̽��ֱ�� of Cambridge - Fiona Gilbert

World first for AI and machine learning to treat COVID-19 patients worldwide

cjb250 — Wed, 15 Sep 2021 16:19:22 +0000

̽��ֱ��research was sparked by the pandemic and set out to build an AI tool to predict how much extra oxygen a COVID-19 patient may need in the first days of hospital care, using data from across four continents.

̽��ֱ��technique, known as federated learning, used an algorithm to analyse chest x-rays and electronic health data from hospital patients with COVID-19 symptoms.

To maintain strict patient confidentiality, the patient data was fully anonymised and an algorithm was sent to each hospital so no data was shared or left its location.

Once the algorithm had ‘learned’ from the data, the analysis was brought together to build an AI tool which could predict the oxygen needs of hospital COVID-19 patients anywhere in the world.

Published today in Nature Medicine, the study dubbed EXAM (for EMR CXR AI Model), is one of the largest, most diverse clinical federated learning studies to date.

To check the accuracy of EXAM, it was tested out in a number of hospitals across five continents, including Addenbrooke’s Hospital. ̽��ֱ��results showed it predicted the oxygen needed within 24 hours of a patient’s arrival in the emergency department, with a sensitivity of 95 per cent and a specificity of over 88 per cent.

“Federated learning has transformative power to bring AI innovation to the clinical workflow,” said Professor Fiona Gilbert, who led the study in Cambridge and is Honorary Consultant Radiologist at Addenbrooke’s Hospital and Chair of Radiology at the ̽��ֱ�� of Cambridge School of Clinical Medicine.

“Our continued work with EXAM demonstrates that these kinds of global collaborations are repeatable and more efficient, so that we can meet clinicians’ needs to tackle complex health challenges and future epidemics.”

First author on the study, Dr Ittai Dayan, from Mass General Bingham in the US, where the EXAM algorithm was developed, said:

“Usually in AI development, when you create an algorithm on one hospital’s data, it doesn’t work well at any other hospital. By developing the EXAM model using federated learning and objective, multimodal data from different continents, we were able to build a generalizable model that can help frontline physicians worldwide.”

Bringing together collaborators across North and South America, Europe and Asia, the EXAM study took just two weeks of AI ‘learning’ to achieve high-quality predictions.

“Federated Learning allowed researchers to collaborate and set a new standard for what we can do globally, using the power of AI,'' said Dr Mona G Flores, Global Head for Medical AI at NVIDIA. “This will advance AI not just for healthcare but across all industries looking to build robust models without sacrificing privacy.”

̽��ֱ��outcomes of around 10,000 COVID-19 patients from across the world were analysed in the study, including 250 who came to Addenbrooke’s Hospital in the first wave of the pandemic in March/April 2020.

̽��ֱ��research was supported by the National Institute for Health Research (NIHR) Cambridge Biomedical Research Centre (BRC).

Work on the EXAM model has continued. Mass General Brigham and the NIHR Cambridge BRC are working with NVIDIA Inception startup Rhino Health, cofounded by Dr Dayan, to run prospective studies using EXAM.

Professor Gilbert added: “Creating software to match the performance of our best radiologists is complex, but a truly transformative aspiration. ̽��ֱ��more we can securely integrate data from different sources using federated learning and collaboration, and have the space needed to innovate, the faster academics can make those transformative goals a reality.”

Reference
Dayan, I et al. Federated learning for predicting clinical outcomes in patients with COVID-19. Nat Med; 15 Sept 2021; DOI: 10.1038/s41591-021-01506-3

Addenbrooke’s Hospital in Cambridge along with 20 other hospitals from across the world and healthcare technology leader, NVIDIA, have used artificial intelligence (AI) to predict COVID-19 patients’ oxygen needs on a global scale.

Creating software to match the performance of our best radiologists is complex, but a truly transformative aspiration

Fiona Gilbert

Mehrnews.com

A clinician helping a COVID-19 patient with an oxygen mask in a hospital in Iran

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Licence type:

Attribution

Cambridge extends world leading role for medical imaging with powerful new brain and body scanners

cjb250 — Mon, 24 Oct 2016 07:22:25 +0000

̽��ֱ��equipment, funded by the Medical Research Council (MRC), Wellcome Trust and Cancer Research UK, sits within the newly-refurbished Wolfson Brain Imaging Centre (WBIC), which today celebrates two decades at the forefront of medical imaging.

At the heart of the refurbishment are three cutting-edge scanners, of which only a very small handful exist at institutions outside Cambridge – and no institution other than the ̽��ֱ�� of Cambridge has all three. These are:

a Siemens 7T Terra Magnetic Resonance Imaging (MRI) scanner, which will allow researchers to see detail in the brain as tiny as a grain of sand
a GE Healthcare PET/MR scanner that will enable researchers to collect critical data to help understand how cancers grow, spread and respond to treatment, and how dementia progresses
a GE Healthcare hyperpolarizer that enables researchers to study real-time metabolism of cancers and other body tissues, including whether a cancer therapy is effective or not

These scanners, together with refurbished PRISMA and Skyra 3T MRI scanners at the WBIC and at the Medical Research Council Cognition and Brain Sciences Unit, will make the Cambridge Biomedical Campus the best-equipped medical imaging centre in Europe.

