABSTRACT FROM AUTHOR]

As a society, we should provide support and care for individuals who are advanced in age or have special needs. We must offer them the opportunity to lead a safe and comfortable life. Sometimes, addressing simple things can mitigate critical problems. For instance, an improperly functioning heating system can put the users' lives at risk, and a gas leak can prove fatal if not detected in time. Hence, this study aims to identify solutions that can help prevent accidents in such vulnerable populations. To achieve this, a safety device equipped with essential sensors continuously monitors the living environment of people with special needs. If there are any anomalies, the system will immediately alert the concerned family members and competent security authorities, ensuring a quick and efficient response to potential emergencies. The proposed safety device aims to provide extra protection and peace of mind for special care individuals. By implementing such safety measures, we can make significant strides in promoting the well-being and safety of vulnerable populations. [ABSTRACT FROM AUTHOR]

Mobile users can use a mobile sensor to record ECG data, and on-device ECG classification can provide a more efficient diagnosis than standard care. Deep Neural Networks (DNNs) have excelled in artificial intelligence (AI)-based applications to extract patterns from the complex waveform to analyze the ECG waveform. Despite their superior performance, their high computational complexity severely limits their use in resource-constrained mobile phones. This study aimed to analyze various lightweight deep learning models for ECG classification on edge devices while maintaining comparable performance. Five DNNs, including SqueezeNet, ShuffleNet, MobileNet, EfficientNet, and NASNet, designed explicitly for resource-constrained mobile, are tested and compared for ECG classification. Time-series ECG signals are converted to a 2D scalogram using the continuous wavelet transform. These generated scalograms are classified into three categories: cardiac arrhythmia (ARR), congestive heart failure (CHF), and normal sinus rhythms (NSR). A comprehensive comparison is presented using accuracy analysis and the required trainable parameters among these models. The experimental analysis of PhysioNet's dataset shows that ShuffleNet has the fewest learnable parameters, with a validation accuracy of 87.50%. In contrast, SqueezeNet performed better, with 93.75% validation accuracy and 724K learnable parameters. [ABSTRACT FROM AUTHOR]

Depression is a mental disease affecting 5% of the population, and its prevalence is increasing. Depression is characterized by feelings of worthlessness, hopelessness, disinterest in enjoyable activities, and sadness, which can result in suicidal thoughts. Traditional approaches to recognizing depression have relied on manually crafted techniques to extract facial expressions, which have their limitations. To address these limitations, this paper proposes using convolutional neural networks (CNNs) as a practical approach for depression recognition. The proposed model in this study involves an eight-step process that includes input data, preprocessing, rescaling, model training, multi-classified results, selecting emotions based on accuracy, retraining the model, and finally, multi-classified results to determine the percentage of depression with greater accuracy. [ABSTRACT FROM AUTHOR]

Web Accessibility refers to apps, websites, and digital tools designed to be accessible by all users, including those with disabilities. The web is a source of public information, education, employment, government, commerce, health, entertainment, and many others, and every person, regardless of ability, has the right to access it. Imagine missing out on thousands of new customers because your competitor has an easier-to-use, WCAG-compliant website. This systematic literature review aims to analyze the empirical methods of evaluating accessibility to educational websites, disabilities, and errors. Some of the papers we will review here use automated tools, real users, or experts to evaluate web accessibility. After analyzing the results, we will determine if the websites comply or not with the Web Accessibility Content Guidelines (WCAG). Also, what are the differences between automated tools, real users, and experts in evaluating websites. [ABSTRACT FROM AUTHOR]

• The article presents a semi-automated tool for drawing complex Data Flow Diagrams. • A Natural Language Interface for processing user query was proposed. • Web scrapping and extraction techniques were used to collect software descriptions. • Using rule-based algorithms for mapping user query with components of DFDs. • The mechanism of graph theory was applied to gauge the accuracy of output. To model the data and functions in various computer science applications, the researcher uses a Data Flow Diagram (DFD). DFD has been constructed using [open-source software tools that provide users with different shapes and environments. However, the existing approaches require substantial human effort, the validity of the generated output is still a loophole, and they have never gained traction in practice. Our research objective is to develop a semi-automated tool for drawing complex Data Flow Diagrams in the shortest time according to the specified features of the intended system. We developed a Natural Language Interface (NLI) that allows the user to compose a query and identify the system functionality and constraints for the composition of DFD. Natural Language Processing (NLP) techniques are applied to scrapped data to extract the keywords and develop a data repository. Also, we developed rule-based algorithms to map user queries onto respective token shapes to draw the required functionality into appropriate levels of DFD. For verification, output DFDs were converted into conceptual digraphs using adjacency and permutation matrices to evaluate isomorphism. The empirical results reflect that the DFDs generated by the system are correct, complete, and significant. [ABSTRACT FROM AUTHOR]

COVID-19 has infected several million of individuals while claiming numerous lives. This fact raised the need to apply the measure to prevent its transmission. The use of disinfection products, wearing masks, and avoiding touching doors are important measures to avoid its spread. Thus, this work proposes a framework supported by a Convolutional Neural Network (CNN) model checking the hygienic conditions of the individuals requiring authorization to access facilities. The experimental work takes IoT devices with sensors to check: whether the users have disinfection product in their hands and a trained model to check whether individuals are also wearing masks. The achieved results highlighted the effectiveness of the proposed framework. [ABSTRACT FROM AUTHOR]

