42337 research outputs found
Sort by
Making the past readable: a study of the impact of handwritten text recognition (HTR) on libraries and their users
Get PDFThis thesis considers the socio-technical infrastructure surrounding AI-enabled Handwritten
Text Recognition (HTR), the process of converting images of historical textual materials into
computer-readable text. Despite sectoral awareness of the technology, institutional
approaches still display gaps between conceptualising and operationalising HTR. The
following main research question is answered: What is the impact of HTR on libraries, users,
and the wider information environment? Foundational knowledge is provided by exploring
HTR's relation to other computational methods and traditional areas of palaeography and
papyrology. Through a mixed methods approach, including interviews with National Library
of Scotland (NLS) staff, survey methods and thematic analysis, this thesis approaches HTR
from multiple vantage points. It critically reflects on HTR’s affordances regarding library
audience development by adapting digital engagement strategies, as well as the practicalities
in embedding the technology within content-holding institutions. In aligning insights from
HTR developers, providers and users, a technical fluency of how to operationalise HTR
institutionally is presented. This informs an understanding of HTR’s potential near future
implications on the information environment and research. Such analysis results in a set of
recommendations for HTR’s future provision, directed thematically at libraries, HTR users
and developers. These recommendations include ways to enhance institutional processes,
such as digital preservation and audience engagement, in relation to HTR outputs; as well as
assessing general HTR capability locally. Other recommendations involve enabling greater
collaboration in HTR projects: through dataset sharing principles, standardised metadata and
flexible tool usage
Condition monitoring of a direct-drive permanent magnet synchronous generator under demagnetisation and eccentricity faults
No full textThis research investigates fault diagnosis challenges in 20 kW direct drive air-cored
permanent magnet generators to enhance reliability and performance. It addresses gaps in
harmonic behaviour and fault detection, focusing on issues like demagnetisation and
eccentricity, which impact efficiency and lifespan. The study highlights harmonic
cancellation and how pole and phase coil configurations influence the detectability of fault
signatures. Results indicate that increasing phase coils reduces detectable fault harmonics,
while three-phase configurations enhance harmonic cancellation, improving understanding
of fault-related patterns.
Diagnostic methods such as Current Signature Analysis (CSA), the Park Vector Approach
(PVA), the Extended Park Vector Approach (EPVA), and Flux Monitoring were adapted from
induction machines to evaluate their application in PM generators. Using Finite Element
Modelling (FEM) alongside experimental validation, the results prove that CSA, in
combination with FEM, provides increased sensitivity to faults like demagnetisation and
eccentricity, making it effective for practical use in real-world systems.
This study makes an important contribution by examining how partial demagnetisation
affects electromagnetic and thermal performance. Results indicate that magnetic
imbalances caused by demagnetisation lead to circulating currents, which increase losses,
reduce efficiency, and aggravate fault severity. The research also identifies critical
thresholds for electromagnetic and thermal stresses, better understanding how faults
develop and progress in PM machines.
The investigation also includes the effects of static, dynamic, and mixed eccentricity faults
in air-cored PM generators, analysing their impact on efficiency and performance across
different severity levels. The research identifies effective diagnostic methods by studying
fault behaviour under varied conditions, extending the work traditionally done with induction
machines.
This thesis contributes to the field by improving the understanding of harmonic interactions,
fault-related behaviours, and diagnostic methods for PM generators. The findings support
the development of more reliable fault detection strategies, reduced maintenance needs,
and improved operational performance, particularly in renewable energy applications
Securing soils in a changing climate: A soil route map for Scotland
Get PDFThere are many risks threatening Scottish soils across different soil types and land covers. However, unlike air and water, there is no single overarching soil policy providing security and governance for Scottish soils. Soils are spread across multiple policy divisions, which results in a lack of cohesive leadership in tackling threats to soils. The aim of this route map is to consolidate the challenges of managing soil systems to develop an overarching strategy for delivering improved soil security across Scottish landscapes
Understanding visualization discourse and building discursive visualizations
Get PDFThis doctoral thesis investigates the potential of discursive visualizations to entice readers of visualizations to discussions about data. Through a suite of surveys, interfaces, and user studies it explores key aspects of how people talk about data and demonstrates how these conversations can be integrated into visualizations. Discursive visualization describes a genre of visualization interfaces, engagement practices, and representation methods that aim to integrate discussions about visualizations into visualizations.
