69238 research outputs found
Sort by
The perceived sound quality of different genres of music when streamed versus played free field to hearing aids
No full textAim: Technological advancements have changed the way individuals listen to music through
hearing aids. Music can be delivered in a free field to the hearing aid microphones or
wirelessly streamed. The current study has three aims - 1) to investigate the effect of the music
delivery mode on the quality of music through hearing aids, 2) to investigate the effect of the
vent condition (open versus closed) on the perceived quality of music delivered in the free field
and streamed, and 3) to examine whether the effects of music delivery mode and vent condition
on the quality ratings differ between genres.
Method: Participants (n = 21) with self-reported or diagnosed hearing loss were presented
with classical and pop music through the hearing aids via free-field delivery through
loudspeakers or wireless streaming. ActiveVentTM was used, and it was opened and closed for
each delivery mode. Quality ratings were obtained on clarity, fullness, brightness, loudness,
and total impression. Real-ear measurements were obtained for each test condition to correlate
the real-ear outputs with the quality ratings.
Results: Wireless music streaming resulted in significantly lower fullness, loudness and total
impression ratings compared to free-field presentation when the vent was closed. The closed
vent resulted in significantly lower fullness, loudness and total impression ratings than the
open vent when the music was streamed. Thus, closed-vent streaming resulted in the lowest
fullness, loudness and total impression ratings, and this was likely associated with significantly
lower real-ear outputs at 250 and 500 Hz. No genre effects on the quality ratings were observed
when the classical and pop samples were presented under the same hearing aid conditions.
Conclusion: The results showed that the delivery mode and the vent condition interact and
affect the perceived quality of music through hearing aids. The results also demonstrated the
importance of an extended low-frequency bandwidth and a greater preference for lowfrequency
amplification in music listening. Understanding how the quality of music is affected
by the delivery mode and the vent condition will be helpful for audiologists when making
hearing aid adjustments and recommendations
Advanced Façade Design and New Metal Envelopes Towards a Net-Zero-Carbon Future
No full textThe façade is an essential component of commercial and public build-ings and plays a fundamental role in their energy efficiency and on their carbon footprint. This research aimed to investigate the most recent trends in advanced façade designs that apply metal components to achieve high environmental per-formances and minimise environmental impacts. Goal of the research was to bet-ter understand current and future design strategies of metal facades for high-performance buildings. More than 100 precedents, built between 2015 and 2023, were examined to better understand the recent application of these materials and products in advanced façade designs. Parameters such as type of façade compo-nents, alloys, finishing, surface albedo, context etc., were taken into consideration. A literature review on environmental impacts of new metal facades and claddings was carried out to inform and validate the results of the survey on precedents. Results showed a higher incidence of aluminium over other materials. Claddings tend to be more lightweight, de-materialised, stratified and more permeable, also to reduce embodied carbon and improve acoustic performances. Solar shading is becoming a prevalent application and new smart materials offer zero-energy systems. Light albedo surfaces are preferred to reduce the urban heat island effect generated by these materials.https://www.sdsbe2024.com
Recovering the Matrix: Hyaluronidase Inhibition as a Potential Neuroprotective Strategy for Hypoxic-Ischaemic Brain Injury
No full textIntroduction: Perinatal hypoxia ischaemia remains an important contributor to neonatal brain injury and is associated with adverse neurodevelopmental outcomes, such as cerebral palsy. Established
evidence supports a role for hyaluronidase family of extracellular matrix remodelling enzymes in the
pathogenesis of evolving brain injury, including the development of seizures and injury to the white
and grey matter. The overall aim of this thesis was to examine the therapeutic potential of Sulfuretin, a
selective hyaluronidase inhibitor, after acute hypoxia ischaemia in chronically instrumented preterm
fetal sheep.
Methods: Fetal sheep at 0.7 gestation (day 103; term ~145 days) received sham asphyxia (n=6) or
asphyxia induced by umbilical cord occlusion for 25 minutes. Immediately after occlusion, fetuses
received either a continuous intracerebroventricular infusion of vehicle (n=8) or Sulfuretin (3.3mg;
n=6). Fetuses were continuously monitored until 7 days recovery, where ewes were euthanised and fetal
brains were recovered.
