index.xml

<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Sheffield AI Research Engineering</title><link>https://shef-AIRE.github.io/</link><atom:link href="https://shef-AIRE.github.io/index.xml" rel="self" type="application/rss+xml"/><description>Sheffield AI Research Engineering</description><generator>Wowchemy (https://wowchemy.com)</generator><language>en-us</language><lastBuildDate>Mon, 01 Jan 2024 00:00:00 +0000</lastBuildDate><image><url>https://shef-AIRE.github.io/media/icon_hu885f0cfeed998fbf7dc869c80f754bbd_18100_512x512_fill_lanczos_center_3.png</url><title>Sheffield AI Research Engineering</title><link>https://shef-AIRE.github.io/</link></image><item><title>Multimodal AI for Parkinson's Disease</title><link>https://shef-AIRE.github.io/project/parkinson/</link><pubDate>Mon, 01 Jan 2024 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/project/parkinson/</guid><description>&lt;p>&lt;strong>University of Sheffield Collaborating Faculties&lt;/strong>: Sheffield Institute for Translational Neuroscience (SITraN) and Faculty of Engineering&lt;/p>
&lt;p>&lt;strong>Overview&lt;/strong>: Our project is dedicated to the development of artificial intelligence tools for Parkinson&amp;rsquo;s Disease, designed to elucidate the underlying mechanisms of the condition and predict its progression. This initiative integrates data from multiple modalities—including genetic data, biomarkers, environmental factors, and medical examinations—utilizing advanced AI methodologies such as contrastive learning, foundation models, and causal discovery.&lt;/p>
&lt;p>&lt;strong>Motivation&lt;/strong>: The growing interest in applying artificial intelligence (AI) to tackle Parkinson&amp;rsquo;s Disease reflects a comprehensive appreciation of the condition&amp;rsquo;s widespread impact, its escalating incidence, the existing gaps in our comprehension, and the extraordinary research opportunities afforded by current data repositories.&lt;/p>
&lt;p>As the second most prevalent neurodegenerative disorder in the UK, Parkinson&amp;rsquo;s Disease presents a formidable public health challenge. Its widespread nature underlines the pressing demand for novel treatment and management strategies, positioning AI as an ideal candidate to drive forward innovative solutions. Globally, the frequency of Parkinson&amp;rsquo;s Disease is on the rise, affecting an ever-growing number of individuals either directly or putting them at a significant risk of developing the condition. This increasing trend accentuates the urgent necessity for interventions that are both scalable and efficacious, areas where AI technology shines with potential. Despite thorough research efforts, the core mechanisms behind Parkinson&amp;rsquo;s Disease remain a mystery. Here, AI&amp;rsquo;s ability to sift through and analyze large, complex datasets could unlock new understanding, setting the stage for transformative developments in how we treat and prevent the disease.&lt;/p>
&lt;p>The &lt;a href="https://www.ukbiobank.ac.uk/" target="_blank" rel="noopener">UK Biobank&lt;/a>, with its comprehensive data, emerges as a pivotal resource for AI research. The extensive scope of this data lays a robust groundwork for the creation and refinement of AI models, fostering considerable progress in our grasp and handling of Parkinson&amp;rsquo;s. Leveraging AI, researchers are not just shedding light on the elusive causes of Parkinson&amp;rsquo;s Disease but are also crafting predictive models that foresee the disease&amp;rsquo;s trajectory, pinpoint therapeutic targets, and ultimately, instill hope in the countless lives touched by this afflictive illness.&lt;/p>
&lt;p>Thus, the convergence of AI and Parkinson&amp;rsquo;s Disease research heralds an exciting era in the battle against neurodegenerative diseases. It offers a beacon of hope for enhancing patient outcomes, marking a vital step forward in our journey towards understanding, managing, and eventually overcoming such conditions.&lt;/p></description></item><item><title>Digital Materials Discovery</title><link>https://shef-AIRE.github.io/project/digital-materials-discovery/</link><pubDate>Thu, 14 Dec 2023 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/project/digital-materials-discovery/</guid><description>&lt;p>&lt;strong>University of Sheffield Collaborating Faculties&lt;/strong>: Faculty of Engineering, Royce Institute&lt;/p>
&lt;p>&lt;strong>Overview&lt;/strong>:
This project uses AI to accelerate the discovery of new materials crucial for green technologies. It explores both data and modelling aspects to predict material composition and properties, through two case studies: permanent magnetic materials and novel corrosion-resistant coatings.