Data visualization has become a popular, ubiquitous medium in science, journalism, entertainment, and across a variety of online environments, including news outlets, social media, dedicated blogs, or expert groups. When discussing data visualizations, expert and novel audiences alike express their thoughts and observations about the data.
Existing annotation systems have explored various facets and scenarios of these discussions about data visualizations, including collaborative sensemaking, public communication, or participatory data correction.
However, considering the ongoing increase in potential audiences, use cases, and venues to encounter and talk about data visualizations, the potential of understanding discourse and the interfaces required for its integration remain under-explored.
To explore the potential of existing discussions about data, I provide a structured analysis of public audience reactions to visualizations observed outside of controlled lab environments. My results describe ten reaction types on four reaction scopes as well as a description of intrinsic and extrinsic engagement drivers that motivate people to participate in discussions.
Building on this knowledge, I design an interactive prototype that allows to annotate data visualizations with personal stories.
Small but important details, for example where an annotation is placed or who write it are generally overlooked in prior research. My interface offers different viewing conditions to better understand the impact of such details on the reading experience, particularly focusing on the role of embedded audience annotations.
My results inform the notion of 'collective storytelling' that describes a genre of visualizations providing a space in which audience annotations become the primary concern of visualizations.
In a follow-up study, I design another prototype that introduces 'anchored annotations' and 'discursive patinas' as core mechanics to integrate discussions into visualizations, regardless of their specific chart types, topics, or desired reaction types. My evaluation shows that discursive patinas improve the ability to navigate discussions and guide people to comments that help to understand, contextualize, or scrutinize the visualization.
My studies demonstrate how discursive visualization supports collaborative sensemaking and allows for novel engagements with data and data visualizations, including collective storytelling, distant readings of discussions, and close readings of visualizations.
Audience annotations are usually not required in visualization design. Yet, they are practical devices that are at the core of discursive engagements, as they allow to guide interpretation, offer contrasting perspectives, offer feedback, establish personal relevance, promote empathy for others, and help people reflect on their own experiences. Conceding space and visual presence to the audience challenges existing power dynamics in the relation between visualization creators and consumers. Offering a platform for discussions that are not considered when designing visualizations allows for a multiplicity of narratives to coexist. Integrating local knowledge into visualizations can point to missing, biased, or simply incorrect data, and improve the validity of the data as a whole.
Based on these considerations, I see great potential for discursive visualization to provide the foundation for polyvocal, personal, and critical engagements with data
Electrical impedance tomography aided by deep learning for applications in lung imaging and two-phase flow measurement
No full textElectrical impedance tomography (EIT) is an emerging, agile, and non-invasive imaging technology
that has attracted extensive attention in both medical imaging and industrial inspection.
In medical applications, the principal challenge lies in the highly nonlinear and ill-posed nature
of EIT, which makes image reconstruction particularly sensitive to imperfect measurement
signals and less robust to model inaccuracies. Although traditional model-based reconstruction
methods have shown success in certain scenarios, they often exhibit limitations in terms
of image quality, conductivity reconstruction accuracy, and computational efficiency, thereby
constraining their practical application. In the industrial domain, a key issue is the high sensitivity
of the reference voltage to environmental changes; without timely updates, the reference
voltage inevitably effects the reconstructed conductivity distribution’s accuracy, ultimately
leading to suboptimal measurements of dispersed-phase volume fractions in two-phase flow
systems. Recently, data-driven approaches have become a research frontier in tomographic
imaging, widely applied to computed tomography (CT) and magnetic resonance imaging
(MRI). Nevertheless, learning-based reconstruction methods for EIT sensor systems are still
at an early stage, and the seamless integration of deep learning algorithms with EIT hardware
remains largely unexplored. Aiming to address this gap, this thesis systematically investigates
the development of robust EIT algorithms based on deep learning, high-accuracy data-driven
reference voltage estimation, and precise measurement of dispersed-phase volume fractions
in industrial two-phase flow systems. In so doing, we provide theoretical insights and practical
solutions for incorporating deep learning into EIT.
