Opin vísindi

 

Nýlega bætt við

Verk
Advancing Human Soft Tissue Pathology Assessment Using Artificial Intelligence in Medical Imaging
(2024-12) Khatun, Zakia; Gargiulo, Paolo; Department of Engineering
Artificial Intelligence (AI) has revolutionized various fields by automating complex tasks, uncovering patterns in large datasets, and making accurate predictions. Machine learning, particularly deep learning, plays a pivotal role in enabling AI to emulate human intelligence for tasks such as image segmentation, classification, and recognition. In the realm of medical imaging, modalities like Magnetic Resonance Imaging (MRI), Computed Tomography (CT), and Ultrasound are indispensable tools for diagnosing pathologies, detecting abnormalities, guiding treatment plans, and monitoring disease progression. The integration of AI with medical imaging offers the potential to enhance the accuracy and efficiency of these processes. In particular, AI’s application in the assessment of tendon-related conditions, such as tendinopathy, presents a promising avenue for improving patient care. Tendinopathy can significantly impact a patient’s quality of life, and early detection is crucial to optimize treatment outcomes. This thesis focuses on the development of advanced AI-driven methods for analyzing human soft tissue pathologies, with a primary emphasis on tendon segmentation, pathology detection (classification), and tendon reflex response assessment. By automating the analysis of tendons and other human soft tissues, these methods aim to reduce human error and variability, thereby enabling more consistent and reliable clinical decisions. Ultimately, the goal is to support earlier, more accurate diagnoses and interventions, leading to better patient outcomes and more personalized treatment strategies. This thesis begins with a study that analyzes MRI and CT scans from 47 participants to investigate the relationships between the tendons, cartilage, and muscles in the knee. This study has two primary objectives: first, to predict knee cartilage degeneration, and second, to predict patellar tendinopathy. For both objectives, predictions are made using features extracted solely from the patellar tendon and quadriceps, rather than directly from the cartilage itself. This approach explores the potential of using features from surrounding tissues as indirect predictors of knee-related pathologies. This study demonstrates that both knee cartilage degeneration and patellar tendinopathy can be predicted using these features from adjacent structures, highlighting the importance of surrounding tissues as potential indicators of pathology. Traditional machine learning models are employed to identify the most relevant features for each prediction task, highlighting their importance in the diagnosis of these conditions. This foundational research deepens our understanding of the interrelationships between knee soft tissues, contributing to more accurate diagnostic approaches in musculoskeletal health and enhancing clinical decision-making and treatment strategies. A central focus of this thesis is the development of an end-to-end tendon segmentation module. This system integrates a superpixel-based coarse segmentation step that serves as a foundation for the final, more precise segmentation. In this approach, the segmentation task is framed as a superpixel classification problem. To achieve this, two distinct approaches are developed: (1) Random Forest (RF) and Support Vector Machine (SVM) classifiers for superpixel categorization, and (2) a Graph Convolutional Network (GCN) for transforming superpixels into graph structures for node classification. The RF and SVM classifiers demonstrate exceptional performance, achieving Area Under the Curve (AUC) scores of 0.992 and 0.987, respectively, with high sensitivity, indicating their effectiveness in accurately classifying superpixels. Although the GCN approach yields slightly lower performance, it showcases the potential of deep learning methods for improving segmentation by leveraging the structural relationships between superpixels. The findings suggest that both traditional machine learning and deep learning techniques offer promising avenues for advancing tendon segmentation, with superpixel-based methods offering a pathway to more reliable and automated segmentation in medical imaging. Another key component of this thesis is the development of an end-to-end tendon pathology detection module, utilizing the same MRI dataset. This module adopts a graph-based approach, where superpixels are treated as nodes and connected by edge relationships. Each MRI scan is transformed into a graph, with the task framed as a graph classification problem to determine the presence or absence of pathology. To achieve this, a Graph Echo State Network (GESN) is employed. Known for its ability to efficiently represent data without the need for iterative backpropagation, the GESN leverages both temporal