بحوث سكوبس — م.د. مياس محمد مهدي عبد علي الجباوي

This article uses advanced signal processing and machine learning to improve medical contactless human sensing. Advanced medical algorithms enable multi-source sensor data collection and analysis in the suggested strategy. In the beginning, the system combines inputs from multiple sensors and dynamically adjusts them based on SNR. Adaptive Z-score algorithms identify anomalies and resist noise. Iterative adjustments then achieve perfect temporal synchronization of synchronized signals, ensuring data integrity for accurate medical analysis. Advanced feature extraction methods normalize and reduce dimensionality while also capturing temporal and spectral features to improve discrimination. A revised and cross-validated machine learning model uses extracted attributes to identify medical issues in real time. The method is better than others based on performance evaluation, which shows high accuracy (88.4%), precision (83.9%), recall (86.7%), F1-score (85.1%), and error metrics like MAE (3.2), RMSE (3.9), and Cohen's Kappa (0.80). These studies show its accuracy in medical data processing and classification, helping doctors make better judgments. © 2025 IEEE.

This paper describes a novel technique to employ sophisticated AI algorithms with hospital IoT devices to simplify diagnostics and follow patients in real time. Convolutional neural networks (CNN) and long short-term memory (LSTM) networks can handle healthcare images and sequential patient data. This allows for precise diagnosis and temporal connection monitoring. CNNs extract and aggregate significant medical characteristics using multi-layered convolutional processes, pooling layers, and activation functions. However, LSTM networks use long-term associations to detect patterns and issues in real time. The recommended solution outperforms current healthcare monitoring methods in data accuracy, reaction speed, scalability, energy efficiency, security, user happiness, system reliability, and cost-effectiveness. The approach might enhance healthcare by transmitting accurate, safe, and current information at the proper moment, enabling proactive treatment and smarter judgments. AI in healthcare IoT networks improves patient care and simplifies tasks for doctors and nurses. This makes AI vital to current healthcare. This research emphasizes the importance of AI and IoT in improving healthcare by combining emerging technologies. © 2025 IEEE.

High-frequency stock trend forecasting is a critical component of financial decision-making that requires accurate and rapid prediction capabilities. Traditional optimization and prediction models often fall short in capturing the non-linear, volatile, and time-sensitive nature of high-frequency financial data. Existing swarm-based and deep learning techniques suffer from slow convergence, local optima entrapment, and lack of adaptability to dynamic market conditions. To overcome these limitations, this paper proposes a novel Quantum-Inspired Swarm Forecasting Algorithm (QISFA) that integrates principles of quantum mechanics with particle swarm optimization (PSO) and deep learning to enhance forecasting performance. QISFA utilizes quantum-inspired behavior to improve population diversity and exploration capability while coupling with Long Short-Term Memory (LSTM)-based neural networks for capturing temporal dependencies in stock prices. The proposed method is tested on real high-frequency stock datasets and benchmarked against standard models. Experimental results demonstrate that QISFA significantly outperforms traditional PSO, LSTM, and hybrid models in terms of prediction accuracy, convergence speed, and robustness under varying market conditions. This approach presents a promising advancement for traders and analysts seeking intelligent tools for real-time market prediction. © 2025 IEEE.

This research proposes a merger of historical time series data with sentiment analysis for increased prediction of stock market dynamics using an LSTM network, which is a deep learning architecture. For the previous five years, we gathered the daily stock prices coupled with text mining based on news articles and social media to cover the numerical and psychological components underlying market movements. Extensive investigation reveals that using an emotion score boosts the predictive modeling accuracy to a bigger extent as compared to a baseline LSTM using price and volume only. Specifically, the model exhibited better sensitivity to abrupt variations in public mood, enabling it to predict short-term price fluctuations and possible market disruptions. Furthermore, our data reveal that stocks with higher traded volumes have a steadier trajectory pattern, whereas mid-cap and smaller companies are more vulnerable to sentiment. All in all, this suggests that integrating quantitative metrics with qualitative cues from the market can lead to improved predictions. Connecting the dots between numbers and human thinking can create space for more machines to control market risk and better portfolio optimization strategies. Our research thus adds to the huge body of literature on the potent combination of new deep learning, stock market prediction, time series, sentiment analysis, LSTM, market volatility, and prediction approaches. © 2025 IEEE.