To tackle the trade-off between temporal resolution and imaging accuracy in pulmonary imaging
and industrial EIT, we propose a high-resolution reconstruction algorithm based on structureaware
hybrid fused learning. Unlike traditional data-driven methods that rely on a singlebranch
neural network to reconstruct multiple targets with different conductivity levels, our
algorithm decouples the measured voltage by a segmentation branch and a conductivity
reconstruction branch. It separately extracts structural features and conductivity information,
then fuses the two features to reconstruct the conductivity distribution of target region, thus
achieving a significant improvement in imaging accuracy. Importantly, we have thoroughly considered
network complexity in the algorithm design, striving to balance high-accuracy imaging
with real-time performance. Numerical simulations and experimental validations on multiple
datasets, including three datasets with different geometric shapes and three experimental
datasets, show that our proposed method excels in reconstructing multi-level, continuous
conductivity distributions. Furthermore, analysis on real lung breathing data demonstrates
the algorithm’s ability to accurately capture dynamic physiological changes over a complete
respiratory cycle, thereby confirming its reliability and effectiveness in practical applications.
Given the sensitivity of industrial time-difference EIT reference voltage to environmental changes,
current research on this key problem remains limited. Existing methods often use data fitting
to estimate the reference voltage; however, their accuracy deteriorates in complex and dynamic
environments. To address this challenge, we introduce two novel methods for reference
voltage estimation: a multiple measurement (MM) approach and a convolutional neural network
(CNN)-based approach. The MM method exploits the fluctuation characteristics of the
measurement voltage, allowing for rapid data fitting in dynamically changing scenarios. Meanwhile,
the CNN-based approach leverages deep learning to establish a nonlinear mapping
between the measurement voltage and the reference voltage, offering superior robustness in
both static and dynamic environments. Comprehensive experiments, using an NEL oil–water
dataset, a gas–water two-phase flow dataset from the University of Leeds, and a static water
tank dataset from the University of Edinburgh, demonstrate that our methods significantly
outperform traditional approaches in terms of reference voltage estimation accuracy, reconstruction
quality, and robustness. Notably, under dynamic conditions, the CNN method exhibits
exceptional performance, effectively meeting the real-time requirements of complex industrial
inspection scenarios.
Building on the remarkable capability of deep learning in nonlinear fitting and feature representation,
its application to EIT for accurate dispersed-phase volume fraction measurement in
the industrial sector presents exciting opportunities. By reducing the dependency on direct
reference voltage measurements, deep learning not only optimizes two-phase flow volume
fraction measurement processes but also greatly enhances measurement precision and efficiency.
To this end, we propose an approach called attention-UNet with a fully connected network
(AU-FC). By incorporating an attention mechanism, the method effectively suppresses
voltage regions unrelated to the volume fraction, thereby enhancing feature extraction of
crucial signals. Through the voltage reconstruction, AU-FC delivers highly accurate volume
fraction predictions. Experimental validation on our constructed NEL dataset reveals that AUFC
achieves superior performance in terms of MSE, MAE, and MAPE, surpassing existing
methods and demonstrating real-time inference capabilities, making it an efficient and robust
solution for online industrial monitoring.
