The access to education, employment, health care, finances and civic engagement is increasingly mediated by digital technologies. However, instead of equalizing, digital systems tend to reproduce and increase the already existing social inequalities. This chapter offers a multidimensional and in-depth analysis of digital exclusion beyond traditional ideas of the digital divide, outlines inequalities by three dimensions: access, digital literacy and algorithmic representation. First, the chapter determines some of the existing inequities in physical and economic access to digital infrastructure, such as broadband connectivity, devices, and platform access, and identifies geographic, gendered, socioeconomic, and Global North-South differences. Second, it looks at the issues of digital illiteracy, which does not merely mean being technically literate but also includes the ability to be data-literate, algorithm-aware, and have the ability to democratically engage in the digital realm. .
Healthcare systems generate massive volumes of sensitive patient information and require the storage, sharing and access of the data in healthcare with maximum levels of reliability, integrity and confidentiality. Traditional forms of storage come with single points of failure, lack redundancy and are prone to cyber threats. This chapter introduces a new framework to exploit the power of approximate quantum cloning and blockchain technology in order to achieve secure and fault-tolerant data redundancy in healthcare. Approximate quantum cloning provides a mechanism for the replication of blocked quantum states with high fidelity which may be good for the creation of a secure backup and redundancy of medical data with a minimal loss of information. By combining these quantum methods with the use of some decentralized ledger system like block chain, the system has a tamper proof and unalterable storage and transparent access management. The synergy gives better data integrity, fault tolerance and availability and avoids the possibility of unauthorized changes. Furthermore, blockchain based consensus protocols and smart contracts are used in regulating access control, audit trails and compliance to healthcare regulations such as HIPAA and GDPR. The chapter also discusses the implementation strategies, architecture design, security analysis and shows how this hybrid approach solves some of the common problems that data redundancy faces like latency, cost and scalability. Real world applications in cases of Electronic Health Records (EHRs), telemedicine and clinical research data protection are discussed, together with possible challenges and limitations and future research directions. This work brings attention to the transformative potential that quantum cloning redundancy can have in combination with blockchain immutability in healthcare informatics, what it can do to revolutionize secure healthcare data ecosystem for next generation resilient healthcare data ecosystems.
The rapid introduction of Large Language Models (LLMs) into clinical practice offers both opportunities and threats that require the use of stringent evaluation, governance, and management, and constant control. In this chapter, the relationships between red-teaming and MLOps practices and pervasive monitoring of models get researchers closer to all the key building blocks of the safety, reliability, and compliance with regulations of medical LLM in practical healthcare facilities. Even before clinical implementation, red-teaming has probably been identified as a pre-emptive strategy that needs to be taken to draw attention to an asset of vulnerabilities, hallucinations, unsafe guidance and adversarial hazards in the behaviour of LLM. However, it is impossible to use red-teaming without formal operational systems.
The rapid advancement of smart textiles, flexible electronics, Edge Artificial Intelligence, and Tiny Machine Learningis transforming wearable healthcare systems from passive sensing platforms into intelligent, real-time health monitoring ecosystems. This chapter explores the integration of e-textiles, textile-based biosensors, conductive fabrics, and flexible electronics for continuous acquisition of physiological and environmental data. It examines the role of Edge AI in enabling low-latency, privacy-aware healthcare analytics and discusses how TinyML facilitates on-device intelligence within resource-constrained wearable platforms. The chapter highlights the importance of explainable artificial intelligence, uncertainty quantification, and trustworthy AI frameworks for improving transparency, reliability, and confidence-aware decision-making in healthcare applications. Additionally, privacy-preserving approaches, including federated learning, differential privacy, and secure collaborative analytics, are reviewed to address growing concerns regarding healthcare data security.
Ethical, scientific and translational limitations of animal testing have increased the global pursuit of high quality, human relevant, biomedical research, toxicological, and drug development alternatives. The combination of artificial intelligence (AI) and cell-based assays, organ-on-chip (OoC) technologies, and computational modelling, is transforming the paradigm of experimental research by promoting predictive, scalable, and ethically responsible research. The chapter gives a general but inclusionary review of AI-based substitutes of animal testing in reference to three pillars of change, including advanced cell-based systems, micro physiological organ-on-chip technology, and AI-enabled virtual bodies (in silico models).The chapter starts with the analysis of the historical dependence on animal models, regulatory/ethical forces behind replacing, minimizing, and refining (3Rs).
Application of Artificial Intelligence (AI) to the sphere of education has already offered the opportunities of a self-reliant and personalized process of learning, however, it does not cause certain critical issues of morality, privacy, and control. Being intelligent tutoring systems, chatbots, adaptive learning platforms become a part of the classroom and online world, the concerns regarding the safety, transparency, and effectiveness of the algorithm and the ethics of the developers of the algorithm creation become an increasingly more evident one. The AI agents will most likely accumulate and manipulate sensitive student information which must be strictly controlled by the policies in order to promote accountability, fairness and adherence to international privacy laws including GDPR and FERPA.