جامعة المعرفة
المصدر المعتمد للمعلومات البحثية والمخرجات هو قاعدة بيانات سكوبس (Elsevier Scopus) مع تدقيق بيانات الباحثين من عمادة البحث العلمي.
AD

Ashit Kumar Dutta

اشيت كومار دوتا

— Computer Science

كلية الهندسة وعلوم الحاسببانتظار توثيق الانتساب2021–2026
220وثائق Scopus
4,399استشهاداً
27مؤشر h
203منها بانتماء الجامعة
إجمالي Scopus (كل الجهات)ملف الباحث في Scopus ↗آخر تحديث: 2026-11-09

كل وثائق Scopus للباحث

220
مقالوصول مفتوح202532استشهاد· Semantic Scholar

Multi-scale feature fusion of deep convolutional neural networks on cancerous tumor detection and classification using biomedical images

Scientific Reports · مجلد 15 (1) · 1105-1105

In the present scenario, cancerous tumours are common in humans due to major changes in nearby environments. Skin cancer is a considerable disease detected among people. This cancer is the uncontrolled evolution of atypical skin cells. It occurs when DNA injury to skin cells, or a genetic defect, leads to an increase quickly and establishes malignant tumors. However, in rare instances, many types of skin cancer occur from DNA changes tempted by infrared light affecting skin cells. This disease is a worldwide health problem, so an accurate and appropriate diagnosis is needed for efficient treatment. Current developments in medical technology, like smart recognition and analysis utilizing machine learning (ML) and deep learning (DL) techniques, have transformed the analysis and treatment of these conditions. These approaches will be highly effective for the recognition of skin cancer utilizing biomedical imaging. This study develops a Multi-scale Feature Fusion of Deep Convolutional Neural Networks on Cancerous Tumor Detection and Classification (MFFDCNN-CTDC) model. The main aim of the MFFDCNN-CTDC model is to detect and classify cancerous tumours using biomedical imaging. To eliminate unwanted noise, the MFFDCNN-CTDC method initially utilizes a sobel filter (SF) for the image preprocessing stage. For the segmentation process, Unet3+ is employed, providing precise localization of tumour regions. Next, the MFFDCNN-CTDC model incorporates multi-scale feature fusion by combining ResNet50 and EfficientNet architectures, capitalizing on their complementary strengths in feature extraction from varying depths and scales of the input images. The convolutional autoencoder (CAE) model is utilized for the classification method. Finally, the parameter tuning process is performed through a hybrid fireworks whale optimization algorithm (FWWOA) to enhance the classification performance of the CAE model. A wide range of experiments is performed to authorize the performance of the MFFDCNN-CTDC approach. The experimental validation of the MFFDCNN-CTDC approach exhibited a superior accuracy value of 98.78% and 99.02% over existing techniques under ISIC 2017 and HAM10000 datasets.

skin cancer detectionskin lesion classificationmulti-scale feature fusionconvolutional neural networks
مقالمفتوح (مستودع)202522استشهاد· Semantic Scholar

Sum Rate Maximization for 6G Beyond Diagonal RIS-Assisted Multi-Cell Transportation Systems

IEEE Transactions on Intelligent Transportation Systems · مجلد 26 (10) · 17601-17611

With the rapid evolution toward data-intensive applications and sustainable urban mobility, upcoming sixth-generation (6G) wireless networks must deliver enhanced coverage, high spectral efficiency, and energy optimization across densely populated areas. However, achieving these requirements poses significant challenges due to the dynamic nature of urban environments, high interference in multi-cell systems, and limitations in conventional passive beamforming technologies. To address these challenges, reconfigurable intelligent surface (RIS) is considered a highly promising approach for enabling and improving 6G wireless communications. This is because it has the ability to efficiently manipulate wireless channels at a lower cost. Considerable study has focused on the utilization of conventional diagonal RIS, in which each individual RIS component is linked to its own ground load but not interconnected with other elements. Nevertheless, the uncomplicated structure of classical RIS imposes restrictions on its ability to manipulate passive beamforming. In this study, we consider beyond diagonal RIS (BD-RIS) in the multi-cell transportation system, which goes beyond using diagonal phase shift matrices. In particular, we provide a new optimization framework to maximize the sum rate of BD-RIS assisted multi-cell transportation system by optimizing the power allocation of the base station and phase shift design of BD-RIS in each cell. We employ the block coordinate descent method to transform the original optimization problem and achieve a local optimal based on standard convex approaches. Numerical results demonstrate the benefits of adopting BD-RIS in multi-cell transportation systems compared to the classical RIS architecture.

beyond-diagonal reconfigurable intelligent surfacesum rate maximization6Gpassive beamforming
مقالوصول مغلق202517استشهاد· Semantic Scholar

Efficient Resource Management for NOMA- Enabled UAV Communications in 6G IRS-Assisted Vehicular Networks

IEEE Transactions on Intelligent Transportation Systems · مجلد 27 (2) · 2706-2715

