179566 research outputs found
Sort by
Can we reconstruct a dysarthric voice with the large speech model Parler TTS?
Get PDFSpeech disorders can make communication hard or even impossible for those who develop them. Personalised Text-to-Speech is an attractive option as a communication aid. We attempt voice reconstruction using a large speech model, with which we generate an approximation of a dysarthric speaker’s voice prior to the onset of their condition. In particular, we investigate whether a state-of-the-art large speech model, Parler TTS, can generate intelligible speech while maintaining speaker identity. We curate a dataset and annotate it with relevant speaker and intelligibility information, and use this to fine-tune the model. Our results show that the model can indeed learn to generate from the distribution of this challenging data, but struggles to control intelligibility and to maintain consistent speaker identity. We propose future directions to improve controllability of this class of model, for the voice reconstruction task
Immersive virtual reality assessments of working memory and psychomotor skills:A comparison between immersive and non-immersive assessments
Get PDFObjectiveImmersive virtual reality (VR) enhances ecological validity and facilitates intuitive and ergonomic hand interactions for performing neuropsychological assessments. However, its comparability to traditional computerized methods remains unclear. This study investigates the convergent validity, user experience and usability of VR-based versus PC-based assessments of short-term and working memory, as well as psychomotor skills, while also examining how demographic and IT-related skills influence performance in both modalities.MethodsSixty-six participants performed the Digit Span Task (DST), Corsi Block Task (CBT) and Deary-Liewald Reaction Time Task (DLRTT) in both VR- and PC-based formats. Participants' experience in using computers and smartphones, and playing videogames, was considered. User experience and system usability of the formats were also evaluated.ResultsWhile performance on DST was similar across modalities, PC assessments enabled better performance on CBT and faster reaction times in DLRTT. Significant correlations between VR and PC versions supported convergent validity. Regression analyses revealed that performance on PC versions was influenced by computing and gaming experience, whereas performance on VR versions was largely independent of these factors, except for gaming experience predicting performance on CBT backward recall. Moreover, VR assessments received higher ratings for user experience and usability than PC-based assessments.ConclusionImmersive VR assessments provide an engaging alternative to traditional computerized methods, with minimal reliance on prior IT experience and demographic factors. This resilience to individual differences suggests that VR may offer a more equitable and accessible platform for automated cognitive assessment. Future research should explore the long-term reliability of VR-based assessments
Match and training injuries sustained by professional male rugby union players in Scotland
Get PDFRugby union ('rugby') is a full-contact sport, with previous studies across the globe reporting a high incidence of injury. However, no injury surveillance study of professional male players in Scotland exists in contemporary literature. The current study therefore aimed to describe the incidence, severity, burden and nature of match and training injuries sustained by male professional club rugby players in Scotland. A prospective cohort study of injuries sustained during matches and training across the 2017/18 and 2018/19 seasons was undertaken, with injury and exposure definitions in line with the international consensus statements. Injury incidence was 136.2/1,000 player match hours and 4.1/1,000 player training hours, the median injury severity was 7.0 days (match) and 7.5 days (training) and injury burden was 2,887.0/1,000 player match hours and 102.3/1,000 player training hours. Concussion (match) and posterior thigh muscle injuries (training) were the most common specific diagnoses. Injury incidence in this population was higher than reported elsewhere in previous studies. However, high incidences of tackle injuries and concussion injuries agree with previous literature, reinforcing the need for mitigation strategies targeting these areas.</p
Tackle characteristics and suspected concussion:Recommendations to improve high school girls' rugby tackle safety
Get PDFObjectives To evaluate the association between tackle characteristics and suspected concussion among Canadian high school girls' rugby union players. Methods A case-control video-analysis study was used to examine the association between a priori tackle characteristics (eg, type, height, head position) and suspected concussion. For every concussive tackle event, six non-concussive tackle events were matched for game and team. Penalised logistic regression was used to estimate ORs with 95% CIs for concussion given tackle characteristics. Results 33 concussive tackle events, including 38 concussions, were identified. 20 were ball carrier concussions (53%). 