The provision of health care is an area that is excessively dependent on the access, processing, analysis, sharing, and storage of data between different stakeholders. Consequently, cloud computing has been adopted in the most diverse areas (e.g., telemedicine, information management, bioinformatics) as a promising model in responding to this need, offering numerous advantages and opportunities. However, the massive migration of the sector to this new paradigm remains far from being achieved. Essential questions and challenges remain unanswered, and the Chief Information Officer (CIO) proves to be a fundamental agent for change to happen. This paper describes a literature review based on 17 survey and systematic review articles published between 2010 and 2021 on the applications, opportunities, and challenges of adopting cloud computing in the health sector. The results discussed may imply different strategies in CIOs' decision-making and contribute to organizational success in this context. [ABSTRACT FROM AUTHOR]

The evolution of knowledge and information has changed the practices of health services over the years. The use of Information systems (IS), with the underlying Information Technology (IT), plays a vital role in managing medical information and associated resources, always to provide a better quality of health care. IS/IT is increasingly developing globally, with new features and tasks greatly expanded, leaving paper records and voluminous files behind. The ubiquitous information and computing resources increasingly allow the sharing of all types of data between various entities. New IS/IT will have to be thought to correspond to the new types of monitoring and analysis of biomedical data. This work intends to explore the IS/IT in health and its evolution over the last few years. [ABSTRACT FROM AUTHOR]

Currently, heart diseases are prevalent in the population. Machine learning methods may help in the identification of heart diseases in the different people with the analysis of various features of heart rate, such as PPE, spread, spread2, MDVP:Fo(Hz), MDVP:Shimmer, MDVP:Shimmer(dB), Shimmer:APQ3, Shimmer:APQ5, MDVP:APQ, Shimmer:DDA, DFA, RPDE, D2, MDVP:Fhi(Hz), MDVP:Flo(Hz), NHR, HNR, MDVP:Jitter(Abs), MDVP:Jitter(%), MDVP:RAP, MDVP:PPQ, and Jitter:DDP. The analysis was performed with the dataset from the UCI Machine learning repository from the Center for Machine Learning and Intelligent Systems. This paper proposes the use of different methods, such as Neural Network, Decision Tree, k-Nearest Neighbor (kNN), Combined nomenclature (CN2) rule inducer, Support Vector Machine (SVM), and Stochastic Gradient Descent (SGD). The best results on the 20-fold Cross-validation and the 10-fold Cross-validation are reported by DT and SVM methods (87.69%). Also, the best results on the 5-fold Cross-validation are reported by SGD (87.69%). [ABSTRACT FROM AUTHOR]

The accelerometer is available on most of these mobile devices. It allows the acquisition and calculation of different physical parameters. Due to the use of pattern recognition, it also enables the identification of several Activities of Daily Living (ADL), such as walking, running, going downstairs, going upstairs, and standing. The feature extraction step performs the extraction of the five most significant distances between peaks, the average, standard deviation, variance and median of extracted peaks and raw data, and the maximum and minimum of raw data. The focus of this paper is the implementation of multiple artificial intelligence methods for the recognition of ADL, including Logistic Regression, Combined nomenclature rule inducer, Neural Network, Naive Bayes, Support Vector Machine, Decision Tree, Stochastic Gradient Descent, and k-Nearest Neighbor. The Decision tree reported the average accuracy of 85.22% between classes. This method also presents an F1-score value of 85,13% and a precision value of 85,08%. Nevertheless, the study has limitations associated with the use of mobile devices. The position and location of the device in the data collection phase need further investigation, and the system architecture demands higher energy consumption. [ABSTRACT FROM AUTHOR]

Every year, the percentage of people affected by cardiovascular diseases increases drastically. Out of them, a heart attack is the most prominent and painful disease. According to the World Health Organization, approximately 17.5 million people lose their lives yearly due to this disease, which is alarming. The remarkable advancement in wearable technology has opened doors to propose many effective smart solutions to tackle this disease efficiently. Furthermore, early diagnosis of heart attack proliferates the compatibility of meditation and expedites the diagnostic recommendation by clinical experts. Considering this problem's sensitivity, we proposed a wearable smart and early Heart Attack diagnosis system and adopted a decentralized computational phenomenon using hybrid computing architecture. It reflects better response time and minimal latency to detect a heart attack in its preliminary stage for homage patients. The proposed system can monitor and trigger the patient current heart status classified on required heart diagnosis parametric sensors assembled on the patient's body with the help of an android application. In this study, three models are developed using the Support Vector Machine (SVM), Adaptive Boosting (AdaBoost), and Random Forest (RF) algorithms for the classification. Performance measures: accuracy, error rate, and response time are used to evaluate the proposed system. Our research findings promise that it can be implemented on patients diagnosed with the risk of a heart attack to monitor their heart health remotely and prevent sudden heart failure without impeding a person's everyday life. • This paper proposes a wearable smart and early Heart Attack diagnosis system. • It adopted a decentralized computational phenomenon using hybrid computing. • The proposed system can monitor and trigger the patient current heart status. • Efficient resource utilization is a key element of the proposed methodology. • Heart status classification based on required heart diagnosis parametric sensors. [ABSTRACT FROM AUTHOR]

Several types of sensors have been available in off-the-shelf mobile devices, including motion, magnetic, vision, acoustic, and location sensors. This paper focuses on the fusion of the data acquired from motion and magnetic sensors, i.e., accelerometer, gyroscope and magnetometer sensors, for the recognition of Activities of Daily Living (ADL). Based on pattern recognition techniques, the system developed in this study includes data acquisition, data processing, data fusion, and classification methods like Artificial Neural Networks (ANN). Multiple settings of the ANN were implemented and evaluated in which the best accuracy obtained, with Deep Neural Networks (DNN), was 89.51%. This novel approach applies L 2 regularization and normalization techniques on the sensors’ data proved it suitability and reliability for the ADL recognition. [ABSTRACT FROM AUTHOR]