Intelligent reconfigurable surfaces (IRS) have emerged as a promising technology to enhance wireless communications by dynamically controlling the propagation environment. Despite their potential, practical challenges such as effective integration with existing systems and efficient optimization remain critical. This paper investigates the sum capacity enhancement of NOMA-enabled uncrewed aerial vehicle (UAV) communications in vehicular networks assisted IRS. In urban environments where direct links from UAV to vehicles are often obstructed by buildings or other obstacles, the IRS plays a critical role in improving signal quality by reflecting signals toward vehicles. We consider a downlink NOMA transmission scenario, where the UAV serves multiple ground vehicles, and signals are delivered through both direct and IRS-assisted links. A joint optimization problem is formulated to maximize the sum capacity by simultaneously optimizing UAV power allocation and IRS passive beamforming while ensuring a minimum signal-to-interference plus noise ratio requirement for each vehicle. To address the non-convex nature and reduce the complexity of the optimization, we first transform the original problem using the first-order Taylor expansion method. Then, we employ a two-step solution based on the fixed-point iteration method for passive beamforming at the IRS and standard convex optimization for UAV power allocation. The proposed solution is compared with a benchmark scheme with direct UAV-to-vehicle communication without IRS assistance. Numerical results demonstrate that our proposed framework converges quickly and significantly outperforms the benchmarks in terms of system capacity.

reconfigurable intelligent surfacenomaUAV communicationvehicular networks
مقالوصول مفتوح202516استشهاد· Semantic Scholar

Integrating intuitionistic fuzzy and MCDM methods for sustainable energy management in smart factories

PLoS ONE · مجلد 20 (1) · e0315251-e0315251

Improving energy efficiency is crucial for smart factories that want to meet sustainability goals and operational excellence. This study introduces a novel decision-making framework to optimize energy efficiency in smart manufacturing environments, integrating Intuitionistic Fuzzy Sets (IFS) with Multi-Criteria Decision-Making (MCDM) techniques. The proposed approach addresses key challenges, including reducing carbon footprints, managing operating costs, and adhering to stringent environmental standards. Eight essential criteria are identified, such as the use of renewable energy, the efficiency of production, and the health and safety of workers, to evaluate energy performance. Using the entropy method for criterion weighting and the CRADIS technique for alternative ranking, we prioritize a range of energy-efficient solutions. The novelty of our approach lies in its comprehensive assessment of complex real-world energy management scenarios within smart factories, offering a robust and adaptable decision-support tool. Our empirical results, validated through sensitivity analysis, show that alternative 5 delivers the most significant improvement in energy efficiency. This study provides valuable information for industry practitioners seeking to transition to more sustainable production methods and supports the broader sustainability agenda.

intuitionistic fuzzy setsmulti-criteria decision makingsustainable energy managementsmart factory
مقالوصول مفتوح202511استشهاد· Semantic Scholar

Enhancing dynamic resource management in decentralized federated learning for collaborative edge Internet of Things

EURASIP Journal on Wireless Communications and Networking · مجلد 2025 (1)

Abstract Dynamic situations and applications are supported by the diverse devices and communication technologies that constitute the Internet of Things concept. Despite this, communication backlogs are common due to rising network demand and insufficient resource allocation. This study provides a way to fix resource allocation problems using Mutable Resource Allocation and Distributed Federated Learning. Inadequacies and backlogs in resources are identified at the edge of the network. As part of this procedure, edge devices are assigned to link resources and users after independently determining which resources cannot be allocated and which shortcomings are linked with them. Adapting to demand and learning suggestions, this allocation is flexible. By classifying resources as sufficient or inadequate, the learning suggestions help avoid backlogs. This enables edge devices to choose between allocation and response, which improves network flexibility by prioritizing inadequate resource allocation. Accordingly, the recommendation factor periodically affects modifications to the edge connection and its interaction with the Internet of Things platform. The suggestion is particularly strong for situations with changing backlogs to ensure that subsequent resource allocations align with preference-based learning. Claiming to improve connection, services, and resource allocation while decreasing backlogs and allocation times, this approach is a hot commodity.

dynamic resource allocationdecentralized federated learningedge computingInternet of Things
مقالوصول مفتوح202511استشهاد· Semantic Scholar

Big data-driven agriculture: a novel framework for resource management and sustainability

Cogent Food & Agriculture · مجلد 11 (1)

As the global population grows, urbanization depletes water resources and significantly reduces cropland available for agriculture. This study proposes a Big Data Analytics-Integrated Agriculture Resource Management Framework (BDA-ARMF) to optimize resource utilization and enhance farm sustainability. The integration of BDA in agriculture offers substantial advantages, including improved management of consumer demand, enhanced farm operations, sustainable food production and better alignment of supply with demand. The framework combines BDA with the Internet of Things and cloud computing to improve accuracy, intelligence and sustainability in agriculture. Efficient data-driven farming requires actionable insights to minimize resource waste and environmental contamination. The proposed model outperforms previous approaches, delivering significant improvements in water management (97.8%), prediction accuracy (97.6%), production efficiency (96.4%), resource consumption reduction (11.5%) and risk assessment enhancement (94.7%). The proposed framework reduces resource waste and mitigates environmental impact, enabling sustainable agricultural systems and efficient, data-driven farming practices.