In conclusion, this thesis addresses the research gap in pulmonary breathing-state monitoring
and two-phase flow dispersed-phase measurement using EIT by developing deep learning–
based imaging and detection methods. The findings significantly advance EIT technology
in terms of image quality, robustness, and application breadth, laying a solid theoretical and
practical foundation for EIT to evolve into a more intelligent and resilient imaging and measurement
technology
Generating porcine stem cell-derived hepatocytes for the study of hepatitis E virus
No full textHepatitis E Virus (HEV) causes approximately 20 million human infections every year globally, resulting in about 3.3 million symptomatic infections and approximately 44,000 deaths annually. Infection is generally self-limiting with a 0.5 to 3% mortality rate in the general population; however, it can cause considerable morbidity and mortality in at-risk populations. Clinical manifestations differ between genotypes; HEV-1 and HEV-2 infections have severe clinical consequences in pregnant women with mortality rates reaching as high as 25% in the third trimester while HEV-3 and HEV-4 infections in immunocompromised groups (transplant recipients, the HIV positive) are linked to chronic hepatitis, which may progress to cirrhosis.
Until the early 2000s, HEV was considered to only be endemic to certain developing countries of Asia and Africa, however, in the last twenty years HEV has been recognized as a clinical issue in developed countries. Of the four genotypes, HEV-3 is the most geographically diverse and is found to be increasingly emerging in developed countries. HEV-3 is primarily a zoonotically transmitted pathogen, typically being acquired by consumption of pork products.
HEV research has been limited by the lack of an efficient culture system. Like many other hepatic viruses, HEV is a slow growing virus that replicates to low titres. The viral replication cycle is still being elucidated and neither HEV-specific antivirals nor a vaccine is available yet. Current research relies heavily on lab-adapted virus strains and cancer cell lines. However, the physiological relevance of these cancer cell lines and strains is limited. Primary hepatocytes have often been employed as a model however these cells are not permissive to genetic edits, have donor-to-donor variability and are a finite resource. Recent research has highlighted induced pluripotent stem cell-derived hepatocyte-like cells (iPSCdHLCs) as a more physiologically relevant model for HEV infection, capable of supporting HEV growth from all four genotypes and clinical isolates. However, PSC technologies have seldom been applied to relevant livestock species, specifically HEV’s main zoonotic reservoir: pigs. This has limited HEV studies concerned with pork food safety, livestock antiviral or vaccine developments, models of zoonotic transmission and xenotransplantation.
This project aimed to establish a reverse genetics system for the study of HEV and a porcine in vitro model to study HEV using pluripotent stem cells. This is the first venture into HEV research conducted within the University of Edinburgh, outside of clinical work. As such, I generated an infectious cDNA clone based on the Kernow C1/p6 genome and several variants with additional fluorescent protein tags. I also acquired the lab-adapted HEV-3 Kernow C1/P6 virus and trialled a variety of infection conditions with these stocks. While producing a novel reverse genetics system did not work, we were able to demonstrate HEV infection in cancer-derived cell lines using the lab-adapted strain.
In parallel, I aimed to adapt several pre-existing human methodologies for differentiating stem cells into HLCs to a porcine iPSC line (Pa9s). Changes to cell morphology, gene- and protein expression were used to develop a reproducible method for generating iPSCdHLCs. The developed method produced a differentiation marker profile similar to that of ex vivo fetal porcine liver samples as determined by RT-qPCR. Furthermore, microscopy of these cells revealed that a subset of the iPSCdHLCs developed in this system exhibit polyploidy and have albumin production, suggesting a mature hepatic fate. Lastly, I further characterised the final iPSCdHLC model we had generated and validated that it was capable of supporting HEV infection. Liver identity was more robustly demonstrated by comparing RNA-Seq profiles of iPSCdHLCs to 67 other porcine cell lines and tissues. Furthermore, successful HEV infection of both immature hepatocytes and mature hepatocytes generated using this methodology was demonstrated by fluorescence microscopy. Altogether, the results of this thesis demonstrate that porcine iPSCdHLCs can provide a novel and physiologically relevant model for the study of HEV in vitro
A social justice perspective on the role of feminist funds in a changing funding context
Get PDFFeminist funds have become a key supporter of intersectional feminist movements in the last decade and a notable actor in the funding context. Emerging grey literature shows how the funding context is currently changing and how feminist funds are adapting to keep resourcing these movements, which warrants a qualitative study to expand on this initial data. This research aims to contribute to the collective reflection taking place about the future of resourcing intersectional feminist movements, and to uncover transformative actions feminist funds are taking. This research asks: 1) how is the funding context changing from feminist funds’ perspective? 2) how and in what ways are feminist funds adapting to the changes in the funding context and 3) what does this tell us about the role feminist funds can play in driving social justice?
