CORA
Cork Open Research Archive (CORA) is UCC’s Open Access institutional repository which enables UCC researchers to make their research outputs freely available and accessible.
UCC Research Communities
Recent Submissions
A machine learning approach to 3D protein structure sonification
(Studio Musica Press, 2024-01) Ronan, Isabel; Mi, Yanlin; Yallapragada, Venkata Vamsi Bharadwaj; Ó Nuanáin, Cárthach; Tabirca, Sabin; Science Foundation Ireland; Munster Technological University
Proteins are intricate structures that can be analysed by biologists and presented to the public using visualisations. However, with an increase in the amount of readily available protein-related information, new forms of data representation are needed. Sonification offers multiple advantages when conveying large amounts of complex data to interested audiences. Previous attempts have been made to sonify protein data; these techniques mainly focus on using amino acid sequences and secondary structures. This paper proposes a novel protein sonification algorithm involving atomic coordinates, B-factors, and occupancies to investigate new ways of displaying 3D protein structure data. This study culminates in creating a cultural showcase involving some of nature's most significant molecular structures. Results of both musical analysis and the showcase indicate that protein sonification has the potential to act as a helpful outreach and engagement tool for biologists while also helping experts in the field glean new insights from complex data.
Fine-tuning generative pre-trained transformers for clinical dialogue summarization
(Institute of Electrical and Electronics Engineers (IEEE), 2025-01-17) Ronan, Isabel; Tabirca, Sabin; Science Foundation Ireland
Automated clinical dialogue summarization can help make health professional workflows more efficient. With the advent of large language models, machine learning can be used to provide accurate and efficient summarization tools. Generative Pre-Trained Transformers (GPT) have shown huge promise in this area. While larger GPT models, such as GPT-4, have been used, these models pose their own problems in terms of precision and expense. Fine-tuning smaller models can lead to more accurate results with less computational expense. In this paper, we fine-tune a GPT-3.5 model to summarize clinical dialogue. We use both default hyperparameters along with manual hyperparameters for comparison purposes. We also compare our default model to past work using ROUGE-1, ROUGE-2, ROUGE-L, and BERTScores. We find our model outperforms GPT-4 across all measures. As our fine-tuning process is based on the smaller GPT-3.5 model, we show that fine-tuning leads to more accurate and less expensive results. Informal human observation also reveals our notes to be of acceptable quality.
Bio-inspired internet of fish towards a climate resilient future
(2024) Cheng, Steven Matthew
This essay reports on the utilization of microwave, optical, and acoustic technology in establishing an underwater interconnected network, internet-of-fish (IoF), to provide comprehensive and real-time information on marine quality for climate monitoring and freshwater preservation. In particular, bio-inspired ocean explorers, capable of traversing different levels of the ocean from shallow to deep waters, are incorporated with communication front-ends to allow inter-explorer communication. Additionally, microwave sensors are discussed as complementary sensing to conventional sensors to provide a more comprehensive and cost-effective characterization of water quality. Moreover, wireless charging mechanisms and mmWave and microwave power beaming are also discussed to provide sustainable and efficient power. Data from these explorers are to be transmitted to nearby hovering unmanned aerial vehicles (UAVs) wherein cross-boundary communication methodology is discussed. With the well-established IoF, a deeper understanding of the earth and its climate is achieved leading to adaptation, mitigation, and prevention of further climate degradation.
Changing oral health trajectories: a 10-year cross-sectional comparison of 2 domiciliary dental care patient groups in nursing homes
(Elsevier Inc., 2025) Janssens, Lynn E.R.; Petrovic, Mirko; Allen, P. Finbarr; Colman, Roos; Janssens, Barbara E.
Objectives
Since 2010, Gerodent has been a comprehensive oral health care program including biannual domiciliary dental care in nursing homes in Flanders, Belgium. Previous research revealed poor oral health among nursing home residents attending the mobile clinic. The objective of this study was to evaluate potential changes in the oral health needs of patients receiving domiciliary dental care, thus providing insights for future cohorts.
Design
This study compares the oral health status and treatment needs of 2 cross-sectional samples.
Setting and Participants
First-time patients attending the mobile dental clinic in nursing homes, collected a decade apart.