The project uses machine learning algorithms for predictions, drawing on both existing and self-generated databases enriched by natural language processing. It addresses inherent challenges like data scarcity and imbalance through data augmentation and the integration of machine learning with physics-based models.
The project will deliver high-quality methodologies, open-source code, and a user-friendly interface to broaden the application of these predictive capabilities.
This has the potential to revolutionize materials discovery by enabling the software to autonomously predict material properties from specified compositions, paving the way for significant breakthroughs.&lt;/p>
&lt;p>&lt;strong>Motivation&lt;/strong>:
The demand for sustainable technologies requires the discovery of novel materials with superior properties.
Traditional methods for materials discovery are time-consuming and resource-intensive due to their reliance on physical experiments and iterative testing, limited by both practical laboratory setups and human imagination.
This project bridges this gap by harnessing the power of AI, using machine learning algorithms, digital databases, and computational tools. This significantly reduces the time it takes to translate a concept into a material with desired properties.
By enabling autonomous prediction of material properties, this project has the potential to revolutionize the field of materials science. In the short term, it can accelerate the discovery cycle, leading to faster development of solar cells and energy-efficient materials. In the long term, this project could pave the way for entirely new materials with unforeseen properties, ultimately propelling advancements across diverse fields like aerospace, medicine, and electronics.&lt;/p></description></item><item><title>Multi-fidelity Fusion and Optimization Theory and Applications</title><link>https://shef-AIRE.github.io/project/multifidelity-fusion/</link><pubDate>Mon, 06 Nov 2023 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/project/multifidelity-fusion/</guid><description>&lt;p>&lt;strong>University of Sheffield Collaborating Faculties&lt;/strong>: Faculty of Science, Faculty of Engineering and AMRC&lt;/p>
&lt;p>&lt;strong>Overview&lt;/strong>: Our project aims to investigate novel methods to leverage a special type of multimodal data—multi-fidelity data—to improve AI model accuracy and efficiency, leading to scalable solutions in computationally intensive optimisation problems in various engineering disciplines.&lt;/p>
&lt;p>&lt;strong>Motivation&lt;/strong>: Our goal is to create innovative techniques and toolsets capable of assimilating and being able to use multi-fidelity data effectively (i.e., providing the same outputs of high-fidelity datasets using lower-fidelity datasets), such as simulation outcomes from both coarse and dense meshes, and signals from high-cost precision sensors and low-cost basic sensors. We aim to enhance AI model accuracy without relying on highly accurate data obtained from high-cost precision sensors only, whilst improving efficiency metrics such as training time and memory cost. We will implement the proposed method in simulation-based testing for autonomous driving systems (ADS) to demonstrate the capability for the safety and reliability of self-driving cars with low-cost computation.&lt;/p></description></item><item><title>AI Brain Imaging for Nerve Pain Detection (UKRI funded)</title><link>https://shef-AIRE.github.io/project/nerve-pain-detection/</link><pubDate>Sun, 01 Oct 2023 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/project/nerve-pain-detection/</guid><description>&lt;p>&lt;strong>University of Sheffield Collaborating Faculties&lt;/strong>: Faculty of Medicine, Dentistry and Health, Faculty of Engineering&lt;/p>
&lt;p>&lt;strong>External Partners&lt;/strong>: University of Oxford, University of Dundee, AstraZeneca&lt;/p>
&lt;p>&lt;strong>Overview&lt;/strong>: Neuropathic pain, arising from damage or disease impacting the nervous system, stands as a significant health concern. Our project aims to develop an AI-based neuroimaging model capable of predicting treatment responses and clinical phenotypes in patients afflicted with neuropathic pain. We plan to conduct extensive external validation studies across multiple sites and conditions to ensure the development of a robust and objective AI-based neuroimaging model.&lt;/p>