and structural dependencies in the data, enhancing classification performance. In this study, the GESN outperforms traditional machine learning models, achieving a mean accuracy of 0.953 and sensitivity of 0.943. These results underscore the potential of the GESN to significantly enhance diagnostic accuracy, offering a powerful tool for early detection and clinical decision-making in tendon pathology assessment. Moreover, the GESN’s ability to handle complex, high-dimensional data suggests its broad applicability to other medical imaging tasks, further expanding its potential clinical utility. The final study of this thesis explores the impact of demographic factors, including age, height, weight, and gender, on reflex response times in healthy individuals. This analysis is based on electromyography (EMG) recordings from 40 participants. The results reveal that elderly individuals, particularly those who are taller, heavier, and male, exhibit delayed reflex onsets. Even after normalizing for height, older participants still demonstrate slower reflex responses. These findings highlight the role of demographic factors in neuromuscular reflexes, aiding in the diagnosis and early detection of related disorders. In conclusion, this research demonstrates the potential of AI, particularly superpixel-based and graph-based models, to advance tendon pathology assessment and exploratory tendon reflex studies, leading to better patient outcomes and musculoskeletal health management.
Verk
EEG-based investigation of cortical activity during Postural Control
(2022-01) Barollo, Fabio; Gargiulo, Paolo
The postural control system regulates the ability to maintain a stable upright stance and to react to changes in the external environment. Although once believed to be dominated by low-level reflexive mechanisms, mounting evidence has highlighted a prominent role of the cortex in this process. Nevertheless, the high-level cortical mechanisms involved in postural control are still largely unexplored. The aim of this thesis is to use electroencephalography, a widely used and noninvasive neuroimaging tool, to shed light on the cortical mechanisms which regulate postural control and allow balance to be preserved in the wake of external disruptions to one’s quiet stance. EEG activity has been initially analysed during a well-established postural task - a sequence of proprioceptive stimulations applied to the calf muscles to induce postural instability – traditionally used to examine the posturographic response. Preliminary results, obtained through a spectral power analysis of the data, highlighted an increased activation in several cortical areas, as well as different activation patterns in the two tested experimental conditions: open and closed eyes. An improved experimental protocol has then been developed, allowing a more advanced data analysis based on source reconstruction and brain network analysis techniques. Using this new approach, it was possible to characterise with greater detail the topological structure of cortical functional connections during the postural task, as well as to draw a connection between quantitative network metrics and measures of postural performance. Finally, with the integration of electromyography in the experimental protocol, we were able to gain new insights into the cortico-muscular interactions which direct the muscular response to a postural challenge. Overall, the findings presented in this thesis provide further evidence of the prominent role played by the cortex in postural control. They also prove how novel EEG-based brain network analysis techniques can be a valid tool in postural research and offer promising perspectives for the integration of quantitative cortical network metrics into clinical evaluation of postural impairment.
Verk
Do Asian Companies Bid Higher in Cross-Border M&A? A Moderating Effect Analysis
(2026-04-12) García-Gómez, Conrado Diego; Farinha, Jorge Bento; Demir, Ender; Díez-Esteban, José María; Department of Business and Economics
This study examines whether Asian companies pay higher premiums in cross-border mergers and acquisitions (M&A) and identifies the institutional factors driving this behavior. Grounded in the concept of Asian institutional logic—characterized by state coordination, relational governance, and long-term strategic orientation—we argue that these features shape distinctive acquisition patterns compared to Western market logics. Using a large sample of cross-border M&A during the period 2003–2021, we first uniquely compare whether the geographical origin of the acquirer firm is a relevant determinant of the premium paid, namely for cross-border operations targeting Asia, Europe, and the United States. We find that Asian acquirers pay significantly higher premiums compared to their European and U.S. counterparts. Employing a moderating effect approach, we find that this relationship is amplified by four mechanisms aligned with the Asian institutional logic. Specifically, we analyze the role of Chinese state-owned enterprises (SOEs) and find that they contribute significantly to the higher premiums paid. Our results are robust across different model specifications and subsample analyses, shedding light on the distinct dynamics of cross-border M&A involving Asian firms.
