Close
The following research areas have been identified as the main areas of expertise at IHI. Expand the accordions to find out more about the research and the experts involved.
- Artificial Intelligence / Machine Learning
Addressing the barriers to translating machine learning algorithms into patient benefit, with a particular focus on data-scarce learning and dataset drift on structured time-series EHR datasets.
Contact:
Dr Xi Bai
- Cardiovascular Disease
Cardiovascular diseases (CVD) and their risk factors, some听of which are diseases in their own right, such as hypertension and diabetes,听represent the largest burden of morbidity and mortality not just in the UK, but worldwide. Our researchers at IHI use 鈥渂ig data鈥� approaches in translational research across aetiology, diagnosis, treatment and prevention of CVD.
Contact:
- Care Homes
Integrating electronic health records with diverse data sources (omics, qualitative research) to improve the management of infection across health and social care settings.
Contact:
- Child Health Informatics
Utilising routine and administrative data to understand the health of children and families, and evaluate policy impact.
Contact:
- Clinical Decision Support / Nudging Trials
Developing knowledge-driven models by deploying AI algorithms and natural language processing techniques to achieve dermatology triage based on GPs' referrals.听The volume of healthcare data continues to grow, as does the work required to analyse it. However, this has yet to produce a significant number of actionable insights that could improve patient outcomes.听We do not yet have a strong evidence base for many treatments and interventions given in routine clinical care. Clinical actions are currently driven by expertise instead of evidence from randomised controlled trials (RCTs) that are expensive, and labour and time intensive to conduct.听Variations in care may therefore reflect both a lack of evidence or suboptimal guideline adherence. However, by harnessing听digital tools and data-enabled trials, it is possible that we can reduce the gap between evidence and practice in healthcare.听The increasing use of clinical decision support systems (CDSS) gives us the opportunity to use听computer prompts called digital 鈥榥udges鈥櫶齮o modify clinician behaviour and improve adherence to guideline-directed therapy.听What is more, 鈥榥udging鈥� gives us the opportunity to initiate data-enabled randomised controlled trials in areas where strong evidence is particularly lacking.
Contact:
Dr Tom Lumbers
Dr Anoop Shah
- Covid-19
Integrating data from routine and administrative sources, as well as from observational studies and clinical trials to describe the epidemiology and outcomes of Covid-19 infection in different populations.
Contact:
Prof Amitava Banerjee
- Data Linkage
Data linkage aims to bring together information from two or more different sources to create a new, richer dataset.听 Data may be linked for different units including individuals, organisations, geographical areas, or time points.
Contact:
- Drug Development
Working on an ontology enrichment project to predict drug targets based on existing knowledge graphs.
Contact:
- Electronic Health Records
An electronic health record (EHR) is the systematised collection of patient health information in a digital format. EHR are commonly derived from general practices and hospitals and may include information on demographics and biometrics such as height and weight, medical history, medication and allergies, immunisations, laboratory test results, medical imaging, vital signs, and billing information.听We analyse听longitudinal patient data from primary and secondary care听electronic health records, extract听concepts and structure information from free text, and develop methods that utilise the limitations and strengths of this valuable data resource. The overall aim is to improve patient care using patient information.
Contact:
Dr Tom Lumbers
Dr Anoop听Shah
- Environmental Health
Aiming to enhance our understanding of how environmental exposures, including climate change, impact human health, particularly those who are the most vulnerable in our population and at critical periods (for example, in utero).
Contact:
- Epidemiology
Epidemiology is the systematic study of the distribution, trends and causes of disease within populations, and can also encompass the study of disease prevention and control that applies epidemiological methods.
Contact:
- Evidence Synthesis
Evidence synthesis is finding, screening, critically appraising, synthesising, analysing, and summarising the existing evidence to answer the research, policy, or practice question.
Contact:
- Genomics
Genomics research focuses on the identification and characterisation of genetic variants that contribute to the risk of developing disease or effect听patient outcome or treatment.听听
Contact:
Dr Tom Lumbers
- Interoperability / Standards / Clinical Data Models
We study ways to structure the information in health records听to ensure that its meaning is preserved, and information can easily flow between systems for better patient care and research.听Software interoperability is the ability of different computer systems to effectively exchange or share data with immediate use and understanding of the exchanged information. Standards for clinical data models refer to the development of a standardised data structure and description to enhance the interoperability of clinical data databases.
Contact:
Dr Anoop听Shah
- Migrant Health / Health Inequality / Digital Health
Studying the health of migrants, including their healthcare utilisation and outcomes. Health inequalities are avoidable differences in health between groups based on certain characteristics, such as age, sex and gender, disability, deprivation or ethnicity. This research identifies those inequalities and proposes strategies to address them.
颁辞苍迟补肠迟:听
- Observational Data / Policy
Health research using observational data involves the description and analysis of data from observational epidemiological studies or routine data collection (as opposed to experimentally-collected data), to improve understanding of the epidemiology of disease and the impacts of exposures and interventions.
Contact:
- Patient and Public Engagement
Aiming to equitably improve the health of the public through the application of data science and public health research.
Contact:
- Phenomics
Phenomics听is concerned with the measurement of observable characteristics of an individual over time and in response to genetic and environmental influences.
Contact:
Dr Tom Lumbers
- Precision Medicine
Precision medicine aims to deliver clinically relevant disease phenotypes and outcomes, scalable insights from real-world evidence from electronic health records and registries and -omics research for drug target discovery, pharmacogenetics and stratified medicine and insights driving drug development and personalised medicine through advanced big-data analytics.
Contact:
Dr Tom Lumbers
- Public Health
Public health has been defined as 'the science and art of preventing disease, prolonging life and promoting health through the organized efforts and informed choices of society, organizations, public and private, communities and individuals'. Reducing unjust and avoidable inequalities in health is also a core element of public health. It encompasses and draws on a broad range of related disciplines including epidemiology and statistics, evaluation and applied health research, sociology, health economics and policy.
Contact:
听
- Risk Modelling / Risk Prediction
Developing methodology to integrate relevant patient information to understand and model risk for diseases on an individual or population basis.
Contact:
- Sex and Gender / Women's and Reproductive Health
Sex and gender听research focuses on differences in health between males vs听females and women vs听men in health, and the epidemiology and management of diseases. In women's health, we aim听to improve the health of women across the life course.
Contact:
- Software Development
Software development refers to computer science activities like software design, coding and programming, testing, and听documenting听performed in order to create a computer program or application.
Contact:
- Statistical Genetics
Contact: