International Journal on Recent and Innovation Trends in Computing and Communication
Not a member yet
8613 research outputs found
Sort by
Optimal Denoising System for Medical Images Using Recurrent Neural Network and SVM
Full text linkImage denoising serves as a crucial preprocessing step in the realm of medical image analysis, with the primary objective of faithfully reconstructing the original image from its noisy counterpart. This process is essential for maintaining the integrity of vital details, such as edges and textures, within the denoised image. Innovatively addressing this challenge, our proposed system introduces a novel approach that seamlessly integrates Recurrent Neural Network (RNN) and Support Vector Machine (SVM). This powerful combination is adept at efficiently eliminating various types of noise, including gaussian, white noise, salt and pepper noise, and speckle noise, from intricate lung CT images. To enhance both learning accuracy and training efficiency, we have incorporated batch normalization in conjunction with residual learning. Notably, batch normalization is executed with the support of Long Short-Term Memory (LSTM). This strategic integration aids in the gradual separation of image structure from the noisy observations, a pivotal aspect in achieving optimal denoising outcomes. This approach not only enhances the accuracy of denoising but also contributes to reducing the overall training time, making it a valuable advancement in medical image preprocessing
Combined Attention-Based Fusion of Multiscale MRI Medical Images for Improving Early Brain Tumor Detection
Full text linkThe effective diagnosis of early-stage brain tumors relies heavily on the analysis of multimodal medical images. To address this need, we propose a novel multimodal medical image fusion approach that utilizes convolutional neural networks (CNNs) for enhanced feature extraction and representation. Unlike conventional CNN-based fusion methods that employ straightforward weighted averaging, our method incorporates a "Multiscale Attention Fusion Module" and a "Visual Relevance Fusion Strategy" to refine the fusion process. Our methodology aims to effectively combine multiple MRI modalities while emphasizing the most crucial diagnostic information, thereby mitigating the issue of non-essential information that often degrades the quality of fused images. By integrating these innovative components, our research contributes to improved early brain tumor detection, ultimately enhancing the quality and efficiency of medical diagnoses
An Efficient CBIR System for Medical Images Using Neural Network
Full text linkThis paper introduces an innovative Content-Based Image Retrieval (CBIR) system that has been specifically developed for medical databases. Its objective is to resolve the drawbacks of conventional keyword-based search approaches when considering the widespread digitization of medical illustrations, diagrams, and paintings. In contrast to conventional methods that rely on textual queries, CBIR systems effectively locate and retrieve relevant images by analyzing image content using computer vision and image processing techniques, as well as information retrieval and database methods.A key challenge in CBIR lies in bridging the semantic gap between high-level user queries, often expressed through example images, and the low-level features of images such as texture, shape, and objects. This paper explores techniques to mitigate this disparity, enhancing the system's ability to accurately interpret user queries and retrieve relevant images.
The proposed CBIR system operates within a medical database containing images of various human organs, including the brain, heart, hand, chest, spine, and shoulder, categorized into six distinct classes. By leveraging low-level image features such as texture and shape, extracted using methods like mean, variance, standard deviation, area, perimeter, circularity, and aspect ratio analysis, the system performs iterative searches to retrieve relevant images.Classification of retrieved images is accomplished using Artificial Neural Networks (ANN), which have demonstrated efficacy in medical image classification tasks based on imaging modalities and the presence of normal or abnormal conditions. Performance evaluation of the CBIR system is conducted using confusion matrices to calculate precision and recall, essential metrics for assessing retrieval accuracy.
By focusing on medical datasets and integrating advanced feature extraction and classification techniques, this CBIR system aims to significantly enhance image retrieval efficiency and accuracy, particularly in the context of medical applications where precise retrieval of relevant images is critical for diagnostic and research purposes.
 
Exploring Portable Multi-Modal Telehealth Solutions: A Development Approach
Full text linkIn the realm of medical healthcare, standard modules are typically utilized in the design of both healthcare units and diagnostic devices. Despite the similarity in operational modes across sensory devices, variations exist in data retrieval methods. Consequently, researchers often develop distinct devices tailored for specific diagnostics. Given that sensors commonly operate in either digital or analog modes, the development of a versatile device capable of supporting multiple sensor types is both feasible and desirable, particularly in resource-constrained settings. The key challenge in such device development lies in software implementation and sensor calibration, ensuring accurate calculation of sensor values. Body statistics encompass various parameters, some directly detected by sensors in real-time, while others require calculation based on standardized procedures such as time-based averaging or differential value analysis. To address these challenges, we propose a system designed to facilitate the execution of multiple functional algorithms on a single device, triggered as needed and based on demand. This research study elucidates the methodology for handling diverse processes on demand and delineates multiple operational procedures pertinent to healthcare devices.
