Search

A systematic review conducted by researchers from the School of Computer Science and Engineering at Vellore Institute of Technology in Chennai, India, explores the use of machine learning and deep learning prediction models for sepsis and neonatal sepsis. Sepsis and neonatal sepsis are significant challenges in global healthcare, as they can lead to life-threatening organ dysfunction. Early identification of sepsis is difficult, but the implementation of machine learning and deep learning technologies shows promise in predicting sepsis at an early stage and reducing mortality rates. The review analyzes various studies and concludes that prediction models can support researchers and clinicians in making better decisions and administering antibiotics earlier. [Extracted from the article]

A recent study conducted at the School of Computer Science and Engineering in Shaanxi, China, focused on the development of a medical dialogue generation system. The researchers aimed to improve the efficiency of clinical consultations and reduce the burden on doctors by creating a model that can automatically generate responses. The study utilized external knowledge and large-scale pre-training models to enhance the accuracy of the dialogue generation. Experimental data showed that the proposed model outperformed traditional methods in terms of response effectiveness and accuracy. For more information, the full research article can be accessed through the provided link. [Extracted from the article]

A recent study conducted by researchers at Beijing Jiaotong University in China explores the impact of teleworkers' psychology on their job performance and how they relieve negative emotions through enterprise social media (ESM). The study surveyed 835 teleworkers from various industries in China and found that family-to-work conflict and social isolation contribute to emotional exhaustion, which in turn negatively affects job performance. However, the study also found that the use of ESM can reduce family-to-work conflict and social isolation, thereby alleviating emotional exhaustion. The research provides practical suggestions for managers and teleworkers to improve job performance while working from home. [Extracted from the article]

A recent report from researchers at the Department of Computer Sciences and Engineering in Kerala, India discusses the importance of early detection and segmentation of brain tumors. The report highlights the use of medical imaging modalities, particularly Magnetic Resonance Imaging (MRI), for analyzing the internal structures of the brain. The study focuses on three approaches for tumor segmentation: classical methods, machine learning strategies, and deep learning techniques. The researchers conclude by discussing future developments and challenges in brain tumor segmentation. This research has been peer-reviewed and provides valuable insights for medical imaging applications. [Extracted from the article]

Researchers at the University of Padova in Italy have developed an Internet of Things (IoT) system that monitors the heart rate of an opera singer in real time and provides feedback. The system also adds effects to manipulate the sounds emitted by the singer's body during a self-healing practice. The system was designed in collaboration with an opera singer who suffered from a carotid aneurysm, and it allows the singer to rest and alternate between demanding and less demanding singing techniques during a performance. The research found that the system had a positive effect on the singer's practice and could have potential applications in different performative scenarios. [Extracted from the article]

Researchers from the School of Computer Science and Engineering in Tamil Nadu, India, have developed a hybrid classification model and game-theoretic approach to optimize patient recruitment in clinical trials. The model combines various machine learning algorithms to improve trial efficiency and patient outcomes. Additionally, the researchers propose a game-theoretic model involving patients, doctors, and research firms to optimize recruitment strategies and foster collaboration. This research has the potential to revolutionize recruitment dynamics in healthcare and lays the groundwork for future studies in this critical domain. [Extracted from the article]

A study conducted by researchers at the School of Computer Science and Engineering in Vellore, India, has developed a novel approach for detecting and classifying myocardial infarction (MI) in echocardiogram frames using an enhanced CNN algorithm and an ECV-3D network. The proposed method aims to improve the accuracy and efficiency of MI diagnosis by leveraging advanced deep learning techniques. Through extensive experimentation, the researchers achieved an accuracy of 97.05% in the detection and classification of MI, indicating the potential of their method to significantly impact the accurate diagnosis of MI and improve patient outcomes. The proposed method is also suitable for real-time clinical applications. [Extracted from the article]

A recent pilot study conducted by researchers at the School of Computer Science and Engineering in Tamil Nadu, India, explores the use of artificial intelligence (AI) to classify mental health disorders. The study utilizes advanced machine learning and deep learning techniques to analyze lifestyle and environmental factors that correlate with the prevalence of mental disorders. The researchers found that their AI models, particularly the convolutional neural network (CNN) architecture, accurately predicted mental health occurrences with a maximum accuracy of 99.79%. The study aims to assist healthcare professionals in diagnosing and treating neurological disorders promptly and provide insights for policymakers in formulating regulations on pollution and addiction. [Extracted from the article]