Data was collected through semi-structured interviews with five employees of feminist funds and two employees from allied feminist organisations, and validated through a member-checking focus group. It finds that feminist funds are facing exacerbated pressures in a changing funding context, yet that they may be able to negotiate these new tensions in a way that can ultimately support the creation and leveraging of opportunities for transformative action. Feminist funds, as a collective and strategic actor, play a key role in social justice as they 1) enable the creation of autonomous spaces for imagination and contention, 2) serve as an example of fostering coalitions across difference by navigating contradictions through reflexivity and solidarity, and 3) enact alternative social relations. This research may be helpful for practitioners navigating the tensions of the changing funding context and for social movement and organisational researchers who seek to inform transformative action
An examination of the role of the systemic inflammatory response in the clinical and biological features of incurable cancer and cachexia
Get PDFBACKGROUND:
The influence of the immune system and inflammation in the development of cancer is widely accepted. The same inflammatory processes are also involved in the evolution of the clinical features of cancer. Cachexia, a syndrome of muscle and fat loss not easily reversed by changes in nutrition, is amongst the most common and devastating features of cancer, especially incurable cancer.
People with incurable cancer are an understudied population, and crucial questions remain regarding the cachexia syndrome in this group. Further foundational work to examine the biochemical and clinical features of cachexia is required in order to inform therapeutic trials and develop meaningful interventions to improve quality of life.
AIMS:
The overall aim of this thesis was to examine the biological and clinical features of cancer cachexia and incurable cancer with a focus on the role of inflammation. In order to achieve this aim, reviews of the current literature were conducted. A prospective observational study (Routine Evaluation of People Living with Cancer- REVOLUTION study) was also undertaken. Five key areas were addressed in the REVOLUTION study- physical function, symptoms, quality of life, body composition and inflammation.
METHODS:
Two systematic reviews were conducted. The first of which, described in chapter two, examined the relationship between circulating cytokines and symptoms in incurable cancer. The second review, described in chapter three, examined the relationship between circulating cytokines and cachexia/ weight loss in incurable cancer. Quality of studies was assessed in both reviews using a modified Downs and Black checklist detailed in chapter two.
The REVOLUTION study was conducted as detailed in chapter four. Patients were recruited from an inpatient hospice setting; were >18 years of age and had a diagnosis of incurable cancer (defined as clinical, histological, or radiological evidence of metastatic cancer, or receiving anti-cancer therapy with palliative intent). Patients were invited to a baseline assessment, with follow-up assessments at six and twelve weeks.
Demographic details and measurements of height, weight and body mass index were collected. Body composition was measured using bioimpedance analysis and (if available) analysis of CT images from routine oncology follow up. Physical function was measured objectively using the Karnofsky Performance Status Scale (KPS) and subjectively using the European Organisation for the Research and Treatment of Cancer – Quality of Life Questionnaire –C30 (EORTC QLQ-C30). The EORTC-QLQ-C30 was also used to gather information regarding symptoms and quality of life. The Functional Assessment of Anorexia/ Cachexia Therapy Scale (FAACT) was included to capture information regarding symptoms related to cachexia. To collect data regarding nutrition and physical function the Patient Generated Subjective Global Assessment – Short Form (PGSGA-SF) was included. Blood samples were analysed for simple markers of the systemic inflammatory response and concentrations of multiple pro- and anti-inflammatory cytokines (e.g., Interleukin (IL)-1α, IL-1β, IL-4, IL-6, IL-10, IL-17, and Tissue Necrosis Factor-α).