261 (228 non-concussive, 33 concussive) tackle events, including 632 individual tackler and ball carrier instances, were coded. All ball carrier head contact intensity types were associated with concussion compared with no head contact. A trip tackle type (OR: 4.41, 95% CI 1.25 to 15.61), illegal tackle type (OR: 4.41, 95% CI 2.67 to 7.29), deceleration (OR: 14.03, 95% CI 4.65 to 42.30) and no change of speed (OR: 18.81, 95% CI 10.04 to 35.24) increased concussion odds for ball carriers. A tap (OR: 10.64, 95% CI 2.00 to 56.62) and trip tackle type (OR: 5.91, 95% CI 3.18 to 10.99), two (OR: 3.38, 95% CI 1.13 to 10.07) or three (OR: 13.91, 95% CI 1.74 to 111.53) tacklers within the event, and a head-down position (OR: 40.54, 95% CI 27.78 to 59.18) increased concussion odds for tacklers. Higher tackler contact on ball carrier (tackle height) increased concussion odds for ball carrier (reference: knee-to-upper-leg; waist-to-sternum OR: 2.27, 95% CI 1.08 to 4.80; sternum-to-armpit OR: 5.65, 95% CI 1.40 to 22.87) and tacklers (sternum-to-armpit OR: 4.20, 95% CI 1.26 to 14.03). Conclusion Numerous tackle characteristics were associated with ball carrier and tackler concussion. This is the first study to identify tackle height on ball carrier as a risk factor for concussion in girls' rugby. Future directions should consider tackle training programmes and lower tackle height as potential concussion prevention solutions.</p
Burnout and relationship quality in residential childcare – Exploring the role of care worker and teacher emotional intelligence
Get PDFThe quality of relationships between staff and children in residential childcare is crucial for fostering healthy psychological and emotional development in vulnerable populations. This study explores the links between staff burnout, emotional intelligence (EI) and the quality of interactions in childcare facilities providing residential care and full-time education in the UK. More specifically, the aim of the study was to assess the role of EI in mitigating burnout’s adverse effects on staff-child relationships. Using cross-sectional survey data from 78 participants (70.5% residential childcare workers; 29.5% residential teachers), the results indicated a significant negative association between burnout and relationship quality, with EI showing a positive correlation with better relationship outcomes. However, contrary to expectations, EI did not moderate the link between burnout and relationship quality, suggesting that even individuals with high EI are not fully protected from the impacts of burnout. Further analysis revealed that client-related burnout had a particularly detrimental effect on staff-child relationships, highlighting the need for targeted interventions
CFD Predictions of Fire Spread over Wood Cribs in Large Open-Plan Compartments under Two Different Ventilation Conditions
Get PDFFire spread behaviours occurring under different ventilation conditions in a large-scale compartment are explored by means of a CFD-based numerical simulator coupled with a multi-component pyrolysis model. The model captures differences in the burning behaviours between the scenarios, showing the role of pre-heating in supporting enhanced burning rates. In the less well-ventilated test this gives rise to a sharp increase in fire spread rates, which are eventually more than doubled, representing a transition to a growing fire, and ultimately a fully-developed fire. The analysis also discriminates the role in stick ignition between the evolving fluxes from the flame in the fire plume and externally (i.e. those from hot layer and compartment boundaries), showing that flame fluxes become progressively less important, more rapidly, in the less well-ventilated test. These results highlight the value of comprehensive modelling approaches which exploit detailed solid-phase pyrolysis models, capturing the fire evolution in different ventilation conditions, thereby permitting exploration of the fire spread behaviours in manners which are inaccessible experimentall
Plug-and-Play Myoelectric Control via a Self-Calibrating Random Forest Common Model
Get PDFObjective. Electromyographic (EMG) signals show large variabilities over time due to factors such as electrode shifting, user behavior variations, etc substantially degrading the performance of myoelectric control models in long-term use. Previously one-time model calibration was usually required each time before usage. However, the EMG characteristics could change even within a short period of time. Our objective is to develop a self-calibrating model, with an automatic and unsupervised self-calibration mechanism. Approach. We developed a computationally efficient random forest (RF) common model, which can (1) be pre-trained and easily adapt to a new user via one-shot calibration, and (2) keep calibrating itself once in a while by boosting the RF with new decision trees trained on pseudo-labels of testing samples in a data buffer. Main results. Our model has been validated in both offline and real-time, both open and closed-loop, both intra-day and long-term (up to 5 weeks) experiments. We tested this approach with data from 66 non-disabled participants. We also explored the effects of bidirectional user-model co-adaption in closed-loop experiments. We found that the self-calibrating model could gradually improve its performance in long-term use. With visual feedback, users will also adapt to the dynamic model meanwhile learn to perform hand gestures with significantly lower EMG amplitudes (less muscle effort). Significance. Our RF-approach provides a new alternative built on simple decision tree for myoelectric control, which is explainable, computationally efficient, and requires minimal data for model calibration. Source codes are avaiable at: https://github.com/MoveR-Digital-Health-and-Care-Hub/self-calibrating-rf
Messages from beyond the grave: How can Drosophila help us to understand the consequences of apoptosis?
