154092 research outputs found
Sort by
Waves of well-being:An exploration of remote health monitoring across species using FMCW radar
Get PDFThe growing demand for advanced health monitoring systems in both human and veterinary medicine has highlighted significant shortcomings in traditional contact-based methods, such as discomfort, inconsistent usage, and stress in vulnerable populations. This thesis explores the potential of Frequency Modulated Continuous Wave (FMCW) radar as a non-invasive, accurate, and privacy-sensitive alternative for health monitoring, addressing these limitations. The research first identifies key health indicators essential for early diagnosis and effective management, including vital signs (heart rate and respiration) and behavioral patterns (activity and posture). Analyzing major health risks across humans, livestock, and pets establishes the need for continuous and reliable monitoring solutions. A comparative evaluation of sensing technologies—including wearable sensors, thermal imaging, and WiFi-based systems—demonstrates FMCW radar as the most promising option due to its high resolution, adaptability to different environments, and privacy advantages. The application of FMCW radar for human health monitoring focuses on advanced signal processing techniques and AI-driven methodologies. The thesis introduces novel algorithms for vital sign estimation, achieving high accuracy even in challenging conditions. Additionally, activity and posture recognition are explored using micro-Doppler signatures and point cloud data, with deep learning models—such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs)—enhancing classification performance. Building on these findings, the research extends FMCW radar applications to veterinary health monitoring, addressing challenges like species-specific physiology, random movement, and environmental factors. Experiments with calves, dogs, and cats refine signal processing pipelines and AI models to account for different motion dynamics. Results confirm the feasibility of FMCW radar in delivering accurate, non-invasive monitoring for animals, improving welfare and farm productivity. The thesis demonstrates FMCW radar’s versatility across human and veterinary applications, contributing advancements in signal processing and AI for real-time monitoring. By bridging biomedical engineering and veterinary science, this research establishes FMCW radar as a transformative tool for non-invasive health monitoring across species, enhancing accessibility, efficiency, and privacy in healthcare solutions
Automatic Robotic Ultrasound Scanning for Muscle Segmentation and Reconstruction
No full textUltrasound (US) imaging is widely utilized for medical diagnostics thanks to its non-invasiveness, real-time imaging capability [1], and cost-effectiveness, making it ideal for visualizing and localizing soft tissues. Recent advancements in US technology allow for real-time 3D model reconstruction, enabling clinicians to dynamically monitor complex anatomical structures, which is especially valuable for analyzing muscle behavior during movement [2,3]. This capability benefits athletes, trainers, and clinicians, and the development of automated scanning systems further supports personalized care through patient-specific model generation.In this study, an automatic robotic ultrasound scanning (ARUS) system was developed to perform real-time segmentation and reconstruction of muscle and bone tissues. The system integrates a robotic arm, a US imaging system, and a stereo camera to perform automatic ultrasound scanning and real-time muscle reconstruction.A custom calibration approach was employed to enhance 3D reconstruction accuracy and consistency [4]. The ultrasound images are transmitted in real-time via the UDP protocol from the Windows workstation to the main processing unit, which operates on an Ubuntu system and is responsible for data processing and robotic control. A hybrid control strategy was employed to ensure stable contact force, while visual servoing allowed real-time adjustments to the probe’s position to accommodate dynamic muscle contours and improve reconstruction quality. The ARUS system was evaluated using a customized muscle phantom.The evaluation was conducted from two aspects, i.e., image segmentation and 3D reconstruction. The segmentation model used was an attention-enhanced U-Net, achieving a high Dice coefficient of 0.84 and an Intersection over Union (IoU) score of 0.85, indicating robust segmentation performance. For reconstruction, the ARUS system yielded an average spatial error of 0.94 mm, with a root mean square error (RMSE) of 1.22 mm, when compared with MRI reference data.In this study, an automatic robotic ultrasound scanning (ARUS) system was developed for real-time segmentation and 3D reconstruction of muscle and bone tissues. Evaluation results demonstrated promising performance in both segmentation and reconstruction, indicating that the ARUS system holds significant potential for dynamic muscle monitoring and personalized diagnostic and therapeutic applications.<br/
2024 EACTS/EACTAIC/EBCP Guidelines on cardiopulmonary bypass in adult cardiac surgery