&lt;p>&lt;strong>Motivation&lt;/strong>: Chronic neuropathic pain afflicts one in ten adults over 30, stemming from injuries to the sensory nervous system. The incidence of this debilitating pain is anticipated to escalate due to ageing, a surge in diabetes cases, and improved cancer survival rates. Neuropathic pain markedly impairs daily functioning, manifesting symptoms like burning or &amp;rsquo;electric-shock&amp;rsquo; sensations, potentially leading to depression and a significant diminution in quality of life. Present medications provide only marginal relief to roughly half of the affected individuals, and are associated with side effects. Over the past quarter-century, the quest for more efficacious drugs for neuropathic pain has stagnated, a likely consequence of the diverse sub-types of the condition and the unpredictable nature of treatment responses. Collaborating with AstraZeneca and other universities, our endeavor is to unearth new biomarkers for neuropathic pain, aiming to invigorate drug development initiatives and enhance their effectiveness.&lt;/p></description></item><item><title>Text Correction for Historical Documents</title><link>https://shef-AIRE.github.io/project/historical-doc-text-correction/</link><pubDate>Thu, 06 Jul 2023 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/project/historical-doc-text-correction/</guid><description>&lt;p>&lt;strong>University of Sheffield Collaborating Faculties&lt;/strong>: Faculty of Arts &amp;amp; Humanities, Faculty of Engineering, Digital Humanities Institute (DHI)&lt;/p>
&lt;p>&lt;strong>External Partner&lt;/strong>: British Library&lt;/p>
&lt;p>&lt;strong>Related Links&lt;/strong>: &lt;a href="https://www.dhi.ac.uk/text-correction-for-mining-historical-documents/" target="_blank" rel="noopener">https://www.dhi.ac.uk/text-correction-for-mining-historical-documents/&lt;/a>&lt;/p>
&lt;p>&lt;strong>Overview&lt;/strong>: This project addresses the critical issue of correcting noisily OCR&amp;rsquo;d historical documents, focusing on the &lt;a href="https://www.gale.com/intl/primary-sources/british-library-newspapers" target="_blank" rel="noopener">British Library Newspapers (BLN) collection&lt;/a>. BLN is a major corpus of over 200 years of scanned British newspapers from over 240 newspapers with textual data, visual data, and metadata available. Scanned newspaper images have undergone OCR (optical character recognition) processing, resulting in inaccurate transcriptions due to the degradation of the original documents. The project aims to employ advanced deep-learning techniques to improve the quality of these transcriptions. The final outputs will be high-quality corrected transcriptions of BLN and open-source code for OCR text correction, both of which would serve as valuable resources for humanities researchers.&lt;/p>
&lt;p>&lt;strong>Motivation&lt;/strong>: Since the early 2000s, significant digitisation efforts have been undertaken to preserve and make accessible historical primary sources such as newspapers, early printed books, and handwritten documents. While these efforts have been instrumental in advancing humanities research, the low quality of OCR transcriptions remains a significant barrier to discovering new historical insights. The successful completion of this project promises both short-term and long-term benefits. In the short term, it will significantly enhance the transcription quality of BLN, enabling accurate and efficient searching within the collection as well as unlocking the potential for text mining, which was previously impractical due to low transcription quality. In the long term, the project&amp;rsquo;s success could revolutionise research on other large collections of historical documents, allowing researchers to track content changes, language evolution, and shifts in thought across different time periods. By being language-independent, the impact could extend to historical documents worldwide, advancing research on a global scale.&lt;/p></description></item><item><title>Multimodal Cardiothoracic Disease Prediction</title><link>https://shef-AIRE.github.io/project/cardiothoracic-disease-prediction/</link><pubDate>Wed, 05 Jul 2023 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/project/cardiothoracic-disease-prediction/</guid><description>&lt;p>&lt;strong>University of Sheffield Collaborating Faculties&lt;/strong>: Faculty of Medicine, Dentistry and Health, Faculty of Social Science, and Faculty of Engineering&lt;/p>
&lt;p>&lt;strong>External Partner&lt;/strong>: Sheffield Teaching Hospitals NHS Foundation Trust&lt;/p>
