Dimitris K. Iakovidis

In the era of artificial intelligence, machines are demonstrating an unprecedented capacity to learn from massive amounts of real-world data to perform human-like cognitive processes, enabling them to recognize environments, objects, and conditions and make critical decisions more accurately than ever. In the medical field, the potential to generate realistic, privacy-preserving, unbiased synthetic data can be the key to unlocking the potential of artificial intelligence in medicine and overcoming the current barriers such as data privacy concerns and high data curation costs. Advanced data-driven solutions could lead towards more robust clinical decision support systems and enhanced clinical training. This Perspective critically examines current and emerging advances in synthetic data generation, and highlights its anticipated transformational effect for early and efficient prevention, diagnosis and treatment of gastrointestinal diseases. Research challenges and directions are identified for leveraging the benefits of synthetic data as well as translating and adopting them in clinical workflows.

Since 1991, Measurement Science and Technology (MST) has awarded best paper prizes. The Editorial Board of this journal believe that such prizes are an opportunity to thank authors for submitting their work, and serve as an integral part of the ongoing quality review of the journal. The current breadth of topical areas that are covered by MST has made it advisable to expand the recognition of excellent publications. Hence, since 2005 the editorial board have presented ‘Outstanding Paper Awards’. This year awards were given in the areas of fluid mechanics, optical and laser-based techniques, precision measurement, sensors and sensing systems, measurement science, and biological, healthcare and life sciences.

Urban environments present accessibility challenges for users with visual impairments due to the presence of everyday obstacles and layout elements such as sidewalks, crossings, street furniture, parked vehicles, and buildings. Although a number of assistive navigation systems have been proposed over the years, they are typically coupled to specific algorithms and hardware and are often designed to operate in predefined or controlled areas, making them costly and difficult to maintain or adapt as new computer vision methods emerge. To address these limitations, we present a modular integrated edge-cloud architecture for uninterruptible assistive navigation. The proposed architecture is structured around three main components: an edge client that handles data acquisition and user interaction, an assistive navigation pipeline that performs the perception of the environment, risk assessment, and path planning using modular algorithmic components, and a cloud component that coordinates the navigation sessions, data management, and bidirectional audiovisual communication with remote users. The assistive navigation pipeline operates on a stream of imagery and geolocation data and supports execution both remotely and locally on the edge device, which enables uninterruptible navigation assistance even when network connectivity is limited. To this end, the proposed architecture aims to provide an answer to the need for cost-effective assistive navigation system, designed to operate even in urban city environments and remain operational under limited network connectivity. The architecture emphasizes modular system orchestration rather than tightly integrated algorithms and devices.

Cruise ships remain a popular vacation choice for millions of travelers annually. However, the on-board environment, characterized by high population density and close-contact settings in ventilated indoor spaces, creates ideal conditions for the spread of highly airborne transmissible respiratory diseases, such as SARS-COV-2. Thus, this study focuses on an effective framework for managing airborne transmission risks of both known and emerging infectious diseases, which are intensified by the growing capacity and density of modern passenger ships, while acknowledging that other mechanisms may also contribute to disease transmission. Hence, early risk assessment (RA) of disease transmissibility is essential for safeguarding public health and enhancing the passenger experience, preparing the cruise industry for future pandemics, and strengthening the overall resilience and safety of cruise operations. Fuzzy cognitive maps (FCMs) offer a valuable method for modeling complex data-driven systems in the context of risk assessment. However, traditional FCM construction is typically performed manually, which limits automation and hampers the accurate representation of uncertainty. We propose Intuitionistic Risk Assessment FCM (IRA-FCM), a novel framework that: (i) employs a hybrid learning framework that integrates unsupervised clustering to reveal hidden patterns within the data with supervised learning based on labeled instances to automatically learn the structure of an intuitionistic Fuzzy Cognitive Map (FCM); (ii) integrates multi-source sensor and system data with varying uncertainty levels; and (iii) models uncertainty via hesitancy, enabling more accurate, uncertainty-aware transmission risk assessment. Additionally, it estimates the confidence levels associated with various degrees of risk, serving as a comprehensive decision-support tool to ensure safety throughout the cruise duration by quantifying the uncertainty in its assessments. The effectiveness of the proposed framework is demonstrated on a dataset representing various airborne and close contact transmission scenarios of SARS-COV-2 within monitored areas, created using a validated agent-based model designed for airborne disease spread. The results validate its capacity to determine the transmission risk with an average accuracy of 88.6%. The applicability of the proposed framework extends beyond cruise ships, to assess the airborne disease transmission risk in ventilated indoor spaces with resident populations, such as apartment buildings, hotels and hospitals.