Clustering is essential for discovering patterns in data, but traditional methods like DBSCAN face challenges with varying densities and overlapping clusters. This study presents an Enhanced Adaptive DBSCAN (ADBSCAN) algorithm that dynamically adjusts clustering parameters based on local density variations and integrates multiple validation metrics for robust performance evaluation. Dimensionality reduction techniques further improve effectiveness on high-dimensional data. Benchmarking against modern clustering algorithms across several complex datasets highlights the improved accuracy, efficiency, and practical utility of the proposed approach. Future studies should concentrate on enhancing adaptation mechanisms to better manage overlapping features and varying data density, enhancing the algorithmś resilience and practicality. A comprehensive sensitivity analysis and comparison of clustering performance in original feature space versus dimensionality-reduced space further underscore the algorithm’s adaptability. Copyright © 2025 International Academic Press

Increases in population and prosperity are linked to a worldwide rise in garbage. The “classification” and “recycling” of solid waste is a crucial tactic for dealing with the waste problem. This paper presents a new two-layer intelligent decision system for waste sorting based on fused features of Deep Learning (DL) models as well as a selection of an optimal deep Waste-Sorting Model (WSM) based on Multi-Criteria Decision Making (MCDM). A dataset comprising 1451 samples of images of waste, distributed across four classes – cardboard (403), glass (501), metal (410), and general trash (137), was used for sorting. This study proposes a Multi-Fused Decision Matrix (MFDM) based on identified fusion score level rules, evaluation criteria, and deep fused waste-sorting models. Five fusion rules used in the sorting process and the evaluation perspectives into the MFDM are sum, weighted sum, product, maximum, and minimum rules. Additionally, each of entropy and Visekriterijumska Optimizacija i Kompromisno Resenje in Serbian (VIKOR) methods was used for weighting selected criteria as well as ranking deep WSMs. The highest accuracy rate of 98% was scored by ResNet50-GoogleNet- Inception based on the minimum rule. However, under the same rule, an insufficient accuracy rate of sorting was presented by ResNet50-GoogleNet-Xception. Since Qi = 0 for Inception-Xception, the final output based on MCDM methods indicates that the fused Inception-Xception model outperforms the other fused deep WSMs, which achieved the lowest values of Qi. Thus, Inception-Xception was chosen as the best deep waste-sorting model based on images of waste, multiple evaluation criteria, and different fusion perspectives. The mean and standard deviation metrics were both used to validate the selection findings objectively. The suggested approach can aid urban decision-makers in prioritizing and choosing an Artificial Intelligence (AI)-optimized optimal sorting model. © 2024 The Authors

Artificial intelligence has many applications in various industries, including agriculture. It can help overcome challenges by providing efficient solutions, especially in the early stages of development. When working with tree leaves to identify the type of disease, diseases often show up through changes in leaf color. Therefore, it is crucial to improve the color brightness before using them in intelligent agricultural systems. Color improvement should achieve a balance where no new colors appear, as this could interfere with accurate identification and diagnosis of the disease. This is considered one of the challenges in this field. This work proposes an effective model for olive disease diagnosis, consisting of five modules: image enhancement, feature extraction, clustering, and deep neural network. In image enhancement, noise reduction, balanced colors, and CLAHE are applied to LAB color space channels to improve image quality and visual stimulus. In feature extraction, raw images of olive leaves are processed through triple convolutional layers, max pooling operations, and flattening in the CNN convolutional phase. The classification process starts by dividing the data into clusters based on density, followed by the use of a deep neural network. The proposed model was tested on over 3200 olive leaf images and compared with two deep learning algorithms (VGG16 and Alexnet). The results of accuracy and loss rate show that the proposed model achieves (98%, 0.193), while VGG16 and Alexnet reach (96%, 0.432) and (95%, 1.74), respectively. The proposed model demonstrates a robust and effective approach for olive disease diagnosis that combines image enhancement techniques and deep learning-based classification to achieve accurate and reliable results. © 2023 by the authors.

Mathematical cryptography applies special properties of integers and number theory in encrypting and decrypting messages in our world which is increasingly in demand for information security, especially for social media and multimedia. The objective of this work is to generalize El-Gamal crypto algorithm to be applied by using symbolic 2-plithogenic integers instead of classical integers. Also, we apply the novel algorithm to the encryption and the decryption of square 3 x 3 fuzzy matrices with rational entries, and fuzzy relations which can be represented as fuzzy matrices with rational entries. In addition, many examples with numerical data will be presented. © 2023, American Scientific Publishing Group (ASPG). All rights reserved.

Streaming data applications are common due to the advancement of technology to continuously capture or produce data, such as sensors for temperature, humidity and precipitation observations, social media or chatbots. These data applications receiving massive data in real-time requires an efficient algorithm and sufficient memory for analytics. Internet-of-Things (IoT) technologies embedded in a system requires a robust algorithm for clustering the streaming data to support decision making by analysing the historical sensor payloads. The MuDi-Stream algorithm, a density-based method, has emerged as one of the important methods for clustering data streams. The main issue with MuDi-Stream is the number of empty grids increased with the dimensional number or the increase of the streaming speed, making it less efficient when handling high-dimensional data. Furthermore, each point that came to a grid in the online phase will be saved, and with time, these points will consume larger memory space. To overcome these issues, we proposed an enhanced version of MuDi-Stream, coded as eMuDiS. Several benchmark datasets have been used in this study, and the performance of eMuDiS is compared to the state-of-the-art methods, including MuDi-Stream. The experimental results show that the proposed eMuDiS has better memory allocation performance than the MuDi-Stream. © 2022 Little Lion Scientific