Results: Electrographically, Sulfuretin was associated with a significant reduction in the number of
seizures between 6 to 24 hours following asphyxia, with a trend for reduction in the total number of
seizures over the 72 hour recovery period. Sulfuretin was also associated with an earlier increase in
recovery of electrophysiological power and spectral edge frequency. Further analysis of specific
frequency bands revealed that Sulfuretin was associated with a faster recovery in the delta, theta, alpha,
beta and gamma frequency bands. Histologically, Sulfuretin was associated with a significant increase
in the number of total and mature oligodendrocytes within the temporal white matter and a significant
reduction in the density of microglia and astrocytes within both the parietal and temporal white matter.
Finally, Sulfuretin was associated with a limited recovery of neurons, with a significant increase only
observed within the CA1/2 region.
Conclusion: Overall, this thesis aimed to evaluate the therapeutic potential of targeting the
extracellular matrix with hyaluronidase inhibitor Sulfuretin following perinatal hypoxia ischaemic
injury in the preterm fetal sheep. Inhibition of hyaluronidase activity suggests therapeutic benefits in
reducing white matter, hippocampal injury and the recovery of electrophysiological power
Machine Learning in Stroke Lesion Segmentation and Recovery Forecasting: A Review
No full textIntroduction: Stroke remains a major cause of disability worldwide, and precise identification of stroke lesions is essential for prognosis and rehabilitation planning. Machine learning has emerged as a powerful tool for automating stroke lesion segmentation and outcome prediction; however, these tasks are often studied in isolation. The two strategies are inherently interdependent since segmentation provides lesion-based features that directly inform prediction models. Methods: This narrative review synthesises studies published between 2010 and 2024 on the application of machine learning in stroke lesion segmentation and recovery forecasting. A total of 23 relevant studies were reviewed, including 10 focused on lesion segmentation and 13 on recovery prediction. Results: Convolutional Neural Networks (CNNs), including architectures such as U-Net, have improved segmentation accuracy on the Anatomical Tracings of Lesions After Stroke (ATLAS) V2 dataset; however, dataset bias and inconsistent evaluation metrics limit comparability. Integrating imaging-derived lesion characteristics with clinical features improves predictive accuracy at a higher level. Furthermore, semi-supervised and self-supervised methods enhanced performance where annotated datasets are scarce. Discussion: The review highlights the interdependence between segmentation and outcome prediction. Reliable segmentation provides biologically meaningful features that underpin recovery forecasting, while prediction tasks validate the clinical relevance of segmentation outputs. This bidirectional relationship underlines the need for unified pipelines integrating lesion segmentation with outcome prediction. Future research can improve generalisability and foster clinically robust models by advancing semi-supervised and self-supervised learning, bridging the gap between automated image analysis and patient-centred prognosis. Conclusion: Accurate lesion segmentation and outcome prediction should be viewed not as separate goals but as mutually reinforcing components of a single pipeline. Progress in segmentation strengthens recovery forecasting, while predictive modelling emphasises the clinical importance of segmentation outputs. This interdependence provides a pathway for developing more effective, generalisable, and relevant AI-driven stroke care tools
The Body Capital: Precariat Contemporary Dancers' Meanings of Wellbeing in New York City
No full textRomanticized in popular culture since the late 18th century, the social concept of the starving artist and the Bohemian lifestyle in Western contexts remains polarized as a badge of honor, or foolish endeavor. Within the hierarchy of artists, dancers are widely acknowledged to be situated at the bottom in terms of economic and social capital. The 21st century model of neoliberal capitalism is further increasing the barrier to entry, cementing the performing arts as a profession only accessible to those situated in a privileged economic class.
This thesis examines the wellbeing of economically precarious contemporary concert dance artists in New York City. The study draws upon a quantitative questionnaire and nine in-depth interviews with self-identified precariat dance artists, applying a constructivist methodology through a capitalist critical lens. It allows emerging-mid career New York City based dance artists to construct their personal meanings of wellbeing, and envisions an artistic world in which the market economy does not dictate access to and career sustainability within the field of dance.
The narratives amongst the dancers demonstrates a collective need for conscious societal intervention divorced from profit, to sustain the lives of dance artists and ensure a rich and diverse artistic milieu of dancemakers and performers. Among the themes identified in the dancers’ visions of wellbeing are financial stability and predictability, the ability to rest and recover, access to healthcare both physical and mental, and a healthy work/life balance. Discussions of agency and the ability to advocate for oneself combined with the often fused identity of dancer and self, demonstrated across the board, the dedication to the artform beyond industry, and the willingness to make great sacrifices to remain in the dance world.