Get PDFClinical practice guidelines consolidate and evaluate all pertinent evidence on a specific topic available at the time of their formulation. The goal is to assist physicians in determining the most effective management strategies for patients with a particular condition. These guidelines assess the impact on patient outcomes and weigh the risk–benefit ratio of various diagnostic or therapeutic approaches. While not a replacement for textbooks, they provide supplementary information on topics relevant to current clinical practice and become an essential tool to support the decisions made by specialists in daily practice. Nonetheless, it is crucial to understand that these recommendations are intended to guide, not dictate, clinical practice, and should be adapted to each patient's unique needs. Clinical situations vary, presenting a diverse array of variables and circumstances. Thus, the guidelines are meant to inform, not replace, the clinical judgement of healthcare professionals, grounded in their professional knowledge, experience and comprehension of each patient's specific context. Moreover, these guidelines are not considered legally binding; the legal duties of healthcare professionals are defined by prevailing laws and regulations, and adherence to these guidelines does not modify such responsibilities. The European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Cardiothoracic Anaesthesiology and Intensive Care (EACTAIC) and the European Board of Cardiovascular Perfusion (EBCP) constituted a task force of professionals specializing in cardiopulmonary bypass (CPB) management. To ensure transparency and integrity, all task force members involved in the development and review of these guidelines submitted conflict of interest declarations, which were compiled into a single document available on the EACTS website (https://www.eacts.org/resources/clinical-guidelines). Any alterations to these declarations during the development process were promptly reported to the EACTS, EACTAIC and EBCP. Funding for this task force was provided exclusively by the EACTS, EACTAIC and EBCP, without involvement from the healthcare industry or other entities. Following this collaborative endeavour, the governing bodies of EACTS, EACTAIC and EBCP oversaw the formulation, refinement, and endorsement of these extensively revised guidelines. An external panel of experts thoroughly reviewed the initial draft, and their input guided subsequent amendments. After this detailed revision process, the final document was ratified by all task force experts and the leadership of the EACTS, EACTAIC and EBCP, enabling its publication in the European Journal of Cardio-Thoracic Surgery, the British Journal of Anaesthesia and Interdisciplinary CardioVascular and Thoracic Surgery. Endorsed by the EACTS, EACTAIC and EBCP, these guidelines represent the official standpoint on this subject. They demonstrate a dedication to continual enhancement, with routine updates planned to ensure that the guidelines remain current and valuable in the ever-progressing arena of clinical practice.[Figure</p
Waveform-Specific Performance of Deep Learning-Based Super-Resolution for Ultrasound Contrast Imaging
No full textResolving arterial flows is essential for understanding cardiovascular pathologies, improving diagnosis, and monitoring patient condition. Ultrasound contrast imaging uses microbubbles to enhance the scattering of the blood pool, allowing for real-time visualization of blood flow. Recent developments in vector flow imaging further expand the imaging capabilities of ultrasound by temporally resolving fast arterial flow. The next obstacle to overcome is the lack of spatial resolution. Super-resolved ultrasound images can be obtained by deconvolving radiofrequency (RF) signals before beamforming, breaking the link between resolution and pulse duration. Convolutional neural networks (CNNs) can be trained to locally estimate the deconvolution kernel and consequently super-localize the microbubbles directly within the RF signal. However, microbubble contrast is highly nonlinear, and the potential of CNNs in microbubble localization has not yet been fully exploited. Assessing deep learningbased deconvolution performance for non-trivial imaging pulses is therefore essential for successful translation to a practical setting, where the signal-to-noise ratio is limited, and transmission schemes should comply with safety guidelines. In this study, we train CNNs to deconvolve RF signals and localize the microbubbles driven by harmonic pulses, chirps, or delay-encoded pulse trains. Furthermore, we discuss potential hurdles for in-vitro and in-vivo super-resolution by presenting preliminary experimental results. We find that, whereas the CNNs can accurately localize microbubbles for all pulses, a short imaging pulse offers the best performance in noise-free conditions. However, chirps offer a comparable performance without noise, but are more robust to noise and outperform all other pulses in low-signal-to-noise ratio conditions.</p
Digital Development Dilemma:From Progress to Control
