University Hospitals Birmingham NHS Foundation Trust (UHB) has partnered with medical technology company, Qureight, to build more diverse artificial intelligence (AI) models that accurately reflect communities; a world-first lung disease research partnership. This will help ensure that patients from minority ethnic backgrounds receive better-tailored and effective treatments.
Idiopathic Pulmonary Fibrosis (IPF) is a type of complex lung disease that affects roughly 50 in every 100,000 people. It causes the lungs to become scarred, leading to severe breathlessness and coughing, and owing to a lack of effective medicines and treatments, it currently has a survival time worse than most cancers.
Applying cutting-edge AI tools, to unlock insights from existing IPF patient data, could help extend the lives of thousands of UK patients each year. However, current research data does not always reflect the true make-up of communities in Britain.
Birmingham’s diverse population will now provide scientists at Qureight with more diverse data than is typically available, enabling the building of AI diagnosis models that better reflect population diversity.
Birmingham is one of the UK’s first super-diverse cities according to the most recent census, which means that the majority of its residents are from Black, Asian or minority ethnic backgrounds.
This is hugely significant for the city’s lung disease clinical teams and researchers, who will be using lung scan data to build on the currently lacking understandings of how complex lung disease manifests and develops in these specific communities.
The aim of the project is to accelerate the development of lung disease treatments to improve outcomes for all patients. This is the first time that significant volumes of data from minority ethnic people will be structured and made available for complex lung disease research and drug development.
Dr Anjali Crawshaw, Consultant Respiratory Physician at UHB, said:
“Complex inflammatory and scarring lung diseases can be challenging to manage; it can be difficult to decide if the disease is responding to treatment, is stable or is getting worse.
“It is currently necessary for specialist radiology doctors to analyse CT scan images of the lungs, as part of the diagnosis and monitoring process. The process is open to interpretation bias and so the outcome may not always be the same.
“In addition, a shortage of these specialists makes this process slow and difficult. At the same time, the limited data we have available comes from predominantly from white, European patients, who may experience lung disease in a specific way.
“Launching this partnership with Qureight is a very significant moment for our team. Allowing access to patient data, that truly reflects the unique diversity of Birmingham’s population, and that will be invaluable to the planning and delivery of more equitable patient care – not just in cities like ours.”
Qureight’s AI, built by their own team of lung experts, will combine the data from patient scans (e.g. lung and airway volume) with lung function data from tests, blood results, and demographic records. This information will be securely and anonymously processed to deliver insights into the presentation, development, and progression of complex lung conditions.
Dr Muhunthan Thillai, Consultant Chest Physician, CEO and co-founder at Qureight, said: “Our partnership with UHB will be pivotal to how we harness our world leading AI solutions, to benefit patients from across the spectrum of complex lung diseases.
“One of the biggest problems with AI in healthcare is the lack of applicability to real world patients. AI is already being used by Qureight in complex lung disease clinical trials in Europe and the USA, but a more diverse patient population is key to its future success.”
“By working in partnership with globally renowned researchers at UHB, we will be able to rely on their clinical expertise to super-charge medical progress and help rebalance inequalities in our understanding of rare and complex lung diseases through the use of the latest AI technologies.”
Questions and answers
The following questions and answers cover the data licence agreement between UHB and Qureight for the Idiopathic Pulmonary Fibrosis (IPF) dataset.
Is UHB selling patient data?
No, UHB is not selling any patient data.
Is Qureright funding any aspect of this agreement?
Qureight are funding the costs of a data analyst to extract, prepare and secure the data for this project. Other costs included in the contract are for administration and business costs only.
Data licence agreement
Why have UHB chosen to enter into this data licence agreement?
Patients with complex lung diseases (such as Idiopathic Pulmonary Fibrosis) can be hard to manage; it may be challenging to decide if their diseases are stable or getting worse, and if they are on a drug treatment whether the disease will respond to a medicine or not.
The main test clinicians use to monitor these patients is a breathing test called Forced Vital Capacity (FVC). As the disease progresses the FVC level drops but there are a number of issues with this measurement as the patient's technique, and whether or not they have co-existing illnesses, can affect the score.
Patients with these diseases also have CT scans (images) of their lungs as part of diagnosis and monitoring but analysis of these scans can be complex and need specialist radiology doctors to interpret them. However, there is a lot of information in the image that computers may be able to analyse that could help improve treatment for these patients.
As part of this collaboration, UHB will work with Qureight who will use their advance computer software to analyse scans and other data using an artificial intelligence technique called machine learning.
Analysis of CT scans will measure lung volume, airway volume, vascular volume and fibrosis levels. Analysis of anonymised longitudinal data will allow us to characterise the patient population and understand how these patients progress with their lung disease. Tracking changes in the patient data and scans will give us more information on how to better treat patients in the long term, compared to how we treat them at the moment.
UHB will provide a rich and diverse dataset for this project, reflective of the diverse patients we care for, and this will help tackle the disparity of care that patients with lung fibrosis have.
Most of the research to date is on patients with European Caucasian ancestry and it is less clear how lung diseases should be managed in patients from other ethnic groups. The higher proportion of ethnic minority patients from Birmingham will help provide a more diverse data set for use in this research.
Data and datasets
What kind of data will Qureight have access to as part of the data licence agreement?
Qureight will be given access to anonymised CT scans, demographic data (such as age and sex) and clinical tests such as lung function results and blood results.
Whose data will be held by Qureight?
The data will be from 2000 patients with complex lung diseases who have been treated at UHB.
How long will Qureight have access to UHB patient data for?
The data licence agreement is in place for five years. After this time the data held by Qureight will be securely destroyed.
How will the patient data be stored?
The data will be stored on the Qureight secure cloud platform – a Trusted Research Environment. This environment has undergone an assessment by UHB to ensure that Qureight have the necessary cybersecurity controls in place.
What safeguards are in place to ensure anonymised data cannot be traced back to identify patients?
In compliance with the national data opt-out policy, UHB will cross-check the list of potential patients against the national data opt-out list to ensure no patients who have chosen to opt-out of sharing their data are included in the data set.
The data will also be anonymised at source at UHB before it is transferred. Qureight has additional mechanisms to check that the data has been anonymised before it analyses them. No identifiable patient data will be held by Qureight.
What checks are in place to ensure patient information is being used appropriately?
UHB adheres to the ‘Five Safes’ for data access, storage and use and the highest standards of security and transparency are applied at every stage of a project from the anonymisation and storage of patient data, through to publishing research outcomes.
The lead researcher and the research team at UHB will oversee this project. They will conduct regular checks and reviews throughout the project and as the data is analysed, to ensure Qureight is remaining compliant with the terms and conditions of the data licence agreement, which direct that the data can only be used for stated research purposes.
Can patients opt out of sharing their information?
Yes. Patients can opt-out of having their information used for medical research purposes via the national opt-out mechanism. Data for patients who opt-out after anonymised datasets have been created, cannot be removed.