This thesis contributes to the field of dance studies and the sociology of mental health through its critical examination of dance in industry, its impacts on dance artist wellbeing, leading to critical ramifications to the field of dance in terms of increased disparities amongst working artists, burnout, disposability, and loss of creative ingenuity and diverse perspectives within the field. It gives voice to artists typically marginalized and punished for discussing topics of economic concern and will contribute to the body of knowledge surrounding the effects of capitalism on artistic professions in which the full value of its contribution to society cannot be measured by capitalist profitability benchmarks
Targeting Ethnic Minority Voters: How National and Labour Use Political Marketing to Target Chinese New Zealanders
No full textThis thesis investigates the utilisation of political marketing by parties in the targeting of ethnic minority voters, employing New Zealand as an empirical case study. It addresses a significant gap in the political marketing and political science literature, which has seldom explored ethnic targeting.
In recent decades, ethnic minority voters have gained prominence in elections of Western democracies, running parallel to the increasing ethnic diversity in numerous countries. The question of ethnic targeting, therefore, is important to understand electioneering in a practical sense. Moreover, this topic carries ethical and democratic implications. For a country to maintain a just and healthy democracy, it is vital that ethnic minority voters are part of politics and considered by political parties during elections.
This is particularly relevant for New Zealand, where 42.4% of the population identifies with at least one minority ethnicity. Chinese New Zealanders, as the fastest-growing segment of the population and the third-largest ethnic group, have been a focal point for major political parties since the 2000s, which presents an ideal opportunity for studying ethnic targeting.
Combining qualitative and quantitative methods, and leveraging data from the New Zealand case study, this thesis develops a market-oriented ethnic targeting model to conceptualise, identify, and evaluate how political parties effectively target ethnic minority voters. The model suggests that a party should adopt a market-oriented targeting approach, emphasising the importance of building relations with the community, understanding their needs and concerns, finding targeting products, and conducting effective outreach.
The empirical research of this thesis, which covers five elections over 13 years, highlights not only an increasing awareness of targeting ethnic voters in New Zealand politics but also identifies a link between a party's market-oriented targeting efforts and the level of support garnered from Chinese voters. This research provides insights into a hidden area of modern politics in New Zealand and presents a ground-breaking model for future research into ethnic targeting
Modelling procalcitonin and other biomarkers for infection and sepsis in neonates
No full textC-reactive protein and procalcitonin are biomarkers which are elevated during infection. An increase after birth and surgery limits their use for the diagnosis of infection. Pharmacometric models were used to describe the time course of both biomarkers following these events in non-infected individuals, quantifying the variability in parameter estimates and concentrations. These models were extended to describe the biomarkers following infection diagnosis.
Procalcitonin (n = 2596) and c-reactive protein (n = 2848) concentrations from 544 and 632 neonates, infants and adults (from birth day to 86 years) with and without infection were collated from six different studies. A method to impute missing observation times based on standard sampling practice at the study hospital was developed. Turnover models were used to describe the time course of biomarkers, with stimulation of biomarker production via birth and surgical events. Postmenstrual age and body size were key covariates. C-reactive protein production rate (0.0433 mg/h/70kg [95% CI: 0.0389 – 0.0448 mg/h/70kg]) and clearance (0.0979 L/h/70kg [95% CI: 0.0923 – 0.101 L/h/70kg]), and the procalcitonin production rate (0.108 μg/h/70kg [95% CI: 0.101 – 0.119 μg/h/70kg]) and clearance (1.25 L/h/70kg [95% CI: 1.19 – 1.59 L/h/70kg]) were estimated in the non-infected state.
These biomarker models were used to generate patient-specific reference ranges of expected concentrations without infection for neonates, infants and adults. A proof-of-concept tool allowing users to generate reference ranges for any combination of patient factors (age, birth-related covariates, weight and time after surgery) was developed. A method by which these reference ranges could support decision-making during infection diagnosis is discussed.
