Raghu Engineering College

Graphene oxide (GO) which is a derivative of graphene, has garnered considerable attention due to its remarkable mechanical, thermal and electrical properties. In this study, GO nanoparticles were synthesized using ball milling method with graphite powder as the precursor potassium permanganate as oxidizing agent for inducing structural modifications. The synthesized graphene oxide was characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD) to confirm the structure, morphology and its functional groups. Subsequently water-based graphene oxide nanofluids of distinct concentrations (1, 3 and 5

Human Action Recognition (HAR) is a significant research area in computer vision with applications in surveillance, healthcare, and human–computer interaction. This study proposes a hybrid deep learning framework that integrates the lightweight MobileNetV1 architecture with the scalable EfficientNetB3 network to achieve a balance between computational efficiency and recognition accuracy. The hybrid design leverages the complementary strengths of MobileNetV1 for efficient low-level feature extraction and EfficientNetB3 for enhanced high-level feature representation, enabling improved discriminative capability while maintaining reduced computational complexity. The model is evaluated on the UCF101 dataset, comprising 13,320 video clips across 101 action classes. A frame-based classification strategy with uniform sampling is adopted to reduce computational overhead, and data augmentation techniques such as rotation, shifting, shearing, and brightness adjustment are applied to improve generalization. Experimental results demonstrate that the proposed model achieves an accuracy of 89.18

Deep learning (DL) and machine learning (ML) approaches are currently being utilized to develop computer-aided diagnosis (CAD) systems that can help radiologists make accurate diagnosis decisions. This paper proposes an optimal ML model for detecting breast cancer from mammography images. Initially, samples are obtained from the Curated Breast Imaging Subset DDSM dataset. The Gaussian Bilateral Filter (GBF) filter is then used during pre-processing to remove noise and improve mammography image quality. The region of interest (ROI) is then calculated employing the Improved Skip progressive attention-based U-Net (ISPU-Net). Higher-level deep features are then extracted using the dilated convolution-based residual densely connected autoencoder (DcDAE) model. Furthermore, handcrafted features (texture, shape, and intensity) are extracted using GLCM and Histogram statistics. Next, deep and handcrafted features are concatenated into a single feature vector. Finally, the concatenated feature is fed into an optimized polynomial kernel Support Vector Machine with a Boost classifier (OP-SVBM) to precisely identify breast cancer. Finally, the model is optimized using the Genetic Beetle Optimization (GBO) method by tuning the parameters. However, the lack of interpretability in DL models hinders their adoption in clinical settings, where explainability is essential for gaining trust and acceptance from healthcare professionals. In this study, an explainable artificial intelligence (XAI) model named gradient-weighted class activation mapping (Grad-CAM) is used to address issues of reproducibility and interpretability of breast cancer images. The experimental results obtained were 99.34% accuracy, 99.15% precision, 99.49% recall, and 99.32% F1-score, which proved that the proposed approach is fast, reliable and good in classification.

ABSTRACT- The presented work proposes a methodology for the automated transition of a solar PV array integrated unified power quality conditioner (PV-UPQC) between standalone and grid-connected modes of operation. The system consists of a shunt and series active filters connected back-to-back with a common DC link, addressing the challenge of integrating power quality improvement with clean energy generation. The automated transition ensures continuous power supply to critical loads, even during grid unavailability. Key innovations include implementing this automated transition in the PV-UPQC system with minimal disturbance to local loads. The system's performance is validated through experimental evaluation under various dynamic conditions, such as automated transition, supply voltage variations, grid unavailability, changes in solar power generation, and load variations - scenarios commonly encountered in modern distribution networks. The results are verified using MATLAB/Simulink simulations. Index Terms—Power Quality, shunt compensator, series compensator, UPQC, Solar PV, MPPT.

MicroRNAs (miRNAs) are small noncoding RNAs that fine-tune gene expression by promoting mRNA degradation or translational repression. In cancer, miRNAs serve as key regulators of tumor progression, acting either as oncogenes or tumor suppressors. A central aspect of their function is the regulation of epithelial-mesenchymal transition (EMT), a process that enables epithelial cells to acquire mesenchymal features, enhancing motility, invasion, and metastatic potential. ZEB2, a transcription factor and primary EMT driver, represses epithelial markers such as E-cadherin while promoting mesenchymal traits. Its overexpression is closely associated with cancer stem cell-like properties, recurrence, and drug resistance. Multiple miRNAs directly target ZEB2 to modulate EMT and metastatic behavior. Notably, the miR-200 family (miR-200a, miR-200b, miR-200c, miR-141, and miR-429) acts as a crucial suppressor of EMT by binding ZEB2 mRNA, thereby maintaining epithelial identity. Conversely, high ZEB2 expression represses miR-200 levels, establishing a feedback loop that dictates the balance between epithelial and mesenchymal states. Other miRNAs, including miR-205, miR-206, and miR-637, also suppress ZEB2 expression, with their downregulation correlating with enhanced invasion, metastasis, and poor patient outcomes. Dysregulation of these miRNA-ZEB2 interactions shifts the cellular phenotype toward mesenchymal dominance, facilitating cancer cell migration, dissemination, and colonization at distant sites. This study aims to explore the molecular and cellular mechanisms underlying the interactions between ZEB2 and miRNAs, with a particular focus on their role in regulating EMT, migration, and metastasis in cancer. This understanding may guide the development of novel therapeutic strategies targeting the miRNA-ZEB2 axis to prevent tumor progression and improve patient outcomes.