A study conducted by the College of Computer Science and Engineering explores the use of machine learning algorithms (MLAs) for the classification of sleep disorders. The study compares deep learning algorithms and conventional MLAs and proposes an optimized method for sleep disorder classification. The results show that the artificial neural network (ANN) algorithm achieved the highest classification accuracy of 92.92%. This research provides valuable insights into the potential of MLAs in improving the accuracy of sleep disorder classification, which can have a significant impact on human health and quality of life. [Extracted from the article]

Researchers from the National University of Computer and Emerging Sciences in Lahore, Pakistan have published a report on speech emotion recognition and its importance in enhancing human-computer interaction and personalized experiences. The researchers propose a novel approach using deep stride convolutional neural networks to accurately identify human emotional states from speech signals. Their model achieved a high accuracy of 95% on the RAVDESS dataset, outperforming the state of the art. This research has been peer-reviewed and may have applications in customer service and mental health. [Extracted from the article]

Researchers at the Beckman Institute for Advanced Science and Technology have developed an artificial intelligence (AI) model that accurately identifies tumors and diseases in medical images. The model is unique in that it provides a visual map to explain each diagnosis, allowing doctors to easily understand and verify the results. This new AI model aims to streamline the process of catching cancer and disease in its early stages and improve communication between doctors and patients. The researchers trained the model on three different disease diagnosis tasks and found that it performed comparably to existing AI systems. [Extracted from the article]

A recent study conducted by the School of Computer Science at Cairo University focused on the early detection of hepatocellular carcinoma (HCC), a type of liver cancer, in Egyptian patients. The study proposed a model that utilizes machine learning algorithms to examine circulating microRNAs (miRNAs) as diagnostic and predictive markers for HCC. The model consists of three phases: data preprocessing, feature selection using a new algorithm called Binary African Vulture Optimization Algorithm (BAVO), and classification with support vector machine and k-folds cross-validation. The results showed that miRNAs-483-5p, 21, and 155 may serve as potential biomarkers for the early detection and prognosis of HCC. [Extracted from the article]

A recent study conducted by the Department of Computer Sciences and Engineering in Secunderabad, India, has developed a new deep learning-based method called the Dynamic Hierarchical Attention for Improved Segmentation and Survival Prognosis (DHA-ISSP) model. This model combines a three-band 3D convolutional neural network (CNN) U-Net architecture with dynamic hierarchical attention mechanisms to accurately segment brain tumors and predict survival outcomes. The DHA-ISSP model achieved impressive performance metrics and outperformed 369 competing teams in the 2020 Multimodal Brain Tumor Segmentation Challenge. The research highlights the potential for clinical applications of this system in personalized treatment strategies for glioma patients. [Extracted from the article]

A new report discusses research findings on the CDC and FDA from investigators at Al Al-Bayt University in Mafraq, Jordan. The research focuses on image segmentation and proposes an improved version of the Arithmetic Optimization Algorithm called AOAa. The algorithm is designed to obtain optimal threshold values for color and gray images using objective functions such as Otsu and Kapur's entropy methods. The results of experiments conducted on benchmark images showed that the proposed method outperformed other established methods. This research has been peer-reviewed and more information can be obtained from the authors. [Extracted from the article]

A recent study conducted at Hunan University in Changsha, People's Republic of China, focused on predicting protein-protein interaction (PPI) binding sites, which are crucial for understanding protein functions, disease mechanisms, and drug design. The researchers proposed a new deep learning model called CLPPIS, which combines CNN and LSTM components to capture spatial and sequential features simultaneously. The model also incorporates a novel feature group and a batch-weighted loss function to address data imbalance. The study found that the CLPPIS model outperformed existing methods for predicting PPI sites. This research was supported by the National Natural Science Foundation of China. [Extracted from the article]

A recent study conducted at the National University of Computer and Emerging Sciences in Lahore, Pakistan, has proposed a novel diagnostic system for detecting diabetic retinopathy (DR). The system utilizes artificial intelligence (AI) and deep learning algorithms to consider all symptoms of DR, including exudates, vessels, and hemorrhages. The study also introduces a new retinal vessel extraction algorithm that improves the performance of the DR classification model. The research achieved an accuracy of 0.88, with 0.91 precision, 0.86 recall, and 0.88 F1 score. This study highlights the potential of AI-based diagnostic systems in improving the timely diagnosis and treatment of DR. [Extracted from the article]

Researchers from the Department of Computer Science have published a study on breast cancer diagnosis using deep learning. The study addresses the challenge of limited data availability by proposing a novel approach that utilizes deep convolutional generative adversarial networks (DCGANs) to produce synthetic mammograms that accurately reproduce patterns observed in real data. The proposed synthesis method was evaluated through experimentation and demonstrated its ability to accurately reproduce diverse viewpoints of the breast. Human experts visually inspected and validated the synthetic images, highlighting the effectiveness of the methodology in identifying outliers. The study provides valuable insights for improving breast cancer imaging and diagnosis. [Extracted from the article]

