We are generating insights into
complex disease patterns, risk trajectories and treatment effects.

Our approach

Data from +10M patients

Our initial sources of data will include:

Large-scale research dataset. An unprecedented wealth of multimodal data from 500,000 participants, with work underway data to collect further imaging data.

Large-scale electronic EHR. The world’s largest database of linked primary and secondary care records.

Pooled randomised trial data. The largest randomised database involving data from over 40 trials and more than 270,000 randomised patients.

We are also exploring other large biomedical datasets across Europe.

Machine intelligence

Combining expertise in healthcare, biomedical data and advanced machine learning, our team investigates whether applying data mining, machine learning and deep learning techniques to large biomedical datasets can lead to better understanding and management of chronic diseases and multimorbidity.

A multidisciplinary team

The programme combines the following areas of expertise: clinical practice in cardiology and expertise in epidemiology and care delivery; Clinical practice in neurology/psychiatry; medical imaging and analysis; health/biomedical informatics; machine learning and big-data analytics; computer vision and deep learning. We believe such a consortium, in close collaboration with Oxford’s scientific community will deliver the analysis tools and clinical solutions that healthcare systems worldwide can benefit from.
Meet the team

Machine Intelligence

Data Mining

We adopt when necessary traditional data modelling and mining techniques.

Machine Learning

We use the most advances techniques in Machine Learning, such as Restricted Boltzmann Machines (RBMs).

Deep Learning

Convolutional neural networks (CNNs) are one of the most important methodologies in Deep Learning that have not yet been extensively applied to neuroimaging and cardiac imaging. Their application in computer vision has been hugely successful.

Research stream 1

Cognitive Decline: detection and prediction.

We will find the markers that can predict the rate of cognitive decline, which can lead to the identification and diagnosis Mild Cognitive Impairment (MCI) and dementia and the prediction of their onset.



Data preparation

We will tackle data extraction processes on large biomedical datasets to transform raw data into meaningful markers or imaging-derived phenotypes through advanced algorithms.


Data analysis

We will apply Data Mining, Machine Learning and Deep Learning methods, as well as some standard statistical techniques to advance our understanding of complex disease patterns and risk.


Solutions development

We will provide data-driven solutions for early prediction of diseases and their precise classification, share new findings and analysis tools with biomedical scientists, and help clinicians with decision-support software.

Innovative solutions for clinicians

Because we know that software engineering is key to put in practice the insights and models we will develop in the programme, we aim to develop innovative solutions based on decision-support systems to help clinicians.

Learn more


We are seeking academic colaborators and commercial partners.
If you are interested in being involved in our programme, don't hesitate to contact us.