This work applied pharmacometrics methodology to model biomarker concentrations in neonates to adults, following birth and surgery and in non-infected and infected states. This advances the field beyond c-reactive protein and procalcitonin patient sub-group reference ranges and cut-off values. The model-based tool predicts reference ranges in non-infected patients that could be applied to the clinical setting, where the benefits of this work will be the greatest
Holding Two Truths: How Counsellors Understand and Navigate Sexual Abuse Survivors' Ambivalence Towards Their Perpetrators
No full textThe experience of sexual abuse survivors feeling ambivalent or conflicted towards their perpetrators is an under-researched phenomenon. Even less is known about how counsellors understand and navigate this ambivalence in their therapeutic practice. This study explores these dynamics through six semi-structured interviews with sexual abuse counsellors in Aotearoa, New Zealand. The central research questions guiding this thesis are: “How do counsellors understand and interpret client ambivalence towards their perpetrators of sexual abuse?” and “How do counsellors navigate their therapeutic work with clients who have ambivalent feelings towards their abusers in cases of sexual abuse?” Reflexive thematic analysis of the data generated five key themes: The Nuanced Dynamics and Contributing Factors to Ambivalence, Internal Conflicts and the Emotional Toll for Clients, Counsellor Beliefs and Responses, Practice Wisdoms: Understanding and Shifting Ambivalence, and The Broader Contexts of Ambivalence. The findings suggest that counsellors perceive ambivalence towards perpetrators of sexual abuse to be a natural and relatively common experience for survivors, often rooted in prior relationships with perpetrators on whom survivors were, in some way, dependent. Counsellors can support clients by facilitating safe, non-judgmental exploration of these feelings and providing psychoeducation on the possible underlying reasons for ambivalence. This research contributes to a greater understanding of ambivalence in counselling practice, highlighting the importance of sensitivity, counsellor self-reflection, and a non-pathologizing approach
Integrating Key-point Information for American Sign Language Translation
No full textOver the past few years, research into Gloss Free Sign Language Translation
(GFSLT) has seen consistent breakthroughs, with state-of-the-art performance
repeatedly surpassed on standard benchmark datasets. However, the potential
for incorporating key point information to improve performance has been largely
overlooked in these studies. In addition, limited research has been conducted on
English sign language datasets.
This thesis introduces the VK-SLT model, a simple and efficient approach to
enhance hand representation by integrating key-point data and video representa
tion on the visual side through a channel attention mechanism. Building on ideas
from the GF-VLP model, we used CLIP to pre-train to align the visual represen
tation and text features. VK-SLT effectively suppresses background noise while
enhancing hand representation. This capability reduces the number of epochs
required to reach convergence and improves the performance of previous models.
Unlike its predecessor, the VK-SLT model integrates a cross-attention mechanism
into the CLIP’s image encoder. This mechanism refines the cls
token represen tation by selectively aggregating global context from video frames and key point
features. This VK-SLT model highlights critical hand gestures while suppress
ing irrelevant background information. This significantly improves the alignment
between visual and textual representations.
Preliminary evaluations demonstrate the effectiveness and efficiency of the
model. Compared to the GFSLT-VLP model, VK-SLT achieves comparable or
better BLEU scores while requiring fewer epochs in both pre-and post-training
(e.g., 1.13 BLEU improvement with 50% fewer epochs on a subset of How to
Sign dataset). These results highlight the potential of VK-SLT as a robust and
resource-efficient framework for promoting the alignment of the visual and textual
representations of sign language, thus paving the way for high-quality, glossless
sign language translation
The effects of adjuvant endocrine therapy on long-term outcomes from ductal carcinoma in situ: a systematic review and meta-analysis
No full textBackground:
Although adjuvant endocrine therapy (ET) is a standard treatment for hormone receptor positive ductal carcinoma in situ (DCIS), its use is variably recommended by clinicians. This paper reviewed the effects of ET in relation to recurrence and survival across diverse populations.
Methods:
PubMed, Embase, Web of Science, and Cochrane were searched for studies that reported outcomes of DCIS treated with ET versus no ET.
Results:
Three randomised trials and 42 cohort studies were included. In the trials, tamoxifen significantly reduced the risk of in-breast recurrence with a pooled hazard ratio (HR) of 0.69 (95 % CI: 0.60, 0.80). In the cohort studies, ET was associated with lower risks of any recurrence (HR 0.67; 95 % CI: 0.55, 0.83), ipsilateral breast tumour recurrence (HR 0.59; 0.51, 0.69), loco-regional recurrence (HR 0.74; 0.53, 1.02) and contralateral breast cancer (HR 0.70; 0.49, 1.00), and improved overall survival (HR 0.85; 0.79, 0.90). No significant association between ET and breast cancer specific survival was observed in either trials or cohort studies. Subgroup analyses within the cohort studies showed no significant differences in the pooled HRs for recurrence and survival by follow-up length, confounding adjustment or treatment type.
Conclusions:
The use of adjuvant ET reduces the risk of recurrence in patients with DCIS in clinical trials, as well as in the real-world setting. Survival benefits, however, warrant further study