MCKV Institute of Engineering

Lung cancer accounts for the highest number of deaths due to cancer in the world. Precise and prompt identification on the basis of histopathology pictures is vital in enhancing survival outcomes. This study proposes to create a fast and intelligible CNN with high accuracy and an easy-to-use CBQA model that can be deployed in the clinical setting. To do this, we created a bespoke model of CNN with four convolutional layers, group normalisation, dropout and SoftMax output. A small balanced dataset of 15,000 histopathology images (5000 adenocarcinoma, 5000 squamous cell carcinoma, 5000 benign) out of Kaggle dataset of Lung Cancer Histopathological Images was adopted with a 70–15–15

This correlative investigation presents a comparative performance analysis on machining characteristics by wire-cut electrical discharge machining (WEDM) outcomes when processing a titanium matrix composite (TMC) under diversifying input parameters. Experimental investigation focuses on metrics such as material removal rate (MRR), surface roughness (SR), kerf width (KW), and overcut (OC). An innovative multi-objective optimization (MOO) algorithm like desirable ant colony optimization (DACO) is proposed here, which is correlated with desirable particle swarm optimization (DPSO). DACO achieved a combined desirability score of 0.804, which rose to 0.813 with the implementation of DPSO. Comparing these two algorithms, DPSO outperformed DACO by approximately 1.119

High-dimensional gene expression data pose substantial challenges for predictive modeling due to noise, redundancy, and the curse of dimensionality. This work introduces a hybrid machine learning framework that integrates a Variational Autoencoder (VAE) for nonlinear dimensionality reduction with an Explainable Boosting Machine (EBM) for transparent multiclass cancer classification. The VAE compresses over 20,000 transcriptomic features into a 64-dimensional latent space while preserving biologically meaningful structure, yielding improved separability across tumor types. These latent representations are then classified using an EBM, enabling inherently interpretable predictions supported by global feature importance and SHAPbased local explanations. Experiments on the TCGA Pan-Cancer dataset demonstrate the effectiveness of the proposed pipeline, achieving 0.98 accuracy, 0.99 macro-precision, 0.97 macro-recall, and 0.98 macro-F1, with per-class F1-scores ranging from 0.95 to 1.00. The results highlight that the VAE-EBM framework provides both high predictive performance and clinical interpretability, offering a promising direction for precision oncology.

The escalating frequency and sophistication of cyber threats pose a critical challenge to individuals, organizations, and societies that have grown deeply dependent on digital technologies for communication, commerce, and daily operations. Despite this growing risk, a significant portion of everyday users remains unaware of the nature and scale of these threats, while existing literature continues to address attack vectors and their countermeasures in a fragmented and disjointed manner — motivating the need for a consolidated, rigorous review. Data has emerged as the most valuable asset of the digital age, underpinning organizational operations, personal identity, and economic activity, making it an increasingly attractive target for malicious actors who exploit both technical vulnerabilities and user complacency with growing precision. This review examines how major cyber-attacks manifest, traces their documented real-world impact across history, and evaluates the effectiveness of current technical and human-based mitigation strategies. In doing so, it provides a comprehensive, up-to-date synthesis that maps specific attack vectors to their historical implications while assessing the emerging role of AI-driven defense mechanisms. The findings aim to bridge the gap between theoretical understanding and practical cyber defense, offering the academic and practitioner community a unified analytical perspective that strengthens collective resilience against an increasingly hostile digital environment.

Content-based medical image retrieval enables searching large medical image databases to find visually similar reference cases, aiding tasks like computer-aided diagnosis, treatment planning, and medical research by utilizing prior data. Convolutional neural networks (CNNs) have shown promising results for medical image analysis, but challenging multi-class problems with limited data require enhancements. This work proposes a novel attention enhanced multi-stage residual networks architecture for extracting more complex visual features from medical images to improve image retrieval performance. The key objectives include designing stage-wise residual architectures to learn hierarchically more complex features optimized for similarity matching and integrating domain knowledge to focus learning on diagnostically relevant regions. Public datasets like chest radiography images (X-Rays) are used for feature extraction, evaluation, and image retrieval. The learned feature representations are evaluated using dimensionality reduction and visualization to assess their quality and diagnostic relevance. The extracted features are then integrated into the image retrieval system, enabling the retrieval of similar reference cases. Performance is assessed using standard retrieval metrics.