International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
Deep Learning Technique for Detecting and Analysing Ischemic Stroke Using MRI Images
Get PDFThe quantitative analysis of cerebral MRI images plays a pivotal role in stroke diagnosis and treatment. Deep learning, particularly CNNs, with their robust learning capabilities, offer an effective tool for lesion detection. To address the unique properties of stroke injuries and automate detection processes, we compiled a dataset of brain MRI images from various medical sources, representing patients affected by ischemic strokes. Different deep learning-based networks, including “Single Shot Multibox Detector (SSD)”, “Region-based CNN with ResNet101 (RCNN-ResNet101)”, “RCNN with VGG16 (RCNN- VGG16)”, and “YOLOV3”, were employed for automated lesion detection. The evaluation focused on achieving optimal precision in comparison to existing methods across Diffused Weight, Flair, and T1 modalities of MRI datasets. The developed technique involves extracting deep features during the encoding stage, followed by the minimization of features using fully connected layers. Significant handcrafted features, such as Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM), were incorporated alongside deep features. The concatenation of these features was implemented to maximize the dimension of the feature vector. This concatenated vector was then used to train and test the performance of various classifiers. Binary classification was employed to categorize brain images into normal or stroke affected. Initially, SoftMax was used as the default classifier. The performance of each classifier was individually evaluated, and the best-performing classifier was selected to confirm the overall effectiveness of the proposed technique. This all-encompassing strategy not only leverages deep learning for automatic lesion detection but also integrates handcrafted features and diverse classifiers to improve the precision and dependability of stroke detection across various brain MRI image modalities
Lesion-Based Detection of Cardiovascular Diseases Using Deep Learning and Red Deer Optimization
Get PDFNowadays, cardiovascular disease is a very concerning health issue in human life. Medical imaging through MRI plays an important role in the detection of many diseases. Magnetic resonance imaging (MRI) is a non-invasive and sophisticated diagnostic tool for cardiovascular disease (CVD) that allows for full visualization of the heart and blood vessels. Through Magnetic resonance imaging, we get high-quality images of blood vessels, which helps in detecting various types of heart-related diseases. With the help of MRI, we can detect various types of heart-related diseases. It also gives us information about their early diagnosis and their preventive measures. Deep learning and its advanced features are proving to be very helpful in this work. Deep learning has brought many new changes in this field. The article presents the Red Deer Optimizer with Deep Learning (ACVD-RDODL) algorithm for automated cardiovascular disease identification using magnetic resonance imaging (MRI). The primary goal of the proposed approach is to use Deep Learning models on cardiac MRI to detect Cardiovascular issues. The dynamic histogram equalisation (DHE) based noise removal model is used in the given approach to pre-process the images. Additionally, the Attention Based Convolutional Gated Recurrent Unit Network (ACGRU) model is used in this approach to classify Cardiovascular diseas
Deep Learning for Dense Interpretation of Video: Survey of Various Approach, Challenges, Datasets and Metrics
Get PDFVideo interpretation has garnered considerable attention in computer vision and natural language processing fields due to the rapid expansion of video data and the increasing demand for various applications such as intelligent video search, automated video subtitling, and assistance for visually impaired individuals. However, video interpretation presents greater challenges due to the inclusion of both temporal and spatial information within the video. While deep learning models for images, text, and audio have made significant progress, efforts have recently been focused on developing deep networks for video interpretation. A thorough evaluation of current research is necessary to provide insights for future endeavors, considering the myriad techniques, datasets, features, and evaluation criteria available in the video domain. This study offers a survey of recent advancements in deep learning for dense video interpretation, addressing various datasets and the challenges they present, as well as key features in video interpretation. Additionally, it provides a comprehensive overview of the latest deep learning models in video interpretation, which have been instrumental in activity identification and video description or captioning. The paper compares the performance of several deep learning models in this field based on specific metrics. Finally, the study summarizes future trends and directions in video interpretation
Using Processing Digital Image Methods for Documenting Tumorogenic Breast Disease Cells
Get PDFWe identified and isolated CD44-CD24/low Lineage tumor cells in eight of nine individuals. In mice, tumors might grow from 100 cells with these traits, but not from thousands with other phenotypes. On each serial passage, tumorigenic cells contributed to the tumor’s CD44-CD24/low Lineage tumor-causing cell as well as phenotypically variable no tumor-causing cell groupings. They are called tumorigenic or cancer-initiating cells because they constantly create malignancies, unlike additional cancer cell types. This work proposes a unique approach to identify breast asymmetry as well as tumorigenic cancer cells utilizing extremely efficient digital image processing methods that are not previously used in this research field. Chi-square tests and t-tests were used for categorical as well as continuous data. All p-values under 0.05 proved significant
Innovative Solutions for Agriculture: Sensor-Driven Soil Parameter Monitoring and Deep Learning in Potato Disease Detection
Get PDFThe primary obstacle facing modern agriculture is the lack of advanced technologies capable of efficiently and proactively identifying crop diseases, a gap that is most noticeable while the crop is at the key stem stage. Taking note of this difficulty, the suggested solution calls for the deliberate insertion of cutting-edge sensors at the root level straight into the soil. The objective of this integration is to offer a comprehensive and in-depth evaluation of crucial factors that are necessary for plant health, including temperature dynamics, moisture content, and nutrient levels of soil. While the temperature sensors serve a dual purpose by monitoring the external environment and evaluating the condition of mechanical assets vital to agricultural operations, the soil moisture and index sensors are essential for precisely determining irrigation needs and assessing soil nutrient levels. The project incorporates a cutting-edge Convolutional Neural Network (CNN) deep learning algorithm designed especially for the identification of potato leaf diseases, which represents a significant improvement to disease detection capabilities. This sophisticated algorithm improves the accuracy and efficiency of disease identification by using deep learning to analyze and comprehend complex patterns found in the leaf of the plant. This comprehensive initiative's main goal is to create a seamlessly integrated sensor system that can monitor crop health dynamically, provide real-time insights into critical soil characteristics, and use state-of-the-art CNN deep learning technology to detect potato leaf diseases in the agricultural landscape with extreme precision