Methods
Generalized estimating equations were used to compare both samples [sample 1 (S1): 2010–2012: n = 1226; sample 2 (S2): 2021–2023: n = 775]. Results were adjusted for age, sex, increased reimbursement for health care costs, number of medications and care dependency. Differences in oral health outcomes were decomposed in a component attributed to the explanatory variables and an unexplained component using the Blinder–Oaxaca decomposition analysis.
Results
There were statistically significant lower edentulism rates in S2 (28%) than S1 [42%; odds ratio (OR), 1.82; 95% CI, 1.34–2.47]. Among dentate patients, S2 had a significantly lower caries prevalence (S1: 70% vs S2: 53%; OR, 0.55; 95% CI, 0.44–0.69) and a significantly higher mean of filled teeth (S1: 1.5 vs S2: 2.8; rate ratio, 1.82; 95% CI, 1.58–2.09). In the dentate sample, 77% of S1 and 54% of S2 residents required extractions and/or restorations (OR, 0.72; 95% CI, 0.53–0.98). The decomposition analysis showed that the change in explanatory variables attributed little to the shift in dental status. The explained component accounted for only 13.5% of the total risk difference in edentulism (bias-corrected and accelerated 95% CI, −0.6% to 30.6%).
Conclusion and Implications
From 2021 to 2023, more dentate care home residents consulted Gerodent with more natural teeth per person, showing lower levels of untreated disease, maintaining dental team treatment demands comparable with 2010–2012. These findings underscore the continued urgency for structured and accessible dental service provision for nursing home residents.
The role of e-leadership on remote decision making: key learnings from the COVID-19 pandemic
(University College Cork, 2024) Alsqeah, Latifah; Adam, Frederic; Treacy, Stephen; Saudi Electronic University
This thesis investigates the role of e-leadership competencies in decision-making processes within private businesses during the COVID-19 pandemic. The pandemic forced a rapid transition to remote work, necessitating a deeper understanding of how leaders adjust to virtual environments and employ e-leadership practices. Despite the growing relevance of remote work, empirical studies on e-leadership remain scarce, particularly in the context of decision-making—a core leadership responsibility. This research addresses this gap by exploring the importance and impact of e-leadership competencies on remote decision-making and the value propositions these competencies present.
The primary objective of this study was to investigate how e-leadership competencies influenced decision-making among private-sector leaders in Saudi Arabia during the COVID-19 crisis. A qualitative field study was conducted through interviews with 19 leaders from various private sectors, including software development, logistics, oil mining, and training services. These interviews provided in-depth insights into how leaders faced the challenges of remote decision-making during the pandemic.
The findings indicated that leaders recognised two key e-leadership competencies, e-communication and e-technology, as crucial for facilitating remote decision-making. These skills enabled managers to maintain clear and consistent communication with their staff, utilising advanced technology to manage remote work successfully. Critical competencies such as e-trust, e-team, and e-change were also recognised as essential, highlighting their significance in building trust, managing teams, and supporting change in a remote environment. Interestingly, 70% of participants considered e-social competency crucial, yet not all leaders agreed, indicating differing viewpoints on the role of social skills in e-leadership.
Through this analysis, two theoretical models emerged from the findings:
i. A preliminary model of e-leadership competencies’ impact on remote decision-making processes.
ii. A preliminary model of the value propositions of remote decision-making.
These models identified seven key impacts influencing decision-making in a remote work environment, alongside five primary value propositions related to remote decision-making.
This research makes several significant contributions to the IS field and practice. Firstly, it empirically validates and extends Roman et al.'s (2019) e-competency framework by applying it to the context of remote decision-making during the COVID-19 pandemic. This application resulted in the development of two preliminary models that explain the impact of e-leadership competencies on decision-making processes in remote work environments. Secondly, the study fills a significant gap in the literature by identifying new value propositions associated with remote decision-making. Lastly, the research broadens the understanding of e-leadership in the context of private businesses, offering practical implications for managing remote work during disruptions and contributing valuable insights to the literature on business continuity and e-leadership.
This thesis provides a comprehensive analysis of how e-leadership competencies influence remote decision-making in a disrupted remote work environment. The findings highlight the essential role of e-competencies in navigating the challenges of remote work, offering both theoretical advancements and practical guidance for leaders facing similar situations in the future. As organisations continue to adapt to the evolving landscape of remote working, the insights from this study will prove valuable in understanding and enhancing the effectiveness of remote decision-making.