RESULTS:
Systematic Reviews:
IL‐6, TNF‐α, and IL‐8 levels were greater in cachectic patients than in healthy individuals. Several symptoms of incurable cancer were associated with elevated levels of circulating cytokines i.e., depression, fatigue, and appetite loss were all linked with increased levels of IL-6. These results support the suggestion of a role for inflammatory cytokines in symptom generation, however it is difficult to draw further conclusions due to limitations in study methodology. The definition of cachexia and weight loss thresholds adopted across the studies included were highly variable, as were the methods described for cytokine analysis and the instruments used for assessment of symptoms.
REVOLUTION Study:
Between July 2020 and June 2021 45 patients were recruited to the REVOLUTION study. Mean age was 65 years, (range 39-91). The majority had a KPS score of 50- 70% (50%- requiring help often, requiring frequent medical care, 70% cares for self, unable to do normal activity or to do active work). The most common primary sites of cancer in the population were breast and colorectal cancer. Thirteen patients participated in assessments at time point two, seven patients participated in assessments at time point three. The most common reason for non-completion was death.
Cachexia was not a pre-requisite for recruitment to the study however at baseline 60% of patients were cachectic using a 5% weight loss threshold. In keeping with this finding, the FAACT and PGSGA-SF mean questionnaire scores revealed a high level of nutritional impact symptoms and a significant need for intervention. At baseline using the EORTC-QLQ-C30, patients had a low overall global health status mean score (37.6/100). In terms of cytokine analysis GDF-15 (r=0.511, p=0.004) and IL-1ra (r=0.492, p=0.002) had a significant relationship with BMI. There was a strong correlation between IL-6 levels and constipation (r=0.663, p=<0.01) and a moderate correlation between IL-8 levels and nausea and vomiting (r=0.486, p=0.002) and between TNF-α and diarrhoea (r=0.409, p=0.34). The study adopted a patient-centred approach in order to optimise patient involvement, however this made it more challenging to obtain consistent standardised measurements.
CONCLUSION:
Results of the systematic reviews suggest links between the systemic inflammatory response, cancer cachexia and the symptoms of incurable cancer. As inflammatory signalling patterns are dynamic, involving cytokines with myriad functions, a standardised longitudinal approach to further research is likely to be required to disentangle these complex relationships.
The findings of the REVOLUTION study reinforce the idea that cancer cachexia extends beyond mere weight loss, involving factors such as inflammation, nutritional status, and muscle wasting. The study's exploration of cytokines, such as GDF-15 and IL-1ra, revealed potential links to BMI and suggested these biomarkers could be crucial in understanding cancer cachexia. In an exploration of the symptoms of incurable cancer it was found that nutritional impact symptoms were common and had links with higher levels of inflammatory cytokines such as IL-6, IL-8 and TNF-α. This study provides a foundation for further research to understand the lived experience of incurable cancer and cancer cachexi
Investigating type I interferon signalling regulation in the context of human Mendelian autoinflammatory disease
Get PDFType I interferons (IFN-Is) are primary antiviral cytokines. Homeostasis of IFN-I signalling is under tight regulation, with too little or too much IFN-I activity resulting in significant pathology. While impaired IFN-I signalling is characterised by immunodeficiency, inappropriate upregulation of IFN-I signalling results in a set of autoinflammatory states termed the type I interferonopathies. In this work, through the study of real-world patients, distinct but related regulatory mechanisms important for homeostatic IFN-I signalling were investigated.