big data analyticsagricultural resource managementsustainable agriculturewater management
مقالوصول مفتوح202510استشهاد· Semantic Scholar

Enhancing wearable sensor data analysis for patient health monitoring using allied data disparity technique and multi instance ensemble perceptron learning

Scientific Reports · مجلد 15 (1) · 29555-29555

Wearable Sensor (WS)-based monitoring systems detect minute patient movements/ demands and abnormalities through periodic sensing and imaging. Sensor data observed over different intervals is not constant or available based on operating sequences. Due to variations in data sequences, the analysis process becomes complex, resulting in less precise outputs. To address this problem, an Allied Data Disparity Technique (ADDT) is proposed in this article. This technique identifies the disparity in different monitoring sequences in coherence with the clinical and previous values. Based on the mean disparity, the data requirement for the WS sequence is decided. This decision uses multiple substituted and predicted values obtained from previous instances. Multi-Instance Ensemble Perceptron Learning is used in this decision process, where the substitution instances for clinical and previous outcomes are performed. The ensemble perceptron selects the maximum clinical value correlating sensor data to ensure high sequence prediction. The ensembles are updated based on the highest precision-based WS values for diagnosis. This diagnosis-focused coalition between clinical and predicted WS values is updated periodically for the new precision levels identified.

wearable sensorspatient health monitoringclinical data fusionmissing data imputation
مقالوصول مفتوح202510استشهاد· Semantic Scholar

Image classification-driven speech disorder detection using deep learning technique

SLAS TECHNOLOGY · مجلد 32 · 100261-100261

Speech disorders affect an individual's ability to generate sounds or utilize the voice appropriately. Neurological, developmental, physical, and trauma may cause speech disorders. Speech impairments influence communication, social interaction, education, and quality of life. Successful intervention entails early and precise diagnosis to allow for prompt treatment of these conditions. However, clinical examinations by speech-language pathologists are time-consuming, subjective, and demand an automated speech disorder detection (SDD) model. Mel-spectrogram images present a visual representation of multiple speech disorders. By classifying Mel-Spectrogram, various speech disorders can be identified. In this study, the authors proposed an image classification-based automated SDD model to classify Mel-Spectrograms to identify multiple speech disorders. Initially, Wavelet Transform (WT) hybridization technique was employed to generate Mel-Spectrogram using the voice samples. A feature extraction approach was developed using an enhanced LEVIT transformer. Finally, the extracted features were classified using an ensemble learning (EL) approach, containing CatBoost and XGBoost as base learners, and Extremely Randomized Tree as a meta learner. To reduce the computational resources, the authors used quantization-aware training (QAT). They employed Shapley Additive Explanations (SHAP) values to offer model interpretability. The proposed model was generalized using Voice ICar fEDerico II (VOICED) and LANNA datasets. The exceptional accuracy of 99.1 with limited parameters of 8.2 million demonstrated the significance of the proposed approach. The proposed model enhances speech disorder classification and offers novel prospects for building accessible, accurate, and efficient diagnostic tools. Researchers may integrate multimodal data to increase the model's use across languages and dialects, refining the proposed model for real-time clinical and telehealth deployment.

Mel spectrogramwavelet transformfeature extractionensemble learning
مقالوصول مفتوح20258استشهاد· Semantic Scholar

Heat recovery unit integrated with biomass gasification for producing hydrogen/power/heat using a novel cascaded ORC with biphenyl/diphenyl oxide mixture; ML optimization and economic evaluation

Case Studies in Thermal Engineering · مجلد 76 · 107397-107397

This work provides a detailed evaluation of a novel biomass-fueled multigeneration system, conceived to contribute to the growing emphasis on sustainable energy solutions. The architecture comprises a biomass gasifier, an innovative cascaded organic Rankine cycle (CORC) incorporating a high-temperature mixture in the top cycle, a proton exchange membrane electrolyzer (PEME), a Brayton cycle, and waste heat utilization units, all operating together to deliver electricity, hydrogen (H 2 ), and thermal output. A comprehensive thermodynamic modeling framework is established to evaluate the system’s performance across various operational scenarios. The framework emphasizes critical metrics, including exergy efficiency, levelized total emissions (LTE), and payback period (PP). These indicators ensure a holistic assessment of energy, exergy, economic, and environmental considerations. Parametric studies demonstrate that enhancements in biomass mass flow rate and combustion chamber temperature significantly increase power output and H 2 production while reducing the payback period, underscoring the system’s flexibility and economic feasibility. Furthermore, the study employs sophisticated machine learning optimization methods, combining artificial neural networks (ANNs) with genetic algorithms (GA), to determine optimal operating conditions with minimal computational effort and maximum efficiency. When evaluated at nominal parameters, the system records an exergy efficiency of 23.72%, achieves a PP of 5.61 years, and yields an LTE value of 0.34 ton/GJ. However, under optimized conditions, these values improve to 35.01%, 3.78 years, and 0.241 ton/GJ, respectively.

biomass gasificationhydrogen productionpolygenerationproton exchange membrane electrolyzer