Loading...
Dr. Kushal Pal Singh is an Assistant Professor in the Department of Mathematics at Maharaja Surajmal Brij University. His academic interests include applied mathematics, mathematical modeling, optimization techniques, data analytics, and the application of artificial intelligence in engineering and interdisciplinary research.
Deep learning has emerged as a revolutionary technology for logistics, transportation, and mobility solutions by facilitating intelligent automation, predictive analytics, and real-time decision-making inside intricate transportation ecosystems. Organisations can enhance traffic management, demand forecasting, route planning, fleet operations, autonomous driving, and public transportation services with sophisticated deep learning architectures, including CNNs, LSTMs, Transformers, and hybrid neural networks. The amalgamation of deep learning with IoT, cloud computing, edge intelligence, digital twins, and linked mobility platforms improves operational efficiency, safety, sustainability, and customer experience. Despite persistent challenges concerning data quality, computational complexity, cybersecurity, interpretability, and infrastructure, continuous advancements in explainable AI, federated learning, autonomous systems, and smart mobility technologies are propelling the evolution of next-generation transportation solutions. Future logistics and mobility ecosystems powered by deep learning will be essential for facilitating efficient supply chains, robust transportation networks, diminished environmental impact, and sustainable urban mobility.
Cardiovascular diseases (CVDs) continue to be a leading cause of illness and death globally, highlighting the urgent requirement for early identification, precise risk assessment, and tailored preventative measures. ML has surfaced as a viable method for discerning intricate connections among demographic traits, clinical data, lifestyle variables, laboratory findings, medical backgrounds, and physiological indicators. This chapter explores the underpinnings of machine learning-driven cardiovascular disease forecasting, pertinent data repositories, preprocessing techniques, feature selection, model formulation, assessment metrics, interpretable artificial intelligence, and clinical assimilation. It offers a suggested methodology for stratifying cardiovascular risk and addresses the difficulties related to class imbalance, absent data, model generalisability, interpretability, equity, privacy, and clinical validation. The chapter wraps up by outlining research avenues related to multimodal learning, continuous patient monitoring, wearable technology, federated learning, and individualised risk assessment. Well-validated machine learning systems can enhance clinical decision-making and preventative care; nonetheless, they ought to supplement not supplant expert judgement and recognised clinical protocols.
The swift expansion of digital banking, online payment systems, mobile financial applications, e-commerce, and fintech services has markedly amplified the number and intricacy of financial transactions. While digital financial services provide convenience and efficiency, they simultaneously create avenues for advanced fraudulent schemes, such as credit card fraud, identity theft, account takeover, payment manipulation, money laundering, phishing-related financial assaults, and unauthorised transactions. This chapter introduces a conceptual multimodal artificial intelligence framework for the identification of financial fraud and the analysis of transaction risk in real time. The chapter examines data collection, preprocessing, feature extraction, multimodal integration, fraud categorisation, risk assessment, and intelligent transaction decision-making. It additionally explores implementations in digital banking, credit card protection, mobile transactions, insurance fraud detection, e-commerce, and anti-money laundering frameworks.