Adoption of Online Home Services an Empirical Study of Consumer Behaviour in Ludhiana City
Full text linkThe advent of technology has revolutionized the service industry, particularly in the realm of home services. This study aims to investigate the nuanced relationship between demographic factors and the adoption of online home services by customers. Recognizing the increasing prevalence of digital platforms offering services ranging from home maintenance to professional consultations, understanding the demographic influences on consumer adoption becomes imperative for service providers and marketers. This research employs a mixed-methods approach, combining quantitative surveys and qualitative interviews, to gather comprehensive insights into the preferences and behaviors of consumers in the online home services sector. The study focuses on key demographic variables such as age, income, education, and geographic location to discern patterns and correlations that may influence the likelihood of customer adoption. The findings are expected to contribute to both academic literature and industry practices by shedding light on the factors that drive or inhibit the adoption of online home services among different demographic groups. Moreover, implications for marketing strategies and platform design may emerge from understanding how customer demographics intersect with their preferences, trust levels, and technological literacy. As the digital landscape evolves, this research aims to provide actionable insights for businesses aiming to tailor their services to diverse customer segments. The outcomes of this study will contribute to a deeper understanding of the socio-economic and cultural factors influencing the adoption of online home services, fostering a more informed and effective approach for businesses operating in this dynamic and rapidly growing sector
Designing AI-Enabled Products for Emerging Markets: Analytics-Driven Strategies Amid Data Scarcity
Full text linkAlthough there is much opportunity for AI-driven products in emerging markets, common machine learning techniques struggle with shortages of data, poor infrastructure and diversity in behavior. In this paper, we discuss using strategies based on analytics, like few-shot learning, producing synthetic data and federated learning to deal with these constraints. Using observations and case studies, we present practical guides for building local and ethical AI systems that are also resilient. We use fairness-aware federated systems and context expertise to provide both AI professionals and product managers with useful techniques for ethical progress in areas where resources are limited and the impact is high
Design and Implementation of Software Engineering Developed Process Models
Full text linkAn Analysis is done of the traditional software life cycle models that are used in the field and current software development practices. It then gives a more in-depth look at the traditional models of software evolution that are used a lot and are thought of as the best way to organize software engineering projects and technologies. There are so many things that go into making software that it's hard to think of a single process model that would work for all projects. This study, however, came up with a generalized model that could help companies make good software. A general goal for evaluation, which means to think about or think about how important it is, is shown. Examining current practices, confirming theories, exploring when the subject isn't well understood, and describing the current state of things are all part of the general evaluation goals. Evaluation helps predict the future and explain why things or sequences are taking place, so it is important to do it. It is important to know about both the software process and what the software does. With this evaluation, we can figure out how to evaluate it
A Survey on Using Machine Learning to Predict Diabetes Early on
Full text linkDiabetes is a category of metabolic disease caused by a prolonged high blood sugar level. It is sometimes referred to as a chronic disease. If accurate early prediction is achievable, it can considerably lower the risk factor and severity of diabetes. Combining data mining methods with machine learning, a subsection of artificial intelligence, offers promise in the field of prediction. Data is widely available in the healthcare industry, and in order to improve prognosis, diagnosis, therapy, medication development, and healthcare in general, information must be extracted from it. Based on the World Health Organisation's 2014 report, diabetes is a type of chronic disease with the fastest global growth rates. To illustrate the widely used techniques for early diabetes detection—which are based on cutting-edge technologies including machine learning, cloud computing, etc.—we have reviewed a few significant pieces of literature in this study. The findings suggested that artificial intelligence-based methods are more effective in the early detection of diabetes in patients. Here, we used the Random Forest model to conduct an experiment using a diabetes dataset. First, the dataset is resampled and then used to train and test the Random Forest model. On all performance criteria, the Random Forest attained values above 96%
Structural Properties of Low Energy Ion Beam Kr Irradiated Sb/Al Bilayers Deposited on Silicon Substrate.
Full text linkIn the present work, Sb(~50nm)/Al (~50nm) thin films were deposited successively on the silicon substrate by e-beam evaporation method under 2×10-5 mbar pressure. The Sb/Al bilayer was then irradiated with beam of 350 KeV Kr+1 with fluence 3×1016 ions/cm2. The sample was then characterised by XRD and Rutherford backscattering spectrometry (RBS). The XRD study reveals AlSb phase formation in Pristine sample. RBS also confirms mixing in Pristine sample
Comprehensive Review of State-of-the-Art Applications of Artificial Neural Networks in Predicting Concrete Compressive Strength
Full text linkConcrete compressive strength prediction is a crucial aspect in civil engineering, with applications ranging from structural design to quality control in construction projects. Traditional methods for predicting concrete compressive strength often rely on empirical formulas or physical testing, which may be limited in accuracy or efficiency. In recent years, Artificial Neural Networks (ANNs) have emerged as powerful tools for predicting concrete compressive strength due to their ability to capture complex nonlinear relationships in data. This paper provides a comprehensive review of the state-of-the-art applications of ANNs in predicting concrete compressive strength. It discusses various architectures, training techniques, input parameters, and datasets used in ANN models, as well as their performance compared to traditional methods. Additionally, challenges and future directions in the field are identified to guide further research efforts