Biomedical Informatics

Biomedical Informatics Research Group is a multidisciplinary group of informaticians, clinicians, psychologists and computer scientists, researching the role of data and knowledge in medical research and practice. Our core area of interest is the concept of the Learning Health System (LHS) which is a part of a growing field of ‘learning systems’ where knowledge acquisition and process improvement become at least semi-automated tasks of the human-cyber-social infrastructure. Thus, applications such as clinical trials, diagnostic support and epidemiological studies, are all redefined in terms of data and knowledge flows through the health system. Through our work, we are committed to improving the quality and safety of clinical care and efficiency and effectiveness of research it is based on.

More information on our software, together with some demos, can be found at https://kclhi.org

Members

Ghita Berrada

Research Associate in Health Informatics

Martin Chapman

Lecturer in Health Informatics

Vasa Curcin

Head of Department, Population Health Sciences

Jesus Dominguez Alvarez

Research Fellow

Davide Ferrari

PhD Student

Tianyi Liu

PhD Student

Iain Marshall

Clinical Senior Lecturer in Population Health Sciences

Frank Soboczenski

Research Associate in Health Informatics

Yijun Wang

PhD Student

Wenjuan Wang

Research Fellow/Senior Data Scientist

Dan Youkee

Co-investigator, Stroke in Sierra Leone

CONSULT: Collaborative Mobile Decision Support for Managing Multiple Morbidities

CONSULT (Collaborative mObile decisioN Support for managing mULtiple morbidiTies) is a 5-year research project aiming to design a novel technology that harnesses the power of artificial intelligence, sensors and mobile devices and supports patient decision making, and thus actively engages patients in managing their health. Using this mobile application, people will benefit by having better personalised information about their health, helping them, and their health providers, make better decisions about their treatments.

Consult Project

RobotReviewer

RobotReviewer is a machine learning system aiming which aims to support evidence synthesis. Systematic reviews now inform everything from national health policy to bedside care. But producing these reviews requires researchers to identify the entirety of the relevant literature and then extract from this the information to be synthesized; a hugely laborious and expensive exercise. Moreover, the unprecedented growth of the biomedical literature has increased the burden on those trying to make sense of the published evidence base. Concurrently, more systematic reviews are being conducted every year to synthesize the expanding evidence base; tens of millions of dollars are spent annually conducting these reviews. RobotReviewer aims to mitigate this issue by (semi-) automating evidence synthesis using machine learning and natural language processing.

RobotReviewer

Phenoflow

Our work in phenomics focuses on capturing the phenotype definitions in an executable form that can be validated, shared and reproduced to achieve research transparency. One-size phenotypes do not fit all datasets, with variations in quality and structure directly impacting the reality of phenotyping. Phenoflow is a database of computable phenotype definitions launched in 2021 as the back-end to the HDR UK Phenotype Library of phenotype definitions that the researchers can access and contribute to.

Phenoflow

Phenotype Library

ROAD2H

ROAD2H is combining resource optimization, argumentation, clinical decision support and clinical and policy analytics techniques to facilitate uptake of health interventions to achieve Universal Health Coverage (UHC) in Low- and Middle-Income Countries (LMICs) with exemplars in Chinese and Serbian hospitals. I have developed a novel guideline-based reasoning engine that can combine several potentially overlapping or conflicting guidelines into a recommended treatment path.

ROAD2H

TrialStreamer

Trialstreamer is an artificial intelligence system and search engine, which finds and summarises new clinical trial publications, registrations, and preprints. A suite of neural network models automatically extracts key data (including characteristics of the participants, interventions, outcomes, biases, sample sizes, and author conclusions). Trialstreamer monitors databases continuously, so that new RCT evidence is findable the moment it is published, and users can surface the biggest and highest quality trials first. Trialstreamer is a collaboration between King's College London and Northeastern University, Boston, and is supported by the US National Library of Medicine (R01-LM012086-01A1).

