Sri Shakthi Institute of Engineering and Technology

This study presents the development and evaluation of a novel eutectic phase change material (PCM) composite for enhanced thermal management in photovoltaic (PV) systems. The composite was formulated with 50 wt

Alzheimer's disease is a disease that causes cognitive impairment. There is no cure for Alzheimer's disease, but the progression can be slowed down or halted. It is possible to treat the symptoms of Alzheimer's disease through the use of different drugs or non-drugs, thereby improving the health of the patient. To treat alzheimer patients at different stages, various techniques have been employed. As a result, it is crucial to predict and classify the various stages of Alzheimer's disease. In conjunction with electronic health records, Machine Learning and Deep Learning algorithms are used to assist in disease detection. The Convolution Neural Network is one of the most effective methods for predicting and diagnosing diseases in the field of deep learning technology. In this paper, several convolution neural network techniques are discussed and their accuracy for the early diagnosis of Alzheimer's disease is compared. It is concluded that this work provides a better accuracy result for CNN algorithms (ResNet50, ResNet101, ResNet50V2, ResNet101V2, InceptionV3, and Inception-ResNetV2) for the prediction of alzheimer disease in different conditions.

Trace metal contamination in marine ecosystems poses risks to both ecological integrity and human health, yet baseline data from remote oceanic regions remain limited. This study presents the first comprehensive multi-tissue assessment of trace metal bioaccumulation in commercially important marine fishes from the Andaman Islands, India. Concentrations of eleven trace elements (Al, Cd, Cr, Co, Cu, Fe, Pb, Mn, Hg, Ni and Zn) were quantified in gills, liver, intestine and muscle of five fish species (Cephalopholis sonnerati, Epinephelus bleekeri, Auxis rochei, Rastrelliger kanagurta and Nemipterus japonicus) collected from major landing centres in South Andaman. Eight metals were consistently detected, while Hg, Cr and Co remained below detection limits across all tissues. Metal accumulation was strongly organ-specific, with liver acting as the primary reservoir for Cu, Fe and Cd, gills reflecting waterborne exposure to Mn and Al, and muscle exhibiting the lowest concentrations. Interspecies variation was limited, indicating broadly similar exposure pathways across trophic groups. Multivariate analyses confirmed clear separation of samples by tissue type rather than species identity. Human health risk assessment based on muscle tissue revealed Estimated Daily Intake and non-carcinogenic risk (HI < 1) values within acceptable limits for all species. However, Total Carcinogenic Risk values for Cd and Pb exceeded the USEPA benchmark, highlighting potential long-term risks for high-frequency consumers. Overall, the Andaman region exhibits a transitional contamination profile dominated by natural lithogenic inputs with emerging anthropogenic signals, underscoring the need for proactive multi-tissue monitoring and context-specific seafood risk communication.

This paper introduces a novel, intelligent wearable platform designed for the early prognostication of cerebrovascular accident (stroke) risk through continuous, multi-parametric physiological surveillance. The system is conceptualized to address the significant lag time between the onset of pathological physiological changes and clinical diagnosis, a primary factor contributing to the severe morbidity and mortality associated with stroke globally. It employs a synergistic integration of non-invasive biosensors to concurrently monitor an ensemble of biomarkers, including cardiac electrophysiology (ECG), heart rate variability (HRV), peripheral capillary oxygen saturation (SpO₂), and tri-axial kinematic data for postural analysis. At its core, an optimized embeddedsystem-on-chip (SoC) performs real-time digital signal processing—encompassing adaptive filtering, feature extraction, and spectral analysis—followed by the application of a lightweight anomaly detection algorithm to identify deviations indicative of ischemic or hemorrhagic precursors. The device's architecture prioritizes a low-power profile and ergonomic design to facilitate seamless, long-term deployment in unstructured, ambulatory settings. Preliminary validation involving controlled pilot studies demonstrates a high degree of accuracy in classifying pre-stroke events, underscoring the system's potential as a robust tool for pre- emptive healthcare. By enabling timely medical intervention, this technology aims to substantially reduce diagnostic latency, improve patient outcomes, and decrease the overall burden on healthcare infrastructure.

In modern electrical systems, power interruptions are a common issue, especially in residential and small industrial applications. To ensure uninterrupted power supply, backup sources such as diesel generators (DG) are widely used. However, manual switching between Electricity Board (EB) supply and DG supply can lead to delays, improper load management, and potential safety issues. This project proposes an Automatic EB–DG Changeover System with Load Priority Control. The system automatically detects EB power failure and switches the load to DG supply using relay-based control logic. Additionally, the system manages multiple loads by prioritizing them. Under normal EB supply, all loads (three bulbs) operate. During power failure, only essential loads (bulb 3) remain ON, while non-essential loads (bulb 1 and 2) are automatically turned OFF The system uses relay modules for switching, ensuring electrical isolation and safety. This design eliminates the need for manual intervention and prevents overload on the generator. The proposed system is cost-effective, reliable, and suitable for small-scale applications such as homes, shops, and laboratories.