本期为大家推荐爱丁堡大学、伦敦国王学院2024最新奖学金介绍。
01、爱丁堡大学
Water security in a changing climate; future patterns of drought hazard
University of Edinburgh | School of Engineering
博导：Prof L Beevers,Dr E Medina-Lopez
截止日期：Saturday, February 17, 2024 资助的博士项目（全球学生）
项目描述： About the Project Background
Hydrological drought hazard results from extreme low flows in rivers, reducing water supplies and thus the capacity for abstraction causing water shortages. Currently, the UK’s vulnerability to drought hazard has reached the warning threshold (20%) on the Water Exploitation Index (water abstraction as a percentage of the freshwater resource); it is thus defined by theEuropean Environment Agencyas a water-stressed country (EEA, 2019).
Recent research (Collet et al., 2018,Visser-Quinn et al., 2019; (Kay et al., 2018,Rudd et al., 2019),suggests that climate change may represent an additional stressor, with hydroclimatological projections indicating an increase in the frequency and intensity of hydrological droughts in the coming decades. It is clear that drought hazard represents a major threat to water security globally, and the UK is no different. However, these hazards are subject to spatial variation, which may change in the future. With regional variations in these hazards, as well as regional population patterns, it is crucial to study such phenomena at a large geographical scale (e.g. country level).In addition to spatial considerations, temporal analyses are needed, using flow or precipitation time series.
Projections of future river flows (runoff) are the product of a long and complex modelling chain: emissions scenarios forceGeneral Circulation Models(GCMs), the outputs of which are downscaled to force hydrological models. Uncertainties in terms of model input, structure, and parameters, cascade through the modelling chain (Clark et al., 2016). Looking forward to future climate projections, any analyses require to consider ensemble projections, thus posing questions around probabilistic analyses.
Aims and Objectives
Droughts can be explored in many ways, one of which is categorising by severity and exploring the transition between severity states. Markov chains are a powerful tool which allow us to explore patterns as an alternative to traditional hydrological models by utilising flows projections either directly from regional climate models (Aitken et al 2022) or through modelling chains such as EFLaG (Aitken et al, 2023).
The aim of this PhD is to explore the regional variation in drought severity transitions regionally across the UK. The research will include developing novel ways to establish the probability of different events occurring, as well as grappling with large datasets to detect patterns of transition or occurrence.
Methods
Some initial work by the research group has established the validity of using different ensemble climate projections for analysis such as this. Similarly some tentative exploration has indicated that using Markov Chains can support pattern exploration in drought occurrences in natural catchments. Consequently this PhD will explore hydrological climate projections at the spatial coverage of Great Britain. At this scale the PhD will analyse:
- Flow derived within the regional climate models (RCMs) - the hydrological component is online coupled, allowing greater feedbacks to be captured.Exploring a wide range of GCMs/ESMs captured - across multiple modelling centres. This will consider a wide representation of processes, and represent a more holistic capture of model structural uncertainties.
- Use Markov Chains to categorise flow transitions and explore drought progression patterns
- Develop and explore methods to examine and detect patterns across spatial and temporal data
Eligibility
Minimum entry qualification - an Honours degree at 2:1 or above (or International equivalent) in a relevant science or engineering discipline*, possibly supported by an MSc Degree. Further information onEnglish language requirements for EU/Overseas applicants.
*An undergraduate degree in Civil or Environmental Engineering; or Physical Geography, or potentially Mathematics or Statistics Further Information
The University of Edinburgh is committed to equality of opportunity for all its staff and students, and promotes a culture of inclusivity. Please see details here:https://www.ed.ac.uk/equality-diversity
Funding Notes Tuition fees + stipend are available for Home/EU and International studentsFurther information and other funding options: https://www.eng.ed.ac.uk/studying/postgraduate/research/phd-scholarships/
02、伦敦国王学院
An Artificial Intelligence approach for epidemiological samples and their implementation in mental healthcare systems King’s College London |Institute of Psychiatry, Psychology and Neuroscience
博导：Dr Paris Alexandros Lalousis,Prof NikolaosKoutsouleris
截止日期：Friday, January 26, 2024
资助的博士项目（全球学生）
项目描述：
About the Project
Artificial Intelligence holds tremendous potential to revolutionise decision-making in mental healthcare systems. By analysing vast amounts of patient data and identifying patterns, AI algorithms can assist clinicians in accurate diagnosis, personalised treatment plans, and timely interventions.
AI-powered tools can also monitor patient progress, provide real-time feedback, and offer support to both patients and healthcare providers. Indeed, such applications have been achieved in other areas of medicine. Data from medical records have been used to identify individuals at risk of intensive care use, imaging data have shown to enhance detection of diminutive adenomas and hyperplastic polyps and treatment response prediction to fifteen distinct cancer types has been achieved using advanced machine learning approaches.
Such advancements applied in mental healthcare systems hold promise to enhance the quality and accessibility of services, leading to improved outcomes and a better understanding of mental illnesses. While AI algorithms have shown remarkable potential in psychiatric research, their full impact on healthcare remains limited without translation into clinical settings. Bridging this gap is crucial to unlock AI's benefits in patient care.
This PhD project will utilise existing multimodal data (neuroimaging, clinical records, genomics, blood-based biomarkers) from the EU FP-7 funded PRONIA study, the UK Biobank, the German National Cohort (NAKO) as well as the Clinical Record Interactive Search (CRIS) system within the NIHR Maudsley Biomedical Research Centre.
The student will be trained in novel machine and deep learning methods to build predictive models of mental health disorders as well as examine their applicability to real-world clinical data (CRIS) and epidemiological data (UK Biobank and NAKO). The student will be based at the Artificial Intelligence in Mental Health (AIM) lab which is co-led by the Chair of Precision Psychiatry Professor Nikolaos Koutsouleris and the Lecturer in Artificial Intelligence in Mental Health Dr Paris Alexandros Lalousis. It is a new lab based in the Department of Psychosis Studies at the Institute of Psychiatry, Psychology & Neuroscience. The student will benefit from access to a strong network of collaborators and research innovation within the Institute of Psychiatry, Psychology & Neuroscience, the NIHR Maudsley Biomedical Research Centre, as well as the Early Psychosis Studies and the Section for Precision Psychiatry in Munich.
This will provide a rich and diverse research environment, helping the student develop their skills and knowledge, and a strong professional network. Specific training will include data analysis using shallow and deep learning techniques and curation/harmonization of existing phenotypic data across the aforementioned cohorts; these are essential skills in mental health research and will therefore be valuable for the student's career development.
The studentship will come with a 2- to 4-month residency in Munich for deep/machine learning training in Professor Nikolaos Koutsouleris’ lab. We are looking for candidates who have strong interpersonal skills, a willingness to learn from and teach others, a desire to be an innovative leader in the field, and strong technical and analytical abilities. For more information, please use the following
link:https://www.kcl.ac.uk/ioppn/study/research-funding/PL-IOPPN-CRIS-NK-24
When applyin, in the
Funding section, please tick box 5 and include the following reference: (PL-IOPPN-CRIS-NK-24)
【竞赛报名/项目咨询请加微信：mollywei007】