Verk
Distributed Energy Resource Management System (DERMS) Cybersecurity Scenarios, Trends, and Potential Technologies : A Review
(2025-01-30) Sugunaraj, Niroop; Ram Abayankar Balaji, Shree; Subash Chandar, Barathwaja; Rajagopalan, Prashanth; Kose, Utku; Charles Loper, David; Mahfuz, Tanzim; Chakraborty, Prabuddha; Ahmad, Seerin; Kim, Taesic; Apruzzese, Giovanni; Dubey, Anamika; Strezoski, Luka V.; Blakely, Benjamin; Ghosh, Subhojit; Jaya Bharata Reddy, Maddikara; Vardhan Padullaparti, Harsha; Ranganathan, Prakash; Department of Computer Science
Critical infrastructures like the power grid face increasing cyber threats due to the widespread integration of interconnected distributed energy resources (DERs). Compromised DER endpoints can trigger cascading outages, intentional device failures, and communication losses. To address these challenges, this paper reviews cybersecurity vulnerabilities in DER management systems (DERMS), including state-of-the-art attacks on data communication protocols, access control mechanisms, and identity management policies. Key threats such as false data injection, malware, denial of service, and network vulnerabilities are discussed. Additionally, realistic threat scenarios are outlined, followed by discussions on advanced security strategies such as the zero trust framework. The paper also introduces new architectural patterns aligned with the IEEE 1547.3 standard, offering a multi-level hierarchical framework for securing DERMS data and assets. Furthermore, it examines cybersecurity threats compromising confidentiality, integrity, availability, and accountability (CIAA) at various levels. This review uniquely consolidates existing research on DER cybersecurity and highlights necessary advancements, particularly intrusion diagnostic units (IDUs), to enhance future DERMS technologies. By ensuring compliance with IEEE 1547.3 standard requirements, this study contributes to improving DER cybersecurity resilience.
Verk
Digital health interventions for women in frontline public service roles : A systematic review of effectiveness in reducing substance use
(2026-01) Williamson, Grace; Khatun, Toslima; King, Kate; Simms, Amos; Dymond, Simon; Goodwin, Laura; Carr, Ewan; Fear, Nicola T.; Murphy, Dominic; Leightley, Daniel; Department of Psychology
Frontline occupations, including military, healthcare, and first responders, often include frequent exposure to traumatic events, increasing the risk of substance use disorders (SUDs). Research has shown that those in high-intensity occupations are at higher risk of developing SUDs compared to the general population. Women face unique experiences related to substance use, including greater functional impairment and barriers to treatment access. Yet, understanding of the effectiveness of digital health technologies in addressing substance use among women in frontline occupations is limited. This systematic review evaluates the effectiveness of digital health interventions in reducing substance use among women in frontline roles. Four databases (PsycINFO, Ovid MEDLINE, Embase, PsycArticles) were searched for English language full-text articles (2007–2024) that (1) evaluated a digital intervention designed to reduce substance use, (2) reported changes in substance use outcomes such as frequency, intensity or duration, using validated tools (3) included current or former frontline public service workers, and (4) included women as the primary target population or as a subgroup within the sample. 13 papers met inclusion criteria, focusing on eight distinct web and mobile-based interventions for alcohol, tobacco and illicit substances. Most studies (n=11) reported substantial post-intervention reductions in alcohol and tobacco use, although results for PTSD symptoms, illicit drug use, and quality of life were mixed. This review highlights the potential of digital health interventions for reducing substance use but underscores significant gaps in research. The scarcity of studies focused on women, small and heterogeneous samples, and focus on veterans limits the generalisability to women in frontline roles. These gaps present a pressing challenge in understanding gender-specific digital intervention efficacy. Future research should prioritise larger, representative samples of women across diverse frontline occupations to drive the development of digital technologies tailored to the unique challenges faced by women in these roles.

Flokkar í Opnum vísindum

Veldu flokk til að skoða.

Niðurstöður 1 - 9 af 9

Öll gögn í Opnum vísindum eru vernduð af ákvæðum höfundalaga og með öllum réttindum áskildum, nema annað sé tekið fram.