Random Forest Algorithm for Real-Time Health Monitoring Throught Iot Data
Get PDFThe last decade made significant progress in the empire of orientation to the health monitoring systems after the invention of wearable devices, simplifying health monitoring on a daily base. Devices combining “Internet-of-Things” and “Machine learning” technologies provide a solution that is persistent, objective, and feasible for remote monitoring, thereby facilitating ambient assisted living. This study aims to utilize a Random Forest machine learning algorithm to address clinical issues after achieving results on ML computations implemented on a dataset. In the subsequent tests, certain data will be collected, e.g., vital signs and body temperature heart rate, blood pressure, etc, utilizing IoT implemented devices. Health tracker devices combined with a series of body sensors revolutionize the system of living and health care regarding patient activity. Smartwatches bring the sensation of being one of the principal devices that often provide information regarding the step counter, heart rate, and sleep pattern, which is also crucial. The combination of the intelligent system of SPO2, heart rate, and body temperature sensors is often integrated with smartwatches find application, collecting the data and transferring it to the cloud for further analysis achieved by ML algorithm and Random Forest Machine Learning algorithm utilization. The testing phase pursues the notion, aiming to identify the level of accuracy in clinical issue detection, which confirms the system demonstrated in the work is efficient for remote monitoring
Futuristic Advancements in AI for Knowledge Management Systems and Multi Model Based Agent
Get PDFIn recent years, especially between the years of 2012 and 2022, the use of Artificial Intelligence (AI) in the education industry has grown significantly. Some of the key areas are Task Automation, Personalized Learning, Smart Content Creation, Teaching The Teacher, Knowledge Repository, Portals, and Maps. AI has already been applied to many educational systems that help develop skills and facilitate easier exchange of knowledge. The Internet Of Things (IoT) based smart schools and collaborative e-learning platforms would be the next big thing, which can take the education sector to the next level, parallel to medical, automobile and other industries. AI can provide many benefits, such as efficiency, and personalization, as well as streamline both automation and admin tasks to allow easier and faster accessibility for knowledge seekers. AI Based Smart Knowledge Management System (KMS) and portal would provide better searching and mapping capabilities, to find the relevant assets for researchers/students, who would like to learn and leverage their knowledge in any domain. Some of the new AI based technologies include MLOps, AutoML, ExplainableAI and Transfer Learning. This research work introduces and explains related literature on these topics to help enable effective, scalable and automatic Knowledge repositories, Systems and K-Map populations
Efficacy of Reflective Questioning Instructional Strategy on Students’ Achievement in Gas Laws Contents of Secondary Schools Chemistry
Get PDFThis study investigated the efficacy of reflective questioning instructional strategy on students’ achievement in gas laws contents of secondary schools Chemistry. A pre-test-post-test, control group quasi-experimental research design was adopted in the study. The participants were 129 students (72 females and 57 males) of four intact classes sampled from two government owned secondary schools in Nigeria using multi-stage sampling procedure. Two intact classes each were assigned to experimental and control groups respectively. Data collection was done using a 25-item gas laws achievement test (GLAT). Data collected were analyzed using mean, standard deviation and analysis of covariance (ANCOVA). The result showed that there was a significant difference between the experimental and control groups indicating that reflective questioning instructional strategy enhances students’ achievement in gas laws. Also, there was no significant influence of gender on the mean achievement scores of students in gas laws and there was no significant interaction effect of questioning instructional strategies and gender on students’ achievement in gas laws. Chemistry teachers and pre-service teachers should be trained on how to adopt reflective questioning instructional strategy in the classroom instructions.
Comparing Two Inclusion Techniques in Timed Automata
Get PDFVerifying the correctness of real-time systems often involves checking language inclusion between timed automata. This problem determines if the language of a system implementation is a subset of the language specified by its design. While the general case is undecidable, recent advancements have proposed techniques for specific scenarios. This paper compares two such techniques: a zone-based semi-algorithm for non-Zeno runs and a time-bounded discretization approach. We analyze their strengths and weaknesses, highlighting cases where each method is advantageous. The comparison highlights the timed bounded discretized language approach's advantages in terms of guaranteed termination and lower memory usage
Nano-Phased Materials and Thin Film Heterostructures: A Pathway to High-Efficiency Solar Energy Conversion Technologies
Get PDFThe sun fuel conversion, solar photovoltaics, bio-catalysis, and solar water splitting are all things that are covered in this road map. Perovskites, organic photovoltaics, and dye-sensitized solar cells (DSSCs) are several components of this category. The distribution and bridging of storage via the use of direct and indirect storage systems is the cornerstone of energy management for this electricity. This will exhibit energy efficiency ahead of several criteria, including mobility and light weight, high energy storage capacity, cheap manufacturing cost, low temperature performance, and quick energy transfer. Consequently, this will demonstrate energy efficiency. When an announcement is made on an increase in capacity, it is common practice to just include the installation of the equipment. Although it may seem like the manufacturing line is functioning well, this does not always mean that it is. It is possible that the installation of the manufacturing line and the actual sale of solar cells will be delayed for a period of time due to the introduction of new technologies. A semiconductor is the fundamental component of dye-sensitized solar cells. This semiconductor is produced by a photoelectrochemical system that consists of an electrolyte and a dye-sensitized anode