&lt;p>&lt;strong>Overview&lt;/strong>: Our project aims to develop a sophisticated Artificial Intelligence (AI) system which can process multimodal, multi-vendor, multi-centre, and multi-pathophysiological cardiothoracic data, such as Chest Radiographs (CXR), Echocardiogram (ECG), Computed Tomography (CT), Magnetic Resonance Imaging (MRI) and Electronic Health Record (EHR), to segment and classify pathophysiological features and improve the diagnosis, prognosis, and therapeutic response prediction of Cardiothoracic Disease (CTD) such as Pulmonary Hypertension (PH), Chronic Obstructive Pulmonary Disease (COPD) and Abnormal Heart Rhythms (AHR) to a level at which advanced methods such as contrastive learning, foundation model, meta-learning, few-shot and zero-shot learning can successfully extract interpretable clinical parameters.&lt;/p>
&lt;p>&lt;strong>Motivation&lt;/strong>: Cardiothoracic disease refers to a variety of conditions that affect the heart and lungs, including coronary artery disease, heart failure, lung cancer, and diseases of the chest wall. These illnesses can impact the overall function and structure of the heart and lungs, and they often require specialised care, potentially including surgery. Despite substantial progress in medical technology, the early diagnosis of these diseases remains a challenge due to the complexity of disease development and the vague symptoms produced in the early stages. In our initial study, we will utilise the &lt;a href="https://archive.physionet.org/physiobank/database/mimicdb/" target="_blank" rel="noopener">MIMIC datasets&lt;/a>, which comprise various data modalities within a comprehensive open-access database. This dataset has been the foundation for high-quality research in areas ranging from intensive care and mortality prediction to disease classification in pathology. Additionally, we will use the in-house dataset, &lt;a href="https://www.hra.nhs.uk/planning-and-improving-research/application-summaries/research-summaries/aspire-registry/" target="_blank" rel="noopener">ASPIRE registry&lt;/a>, which also offers multiple data modalities, to further test our model. In the short term, the project&amp;rsquo;s success could lead to improved diagnosis and monitoring of cardiothoracic diseases, potentially reducing the need for invasive procedures and facilitating personalised treatment plans. In the long term, our AI system could be adapted to more cardiovascular and respiratory diseases, revolutionising the approach to cardiothoracic medicine and benefiting countless patients worldwide.&lt;/p></description></item><item><title>Registration for information session on AI Research Engineer openings</title><link>https://shef-AIRE.github.io/news/23-03-31-information-session/</link><pubDate>Fri, 31 Mar 2023 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/news/23-03-31-information-session/</guid><description>&lt;p style="text-align: justify;">
Register TODAY via &lt;a href="https://lnkd.in/eJv76fRg">https://lnkd.in/eJv76fRg&lt;/a> to attend the hybrid information session on our FIVE 3-year (Senior) AI Research Engineer positions at University of Sheffield from 13:30 to 14:30 (BST, GMT+1) on 31st March. Profession Haiping Lu will explain who we are looking for, how we will work together, and what opportunities we can offer, and answer any questions you may have, either submitted via the form or to be asked during the session.
&lt;br>
To build a team with five full-time staff, we are very keen to have female members on board so we strongly encourage female candidates to apply and highly appreciate if you could share this opportunity with potential female candidates in your network. Thank you!
&lt;br>
&lt;p>Job application deadline: 24th April 2023.
&lt;br>&lt;/p>
&lt;p>Application link: &lt;a href="https://lnkd.in/eJyE5eH4" target="_blank" rel="noopener">https://lnkd.in/eJyE5eH4&lt;/a>&lt;/p>
&lt;/p></description></item><item><title/><link>https://shef-AIRE.github.io/admin/config.yml</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/admin/config.yml</guid><description/></item><item><title/><link>https://shef-AIRE.github.io/home/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/home/</guid><description/></item><item><title/><link>https://shef-AIRE.github.io/q-and-a/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/q-and-a/</guid><description/></item><item><title/><link>https://shef-AIRE.github.io/team/</link><pubDate>Mon, 01 Jan 0001 00:00:00 +0000</pubDate><guid>https://shef-AIRE.github.io/team/</guid><description/></item></channel></rss>