The Silicon Institute of Technology, Bhubaneswar (SIT-BBSR) is a NAAC Grade A grade engineering institution with NIRF 2021 rank of 163 in Bhubaneswar, the capital of Odisha, India. Established in 2001 as an affiliated college of the Utkal University, the college has been an affiliated college of Biju Patnaik University of Technology, since 2002. Students are admitted to the college through the Joint Entrance Examination (main) merit list of Government of India. The college is accredited by the National Board of Accreditation and was granted autonomy by University Grants Commission (UGC) in 2017 for a period of 10 years. In 2009, Silicon Institute of Technology, Sambalpur was established as its sister institution.
Affecting almost 10 million people globally, Parkinson’s disease (PD) is identified as the second most prevalent neuro-degenerative disease by 2025, and 25.2 million people are expected to live with PD worldwide. Current diagnostic techniques often fail to detect early, as symptoms usually start to show after significant neuronal loss. To address the limitations of unimodal diagnostic approaches, we propose a novel multimodal deep learning framework that integrates spiral kinematics, acoustic features, and neuroimaging data to achieve an early stage and accurate detection of Parkinson’s Disease. The framework captures motor dysfunction through spiral drawing tests, speech impairments through acoustic features, and neurodegenerative brain changes via neuroimaging. Dedicated deep learning models, CNNs for spiral image analysis, voice data processed through RNNs, and spatial-temporal features captured via 3D CNNs for MRI data processed each modality. Combining these modalities with an attention-based fusion approach improved diagnosis performance.
Healthcare 4.0 envisions intelligent hospital environments where continuous, non-invasive, and scalable patient monitoring is seamlessly integrated with advanced analytics and ultra-reliable communication. Traditional wearable cardiac monitors often suffer from performance degradation due to motion artifacts, electromagnetic interference (EMI) from nearby medical equipment, and limitations in monitoring multiple patients simultaneously. Fiber Bragg Grating (FBG) sensors, with their immunity to EMI, high multiplexing capacity, and mechanical sensitivity, offer an effective platform for distributed physiological signal acquisition. This paper presents a distributed FBG sensor network deployed across multiple patients and an edge-based Graph Neural Network (GNN) framework for robust cardiac health monitoring. The GNN leverages both temporal features of individual ballistocardiogram (BCG) signals and spatial correlations among patients to suppress artifacts and improve anomaly detection accuracy. A 5G architecture with Ultra-Reliable Low-Latency Communication (URLLC) slices is used for critical alert transmission, while enhanced Mobile Broadband (eMBB) handles periodic waveform uploads. Simulation results show that the proposed system achieves heart rate estimation with a Root Mean Square Error (RMSE) of 2.15 bpm, anomaly detection F1-score of 0.97, and end-to-end alert latency under 4 ms with 99.999
The sliding wear properties of aluminum toughened with nano alumina particulate in a metal matrix nanocomposite have been justified via this work. Non-contact cavitation procedure implemented for the exclusive fabrication of nanocomposite material. In this observation, an unlubricated Multiple Tribo Tester has been installed to recognize the wear properties of an aluminum-backed nanocomposite. Two loads, i.e., a constant load and a load varying stepwise while keeping speed and time constant, have been introduced to conduct the sliding wear test on the system. The very consequences of the above setup reveal that, with the corresponding rise in parameters, i.e., load and speed, the wear rate also rises. A Field Emission Scanning Electron Microscope has been introduced into the system to observe the impacted, wear-out surfaces of both specimens, i.e., unpolluted aluminum and MMNC, and determine the wear mechanism. To optimize wear rate, the Novel Remora Wear Reduction Model (RWRM) was proposed; it is based on the principle of remora optimization. The observation established that amongst the specimens, MMNC exhibits enhanced resistance to rubbing wear; therefore, abrasion and delamination are considered as profound wear mechanisms.
Vehicular ad hoc networks (VANETs), which allow cars and infrastructure to interact in real-time, have transformed intelligent transportation systems. Malicious nodes in the network, however, could send harmful communications that compromise the security and efficiency of the traffic. Current detection systems using fog computing to find rogue nodes, including F-RouND, have limitations including high computational costs, restricted anomaly detection criteria, and reliance on a single node (Paranjothi et al., 2020). In this regard, our unique solution, H-FRND, is exceptional. The enhanced system includes hierarchical fog computing, blockchain-based trust management, support for heterogeneous networks, and deep learning for multi-parameter anomaly detection to avoid these problems. Using thorough simulation in OMNET++ and SUMO, we show that H-FRND outperforms F-RouND by 37
Due to their intermittent nature, the increasing penetration of renewable energy sources (RESs) in microgrids (MGs) causes frequency instability and supply–demand imbalances. Although battery energy storage (BES) and vehicle-to-grid (V2G)-enabled electric vehicles (EVs) can support frequency regulation, existing control and optimization methods often exhibit limited performance under system uncertainties and nonlinearities. This study suggests a fuzzy tilt integral plus derivative (FTI plus D) controller to close this gap. The controller parameters are adjusted using a novel hybrid salp swarm algorithm (SSA) pattern search (hSSAPS) algorithm. Furthermore, when compared to traditional controllers, the FTI plus D controller improves frequency regulation performance. The effectiveness of the proposed approach is evaluated by comparing it with heuristic techniques such as the teaching learning based optimization (TLBO), SSA, hybrid artificial rabbit optimiser pattern search (hAROPS) algorithm, and hybrid dung beetley optimiser pattern search (hDBOPS) algorithm. The fitness value, mean, and standard deviation are used for comparison. Furthermore, the FTI plus D controller is benchmarked against the FPI plus D, PI plus D, and PID controllers. The results show that the proposed hSSAPS-based FTI plus D controller reduces the fitness value by 94%, 89.94%, 87.12% and 82.97%, respectively, when compared to PID, PI plus D, FOPID and FPI plus D controllers. Furthermore, a Lyapunov-based stability justification verifies the stability of the controller.