A report from the Faculty of Electrical Engineering and Computer Science in Suceava, Romania discusses the potential of virtual reality, augmented reality, and mixed reality technologies in the therapy of children with autism spectrum disorder (ASD). The study conducted a systematic literature review and found that there has been an increase in publications related to apps for children with autism, with a focus on mobile AR solutions for Android devices. The findings suggest that these technologies are appropriate for the therapy of children with ASD, but more extensive testing and comprehensive empirical evaluations are needed. The report emphasizes the need for multidisciplinary cooperation and an integrated research approach, as well as consideration of technology ethics. [Extracted from the article]

A study conducted by researchers from the University of Sao Paulo explores the use of electroencephalography (EEG) to evaluate user experience (UX) with chatbot interactions. The researchers found that EEG, which measures brain electrical activity, offers high resolution and data accuracy for analyzing user-computer interactions. The study identified seven studies that examined commercial EEG devices for UX assessment, with the most popular technique being event induced emotional stimulation. This involves participants passively perceiving emotional stimuli such as images, music, and videos to evoke specific emotions. The research provides valuable insights into the potential use of EEG in analyzing user experience with conversational systems. [Extracted from the article]

Researchers from the University of Science & Technology Oran Mohamed Boudiaf in Algeria have developed a bio-inspired spiking network model based on Hodgkin-Huxley neurons for human motion perception. The model mimics the mechanisms of direction selectivity found in the primary visual cortex and effectively replicates the directional selectivity operation of the visual cortex when presented with time-varying images. The model's motion maps can be used for action recognition in input videos and show competitive performance in motion segmentation tasks. This research has been peer-reviewed and published in Cognitive Neurodynamics. [Extracted from the article]

A study conducted in Hubei, People's Republic of China, explores the impact of psychological impatience on pedestrian behavior and evacuation dynamics. The researchers developed an evacuation model that considers both the self-growth and contagion of impatience among different groups of pedestrians. The results show that impatience contagion significantly affects evacuation efficiency, and pedestrian groups with higher impatience self-growth rates have lower evacuation efficiency. The study concludes that the proposed impatience model has the potential to accurately predict evacuation processes. This research has been peer-reviewed and provides valuable insights into understanding pedestrian behavior during evacuations. [Extracted from the article]

A report from Jagiellonian University discusses the efficiency and potential applications of silicon drift detectors (SDDs) in medical imaging. Compared to conventional detectors, SDDs offer advantages such as high efficiency in capturing X-rays, low electronic-noise levels, and high resolution for accurate diagnosis. The research focuses on evaluating the performance characteristics of SDDs, including linearity, stability, and sensitivity in detecting X-rays, highlighting their suitability for a wide range of applications. The study provides valuable insights into the capabilities and potential of SDDs in the field of medical imaging. [Extracted from the article]

A researcher from Universita degli Studi di Milano has published a new study on computer security, specifically focusing on ensuring cybersecurity for industrial networks. The study highlights the vulnerability of industrial control systems (ICSs) to security risks due to the increased adoption of the Internet Protocol (IP). The researcher proposes a protocol called ArpON, which effectively prevents ARP cache poisoning attempts in real time, without incurring high operational costs or requiring hardware features or cryptography functionality. Simulation results demonstrate the effectiveness of ArpON in avoiding ARP poisoning with minimal communication overhead. The full research article is available for free online. [Extracted from the article]

A recent study conducted by researchers at Annamalai University in Chidambaram, India, has developed a new method for the automated detection and segmentation of gastrointestinal tract tumors using CT scans. The researchers utilized a Bottleneck Feature-based U-Net model, which combines an encoding U-Net with a segmentation U-Net to compress input data and retain vital information for diagnosis. The model was tested using the comprehensive Kvasir dataset and demonstrated an accuracy rate of 98.64% and a specificity score of 99.71%. This research has the potential to advance diagnostic methodologies in clinical settings and reduce diagnosis time. [Extracted from the article]

A research study conducted by the National University of Computer and Emerging Sciences in Karachi, Pakistan, focuses on identifying key genes associated with Type 2 diabetes (T2D) and its complications. The study introduces a multi-layer Knowledge Graph called DbKB, which visualizes and explores the relationships between biological networks related to diabetes. By integrating genetic, molecular, and regulatory information, the researchers aim to understand the interactions among T2D candidate genes and identify diseased genes. The study concludes by constructing a Knowledge Graph that identifies interaction-driven sub-networks involving regulatory genes, functional genes, and protein-protein interactions. [Extracted from the article]