Get PDFWhile the move towards digital futures seems to be inevitable, there are concerning reports about discrimination, exclusion, injustice, repression, and bias backed up by the newest technologies. Many of these problems are portrayed as unintended outcomes, digital harm, political repression, or planning and design mistakes. This chapter takes a brief historical look at the conceptualisation of technology in the decades of development work and the faith in technological fixes for socio-political problems. It argues that without situating ICT4D programmes in their colonial, political, socio-cultural, and economic contexts, their complexities and their “outcomes” cannot be analysed. This chapter introduces the digital development dilemma as a concept describing the inherent dilemma carried in the core of digital development programmes: increasing efficiency, inclusion, and participation on the one hand and paving the way for digital repression, consolidation of exclusion, establishment of new forms of technological dependency, and complicating digital self-determination, on the other. The chapter also includes recent examples of state control and surveillance, the increasing engagement of Big Tech companies in digital development, and new colonial models of platform-based work. In doing so, it aims to scrutinise the neutrality and idealism of ICT4D programmes by highlighting the dilemma between efficiency, control, and dependency at the heart of such initiatives.</p
Developing Trustworthy Artificial Intelligence Models to Predict Vascular Disease Progression:the VASCUL-AID-RETRO Study Protocol
Get PDFIntroduction: Abdominal aortic aneurysms (AAAs) and peripheral artery disease (PAD) are two vascular diseases with a significant risk of major adverse cardiovascular events and mortality. A challenge in current disease management is the unpredictable disease progression in individual patients. The VASCUL-AID-RETRO study aims to develop trustworthy multimodal predictive artificial intelligence (AI) models for multiple tasks including risk stratification of disease progression and cardiovascular events in patients with AAA and PAD. Methods: The VASCUL-AID-RETRO study will collect data from 5000 AAA and 6000 PAD patients across multiple European centers of the VASCUL-AID consortium using electronic health records from 2015 to 2024. This retrospectively-collected data will be enriched with additional data from existing biobanks and registries. Multimodal data, including clinical records, radiological imaging, proteomics, and genomics, will be collected to develop AI models predicting disease progression and cardiovascular risks. This will be done while integrating the international ethics guidelines and legal standards for trustworthy AI, to ensure a socially-responsible data integration and analysis. Proposed Analyses: A consensus-based variable list of clinical parameters and core outcome set for both diseases will be developed through meetings with key opinion leaders. Blood, plasma, and tissue samples from existing biobanks will be analyzed for proteomic and genomic variations. AI models will be trained on segmented AAA and PAD artery geometries for estimation of hemodynamic parameters to quantify disease progression. Initially, risk prediction models will be developed for each modality separately, and subsequently, all data will be combined to be used as input to multimodal prediction models. During all processes, data security, data quality, and ethical guidelines and legal standards will be carefully considered. As a next step, the developed models will be further adjusted with prospective data and internally validated in a prospective cohort (VASCUL-AID-PRO study). Conclusion: The VASCUL-AID-RETRO study will utilize advanced AI techniques and integrate clinical, imaging, and multi-omics data to predict AAA and PAD progression and cardiovascular events. Clinical Trial Registration: The VASCUL-AID-RETRO study is registered at www.clinicaltrials.gov under the identification number NCT06206369. Clinical Impact: The VASCUL-AID-RETRO study aims to improve clinical practice of vascular surgery by developing artificial intelligence-driven multimodal predictive models for patients with abdominal aortic aneurysms or peripheral artery disease, enhancing personalized medicine. By integrating comprehensive data sets including clinical, imaging, and multi-omics data, these models have the potential to provide accurate risk stratification for disease progression and cardiovascular events. An innovation lies in the extensive European data set in combination with multimodal analyses approaches, which enables the development of advanced models to facilitate better understanding of disease mechanisms and progression. For clinicians, this means that more precise, individualized treatment plans can be established, ultimately aiming to improve patient outcomes.</p
Stress-strain analysis of single ultrasound-driven microbubbles for viscoelastic shell characterization
Get PDFMicrobubbles are of great interest both for ultrasound imaging and for ultrasound-assisted therapy due to their nonlinear scattering, which is enhanced by the viscoelastic shell. A full characterization of this nonlinear response is therefore crucial to fully exploit their potential. Current microbubble characterization techniques rely on assumptions regarding the microbubble shell rheology. Here, a stress-strain method is proposed to characterize the viscoelastic shells of single microbubbles with minimal underlying assumptions, which mainly entail separable viscous and elastic contributions. Detailed knowledge of the acoustic driving pressure and frequency, combined with a precise measurement of the bubble oscillations obtained through high-frequency ultrasound scattering, allows to derive the viscoelastic contribution of single microbubbles. To account for experimental uncertainties, we employed a fitting procedure of the surface tension in the buckled and ruptured regimes, which currently limits the applicability of the method to phospholipid-shelled microbubbles. The method was validated through simulations, and used to experimentally characterize 275 individual microbubbles from a monodisperse population, revealing a shell elasticity of (0.49 ± 0.10) N m−1, and initial surface tension of ( 28.7 ± 3.94 ) mN m-1. Besides providing detailed information on single bubble dynamics, this analysis paves the way for the characterization of the viscous dissipation mechanisms of individual microbubble shells.</p
Spectral Characterization and Discrimination of Sorghum (Sorghum bicolor (L.) Moench) Cultivars for Remote Sensing-Based Phenotyping and Selection
Get PDFRemote sensing has immense potential for high throughput phenotyping of agronomic and physiological traits that can be used in the selection of elite lines for productivity and adaptability in crop breeding programs. This is critical for important crops such as sorghum that are important for food security and livelihoods of millions of people in the world. The aim of this study was therefore to use remote sensing for spectral characterization and discrimination of 20 sorghum cultivars that are produced and grown in Southern Africa. Spectral reflectance indices (SRIs) were used in detecting variations in morphological traits (leaf length, leaf width, leaf area, plant height, number of leaves and chlorophyll) at vegetative and maturity growth stages. Our results showed that the morphological traits were all higher at maturity stage than at vegetative stage as expected but varied widely between the evaluated cultivars. Of the 29 indices, 12 and 19 of the evaluated vegetation indices showed significant differences between cultivars (p < 0.05) at vegetative and maturity stages, respectively. There was a significant relationship (p < 0.05) between morphological traits and SRIs, which were dominantly moderate at both stages. There were distinct differences between hybrids and open-pollinated (OPVs) varieties at maturity stage, with high reflectance observed from OPVs in the visible and near-infrared regions of the spectrum. Three principal components identified four indices with high discrimination ability among sorghum genotypes. We conclude that there is great potential for spectral data to be used in phenotyping and cultivar selection in sorghum crop improvement
Charting the landscape of rail human factors and automation:A systematic scoping review
Get PDFAs railway systems in Europe move towards increased integration and automation, understanding the human factors implications is critical. This systematic scoping review examines research on human factors and automation in railways, with a focus on studies involving railway operators such as train drivers and traffic controllers. Following PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) guidelines, we explored six databases and solicited expert recommendations, identifying 65 relevant studies published since 2000. Studies were categorized based on methodology and analysed to identify key themes, measures, and research priorities. The review revealed five main types of studies: empirical simulations (32%), non-simulation studies (25%), literature reviews (8%), analysis of existing technologies (31%), and new technologies (20%). Key research priorities included assessing the impact of automation on operator performance, workload, and situational awareness. Human-in-the-loop simulations emerged as a crucial method for evaluating new automated systems. Nevertheless, gaps emerged, e.g., studies focus mainly on drivers, use small sample sizes, and pay little attention to operators’ communications. Moreover, researchers seem to have scattered goals and assessment practices, with limited cross-contamination among different centres and across domains. If the goal is to integrate the European rail network, policymakers should push not only for technological integration but also for cultural and methodological integration, in which human factors can play a pivotal role.</p
Knowledge Silos as a Barrier to Responsible AI Practices in Journalism? Exploratory Evidence from Four Dutch News Organisations
Get PDFThe effective adoption of responsible AI practices in journalism requires a concerted effort to bridge different perspectives, including technological, editorial, and managerial. Among the many challenges that could impact information sharing around responsible AI inside news organisations are knowledge silos, where information is isolated within one part of the organisation and not easily shared with others. This study aims to study how knowledge silos might affect the adoption of responsible AI practices in journalism through a cross-case study of four Dutch media outlets. We examine individual and organisational barriers to AI knowledge sharing and the extent to which knowledge silos could impede the operationalisation of responsible AI initiatives inside these newsrooms. To address this question, we conducted 14 semi-structured interviews with a strategic sample of editors, managers, and journalists at de Telegraaf, de Volkskrant, NOS, and RTL Nederland. The interviews aimed to uncover insights into the existence of knowledge silos, their effects on responsible AI practice adoption, and the organisational practices influencing these dynamics. Our results emphasise the importance of creating better structures for sharing information on AI across all layers of news organisations and highlight the need for research on knowledge silos as an impediment to responsible AI production.</p