Get PDFCell Death is an integral part of life. Cells strategically die during the embryonic development of an organism to help sculpt developing tissues into their final form and cells are constantly lost and replaced during homeostasis as well as through injury. In each of these scenarios, the resulting corpses must be efficiently cleared and surviving cells must react to the loss of their dying neighbour appropriately such that the health and function of the tissue is maintained. For this to happen effectively healthy cells need to be alerted to a death in their community rapidly and efficiently. Fortunately, cell death is not a silent process, there is no quiet drifting off into the afterlife for the departing cell but instead it announces its demise through the creation of a plethora of signals both secreted and membrane bound designed to make its neighbours aware of its imminent departure. The fruit fly Drosophila melanogaster through its powerful genetics and live imaging has proved an invaluable model in helping researchers understand not only the signalling that drives the process of programmed cell death but also the complex multicellular communications within a tissue that influence a cell’s decision to die as well as the downstream consequences of that death. Here we highlight how research using the humble fly can help to understand the complex multi-cellular coordination required to successfully deal with the loss of a cell within the complex setting of a 3-dimensional living organism
Morphological and Functional Phenotyping of Skeletal Muscle and Bone in the zQ175 Knock-In Mouse Model of Huntington’s Disease
Get PDFBackground: Huntington’s disease (HD) is a progressive neurodegenerative disorder primarily affecting the central nervous system (CNS). However, emerging evidence suggests that peripheral tissues, including skeletal muscle and bone, also undergo pathological changes contributing to disease burden. Objective: To characterize musculoskeletal impairments in the zQ175 knock-in (KI) mouse model of HD, through integrated behavioral, biomechanical, and imaging analyses.Methods: Motor function was assessed using grip strength, rotarod, and open field testing. Ex vivo contractility of the extensor digitorum longus (EDL) and Soleus (Sol) muscles was measured. Muscle fiber cross-sectional area (CSA) was quantified using semi-automated segmentation. Bone microarchitecture was analyzed using high-resolution micro-computed tomography (μCT).Results: Six-month-old homozygous zQ175 mice exhibited significantly reduced muscle strength and impaired contractile properties in both the EDL and Soleus muscles compared to wild-type (WT) controls. µCT analysis revealed decreased trabecular bone volume and alterations in bone structure. Conclusions: These findings provide a comprehensive musculoskeletal phenotyping of zQ175 mice, revealing early-onset muscle atrophy and skeletal fragility. Our study highlights the importance of targeting peripheral manifestations in HD and establishes zQ175 KI mice as a valuable additional model for investigating systemic disease mechanism
FjordRPM v1.0: a reduced-physics model for efficient simulation of glacial fjords
Get PDFInteractions between ice masses and the ocean are key couplings in the global climate system. In many cases these interactions occur through glacial fjords, which are long, deep, and narrow troughs connecting the open ocean to marine-terminating glaciers. By controlling the fluxes of ocean heat towards the ice sheet and ice sheet freshwater towards the ocean, glacial fjords play an important role in modulating ice sheet mass loss and the impacts of freshwater on ocean circulation. Yet, these dynamics occur at small scales that are challenging to resolve in earth system models and so are they often ignored, represented in an ad-hoc manner, or studied using expensive high-resolution models that are limited in scope. Here, we propose a means of capturing glacial fjord dynamics at negligible computational expense in the form of a "reduced-physics" model (FjordRPM) that resembles a "1.5-dimensional" or box model. We describe the design and physical parameterisations in the model and demonstrate its ability to capture important modes of glacial fjord circulation by comparing it against a general circulation model in idealised and realistic simulations. We suggest that the model is a useful tool for understanding fjord dynamics and a promising approach for representing glacial fjord processes within large-scale models or climate and sea level projection efforts
Access Repository Dashboard
Do you manage Edinburgh Research Explorer? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!