A recent study conducted by the Department of Computer Sciences and Engineering in Uttarakhand, India, explores the use of social media data to analyze mental health disorders. The researchers emphasize the importance of accurate diagnosis and early identification of mental health issues, as they can be life-threatening if left untreated. The study proposes a new system that involves preprocessing, feature extraction, and classification to analyze mental health disorders using social media data. The research concludes by verifying the efficacy of the proposed method. This study has been peer-reviewed and provides valuable insights into the potential of using social media for mental health analysis. [Extracted from the article]

A recent study from the National University of Computer and Emerging Sciences in Karachi, Pakistan, explores the potential of personalized medicine in cancer treatment. The researchers propose an intelligent multi-class classifier system that uses Natural Language Processing (NLP) techniques and Machine Learning algorithms to automatically classify clinically actionable genetic mutations in cancer patients. The results show that the proposed model outperforms previous research in terms of accuracy and precision scores, with an accuracy score of approximately 82%. The researchers believe that this system has the potential to revolutionize cancer treatment and lead to significant improvements in cancer therapy. [Extracted from the article]

Researchers from the Department of Computer Science and Engineering at Shri Jagdishprasad Jhabarmal Tibrewala University in Rajasthan, India, have conducted a review of automated models for detecting cardiac arrhythmias. The researchers highlight the importance of early detection and prevention of cardiac disorders, as they are becoming increasingly prevalent due to factors such as stress, anxiety, and depression. The review evaluates the performance of the models based on various parameters and discusses their limitations and potential for improvement. The study emphasizes the use of electrocardiograms as a cost-effective method for diagnosing cardiac arrhythmias. [Extracted from the article]

A study conducted by researchers at the University of Lincoln in the United Kingdom explores the use of deep learning and machine learning techniques for detecting lesions in chest X-rays. The researchers propose a dual attention supervised module called DualAttNet, which combines global and local lesion classification information to improve accuracy. The experimental results show that DualAttNet outperforms other models in terms of mean average precision and average precision at 50%. The research has been peer-reviewed and more technical details can be found on the University of Lincoln's GitHub page. [Extracted from the article]

A study conducted in Wenzhou, People's Republic of China, proposes the use of an explainable deep neural network called XAIRL to segment foot infrared imaging datasets. The study compares the results of XAIRL with other algorithms and finds that XAIRL performs the best in terms of accuracy, precision, recall, F1 score, intersection over union, Dice similarity coefficient, boundary displacement error, Hausdorff distance, and receiver operating characteristic. The researchers suggest that these machine learning-based algorithms can be used to analyze foot temperature and pressure distribution, which can help customize shoes for individual wearers and reduce the risk of foot injuries, especially for those with high blood pressure. [Extracted from the article]

A recent research study conducted by the Institute for Computer Science in Iasi, Romania, focuses on the transliteration of old Romanian books written in the Cyrillic script into the Latin script. The study addresses the challenges faced in automatically processing these documents and proposes a technology that utilizes AI and symbolic methods. The research is a work in progress and is based on data from the former DeLORo project. For more information, readers can refer to the publication "Bringing the old writings closer to us: Deep learning and symbolic methods in deciphering old Cyrillic Romanian documents" in the Memoirs of the Scientific Sections of the Romanian Academy. [Extracted from the article]

A recent report from the University of Sfax discusses the challenges of security in wireless body sensor networks (WBSNs) and the Internet of Medical Things (IoMT) in healthcare technology. The report proposes a solution called the Enhanced Random Forest Classifier with K-Means Clustering (ERF-KMC) algorithm, which combines the accuracy of the enhanced random forest classifier with the clustering capabilities of K-means. This algorithm enhances security in IoMT networks by detecting and categorizing attack messages, specifically distinguishing between man-in-the-middle (MITM) and distributed denial of service (DDoS) attacks. The ERF-KMC algorithm also improves network performance metrics and contributes to the reliability and security of IoMT systems. [Extracted from the article]

Researchers from the Institute of Computer Science in Iasi, Romania, have conducted experiments on iris image classification using texture features and SURF (Speeded Up Robust Features) keypoints. They developed a method called DENOL, which generates a fixed number of SURF keypoints for each image by adapting the Hessian threshold parameter. The experiments used two well-known datasets, UBIRIS and UPOL, and the results were comparable to those obtained using Masek implementation of Daugman's method. For more information, readers can refer to the research article published in Memoirs of the Scientific Sections of the Romanian Academy. [Extracted from the article]