TrialStreamer

Social media big data analysis for mental health research
Wongkoblap, A., Vadillo, M. A. & Curcin, V., 1 Jan 2021, Mental Health in a Digital World. Elsevier, p. 109-143 35 p. Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review. DOIs: https://doi.org/10.1016/B978-0-12-822201-0.00018-6
Real-world effectiveness of steroids in severe COVID-19: a retrospective cohort study
Wang, W., Snell, L. B., Ferrari, D., Goodman, A., Price, N. M., Wolfe, C., Curcin, V., Edgeworth, J. & Wang, Y., 6 Sep 2022, (Accepted/In press) In: BMC infectious diseases. Research output: Contribution to journal › Article › peer-review
Correction: Risk prediction of 30-day mortality after stroke using machine learning: a nationwide registry-based cohort study (BMC Neurology, (2022), 22, 1, (195), 10.1186/s12883-022-02722-1)
Wang, W., Rudd, A. G., Wang, Y., Curcin, V., Wolfe, C. D., Peek, N. & Bray, B., 25 Aug 2022, In: BMC Neurology. 22, 1, 319. Research output: Contribution to journal › Comment/debate › peer-review. DOIs: https://doi.org/10.1186/s12883-022-02840-w
Differences in Clinical Presentation With Long COVID After Community and Hospital Infection and Associations With All-Cause Mortality: English Sentinel Network Database Study
Meza-Torres, B., Delanerolle, G., Okusi, C., Mayor, N., Anand, S., Macartney, J., Gatenby, P., Glampson, B., Chapman, M., Curcin, V., Mayer, E., Joy, M., Greenhalgh, T., Delaney, B. & de Lusignan, S., 16 Aug 2022, In: JMIR Public Health and Surveillance. 8, 8, p. e37668 Research output: Contribution to journal › Article › peer-review. DOIs: https://doi.org/10.2196/37668
Developing a Long COVID Phenotype for Postacute COVID-19 in a National Primary Care Sentinel Cohort: Observational Retrospective Database Analysis
Mayor, N., Meza-Torres, B., Okusi, C., Delanerolle, G., Chapman, M., Wang, W., Anand, S., Feher, M., Macartney, J., Byford, R., Joy, M., Gatenby, P., Curcin, V., Greenhalgh, T., Delaney, B. & de Lusignan, S., 11 Aug 2022, In: JMIR Public Health and Surveillance. 8, 8, e36989. Research output: Contribution to journal › Article › peer-review. DOIs: https://doi.org/10.2196/36989
Clinical prediction models for mortality in patients with covid-19: external validation and individual participant data meta-analysis
CovidRetro collaboration, 12 Jul 2022, In: BMJ (Clinical research ed.). 378, p. e069881 e069881. Research output: Contribution to journal › Article › peer-review. DOIs: https://doi.org/10.1136/bmj-2021-069881
Using Microservices to Design Patient-facing Research Software
Chapman, M., G-Medhin, A., Sassoon, I., Kokciyan, N., Sklar, E. & Curcin, V., 2022, (Accepted/In press) IEEE 18th International Conference on eScience. Research output: Chapter in Book/Report/Conference proceeding › Conference paper › peer-review
What factors influence differential uptake of NHS Health Checks, diabetes and hypertension reviews among women in ethnically diverse South London? Cross-sectional analysis of 63,000 primary care records
Molokhia, M., Ayis, S., Karamanos, A., L'Esperance, V., Yousif, S., Durbaba, S., Ćurčin, V., Ashworth, M. & Harding, S., Jul 2022, In: EClinicalMedicine. 49, 101471. Research output: Contribution to journal › Article › peer-review. DOIs: https://doi.org/10.1016/j.eclinm.2022.101471
LOng COvid Multidisciplinary consortium Optimising Treatments and servIces acrOss the NHS (LOCOMOTION): protocol for a mixed-methods study in the UK
LOCOMOTION consortium, 17 May 2022, In: BMJ Open. 12, 5, p. e063505 Research output: Contribution to journal › Article › peer-review. DOIs: https://doi.org/10.1136/bmjopen-2022-063505
Risk prediction of 30-day mortality after stroke using machine learning: a nationwide registry-based cohort study
Wang, W., Rudd, A., Wang, Y., Curcin, V., Wolfe, C., Peek, N. & Bray, B., 27 May 2022, In: BMC Neurology. 22, 1, 195. Research output: Contribution to journal › Article › peer-review. DOIs: https://doi.org/10.1186/s12883-022-02722-1

View all publications

DECODE: Data-driven, machinE learning aided stratification and management of multiple long-term COnditions in adults with intellectual DisabilitiEs (NIHR, 2022-2024)
LOCOMOTION: LOng COvid Multidisciplinary consortium: Optimising Treatment and Interventions acrOss the NHS (NIHR, 2021-2023)
Improving lives of stroke survivors with data (NIHR, 2021-2026)
HDR UK National Human Phenome Project (HDR UK 2019-2022)
Implementing support for Observational Medical Outcomes Partnership (OMOP) Common Data Model at KCL (IMI EHDEN, 2020-2022)

Elixir All Hands

Applying FAIR principles to computable phenotype libraries, panel at Elixir All Hands 2022.

Applying FAIR principles to computable phenotype libraries

AMIA

Connecting computable phenotypes with multiple Health IT standards using the Phenoflow library, presentation at AMIA Clinical Informatics 2022.

AMIA 2022 Clinical Informatics Conference

Therapeutics Initiative

New models for evidence synthesis using machine learning/artificial intelligence, Therapeutics Initiative(TI) Methods Speaker Series, University of British Columbia, Canada.

Therapeutics Initiative

demonstration of RobotReviewer, Trialstreamer, and RobotReviewer LIVE

New technologies to speed up systematic review production—demonstration of RobotReviewer, Trialstreamer, and RobotReviewer LIVE, Workshops to Support EPA's Development of Human Health Assessments, National Academy of Sciences, US.

National Academies