In the first part of the thesis, I studied a patient presenting with features consistent with a type I interferonopathy, who we identified to carry a rare homozygous missense variant p.(A219V) in STAT2. Through in vitro testing, I showed that while STAT2 p.(A219V) maintained the ability to transduce an IFN-I signal, its negative regulatory function was impaired due to defective binding of USP18.
In the second part of this study, I report a novel cohort of patients with a highly stereotyped clinical phenotype comprising normal early development followed by the onset of subacute neuro-regression, with increased IFN-stimulated gene (ISG) expression in whole-blood and raised neopterin levels in cerebrospinal fluid. We found these patients to harbour rare heterozygous variants in PTPN1 resulting in loss of mRNA and/or protein expression, suggesting that haploinsufficiency of PTPN1 leads to upregulated IFN-I signalling. Indeed, cells deficient in PTPN1 demonstrate an upregulation of ISG expression at baseline, and ‘hypersensitivity’ to stimulation with IFNα2b and the STING agonist diABZI. Underlying mechanisms could involve enhanced STING signalling, as evidenced by an overexpression of IFNB1 upon diABZI stimulation, and/or enhanced IFN-I signalling, where higher levels of STAT1 phosphorylation upon IFNα2b stimulation result in overproduction of ISGs.
Overall, this thesis describes the study of IFN-I mediated autoinflammation, with a particular focus on IFN signalling downstream of the IFN-I receptor. This work thus contributes new knowledge relevant to the physiological regulation of IFN-I signalling in humans, with potentially important implications for clinical testing and treatment
This meal reminds me: using participatory and photographic art practices to explore ways that food is used to connect to and commemorate the dead in the UK
Get PDFThe funeral is emphasised as being the formal goodbye in loss and often the final time mourning an individual will occur as a collective endeavour. Whilst funerals are integral ritualised spaces for the bereaved, they do not minimise the impact that grief has on their daily lives and identities. Many religious and cultural traditions across the world use food to commemorate and connect with the dead in an ongoing manner, while on an everyday level, death ritualization uses mundane objects to help to facilitate the work of memory practices and endeavours to counter the loss caused by death.
This art by practice PhD research employs an interdisciplinary way of working to explore ways that food is used to connect with and commemorate the dead within the UK. Food has been used as a mediator within both participatory and photographic art practices to inform of and reflect upon individual and culturally shaped human behaviours that operate within the wider context of British society. Discussions centred around individual food customs, such childhood memories, familial traditions, and private commemorative practices, provide insight into differing grief experiences. While the development of participatory framed exchanges have been used to inform, challenge, and support the work produced, helping to consistently generate new forms of knowledge from new interpretations.
While initially used to investigate individual memories and narratives around grief, framed exchanges evolved to reflect the needs of the bereaved participants, becoming supportive, inclusive, and safe spaces where both shared experiences and the sharing of experiences could simultaneously occur. These framed exchanges acknowledge the livings’ relationship with the dead and offer comfort in grief by providing them with the opportunity for their grief narratives and experiences to be recounted, witnessed and validated, thereby offering the potential for healing.
In addition to food acting as a catalyst for conversation and the translation of individual grief narratives, still lives depicting food were used to visually explore a range of research themes as they arose from other strands of theoretical and participatory interactions. Set within the photographic studio and darkroom, analogue photography and the process of image-making, became a central research methodology. The methodical and repetitive acts used within traditional analogue photographic processes, as well as both the restricted and heightened effect on sensory experiences these settings create, has led to the development of an immersive, ritualised and self-reflexive embodied practice used to reflect on participatory encounters and explore my own relationship to grief. The experimental approach to photographic practice and exhibiting both informed and transformed the wider investigation and research process, which emphasised process rather that final imagery produced
Access Repository Dashboard
Do you manage Edinburgh Research Archive? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!