Professor Ed Bullmore, Co-Chair of Cambridge Neuroscience and Scientific Director of the WBIC, says: “This is an exciting day for us as these new scanners will hopefully provide answers to questions that we have been asking for some time, as well as opening up new areas for us to explore in neuroscience, mental health research and cancer medicine.

“By bringing together these scanners, the research expertise in Cambridge, and the latest in ‘big data’ informatics, we will be able to do sophisticated analyses that could revolutionise our understanding of the brain – and how mental health disorders and dementias arise – as well of cancers and how we treat them. This will be a powerful research tool and represents a big step in the direction of personalised treatments.”

Dr Rob Buckle, Director of Science Programmes at the MRC, adds: “ ̽��ֱ��MRC is proud to sponsor this exciting suite of new technologies at the ̽��ֱ�� of Cambridge. They will play an important role in advancing our strategy in stratified medicine, ultimately ensuring that the right patient gets the right treatment at the right time.”

Slide show: Click on images to expand

7T Medical Resonance Imaging (MRI) scanner

̽��ֱ��Siemens 7T Terra scanner – which refers to the ultrahigh strength of its magnetic field at 7 Tesla – will allow researchers to study at unprecedented levels of detail the workings of the brain and how it encodes information such as individual memories. Current 3T MRI scanners can image structures 2-3mm in size, whereas the new scanner has a resolution of just 0.5mm, the size of a coarse grain of sand.

“Often, the early stages of diseases of the brain, such as Alzheimer’s and Parkinson’s, occur in very small structures – until now too small for us to see,” explains Professor James Rowe, who will be leading research using the new 7T scanner. “ ̽��ֱ��early seeds of dementia for example, which are often sown in middle age, have until now been hidden to less powerful MRI scanners.”

̽��ֱ��scanner will also be able to pick up unique signatures of neurotransmitters in the brain, the chemicals that allow its cells to communicate with each other. Changes in the amount of these neurotransmitters affect how the brain functions and can underpin mental health disorders such as depression and schizophrenia.

“How a patient responds to a particular drug may depend on how much of a particular neurotransmitter present is currently present,” says Professor Rowe. “We will be looking at whether this new scanner can help provide this information and so help us tailor treatments to individual patients.”

̽��ֱ��scanner will begin operating at the start of December, with research projects lined up to look at dementias caused by changes to the brain almost undetectable by conventional scanners, and to look at how visual and sound information is converted to mental representations in the brain.

PET/MR scanner

̽��ֱ��new GE Healthcare PET/MR scanner brings together two existing technologies: positron emission tomography (PET), which enables researchers to visualise cellular activity and metabolism, and magnetic resonance (MR), which is used to image soft tissue for structural and functional details.

Purchased as part of the Dementias Platform UK, a network of imaging centres across the UK, the scanner will enable researchers to simultaneously collect information on physiological and disease-related processes in the body, reducing the need for patients to return for multiple scans. This will be particularly important for dementia patients.

Professor Fiona Gilbert, who will lead research on the PET/MR scanner, explains: “Dementia patients are often frail, which can present challenges when they need separate PET and MR scanners. So, not only will this new scanner provide us with valuable information to help improve understanding and diagnosis of dementia, it will also be much more patient-friendly.”

PET/MR will allow researchers to see early molecular changes in the brain, accurately map them onto structural brain images and follow their progression as disease develops or worsens. This could enable researchers to diagnose dementia before any symptoms have arisen and to understand which treatments may best halt or slow the disease.

As well as being used for dementia research, the scanner will also be applied to cancer research, says Professor Gilbert.

“At the moment, we have to make lots of assumptions about what’s going on in tumour cells. We can take biopsies and look at the different cell types, how aggressive they are, their genetic structure and so on, but we can only guess what’s happening to a tumour at a functional level. Functional information is important for helping us determine how best to treat the cancer – and hence how we can personalise treatment for a particular patient. Using PET/MR, we can get real-time information for that patient’s specific tumour and not have to assume it is behaving in the same way as the last hundred tumours we’ve seen.”

̽��ֱ��PET/MR scanner will begin operation at the start of November, when it will initially be used to study oxygen levels and blood flow in the tumours of breast cancer patients and in studies of brain inflammation in patients with Alzheimer’s disease and depression.

Hyperpolarizer

̽��ֱ��third new piece of imaging equipment to be installed is a GE Healthcare hyperpolarizer, which is already up and running at the facility.

MRI relies on the interaction of strong magnetic fields with a property of atomic nuclei known as ‘spin’. By looking at how these spins differ in the presence of magnetic field gradients applied across the body, scientists are able to build up three-dimensional images of tissues. ̽��ֱ��hyperpolarizer boosts the ‘spin’ signal from tracers injected into the tissue, making the MRI measurement much more sensitive and allowing imaging of the biochemistry of the tissue as well as its anatomy.

“Because of underlying genetic changes in a tumour, not all patients respond in the same way to the same treatment,” explains Professor Kevin Brindle, who leads research using the hyperpolarizer. “Using hyperpolarisation and MRI, we can potentially tell whether a drug is working, from changes in the tumour’s biochemistry, within a few hours of starting treatment. If it’s working you continue, if not you change the treatment.”

̽��ֱ��next generation of imaging technology, newly installed at the ̽��ֱ�� of Cambridge, will give researchers an unprecedented view of the human body – in particular of the myriad connections within our brains and of tumours as they grow and respond to treatment – and could pave the way for development of treatments personalised for individual patients.

By bringing together these scanners, the research expertise in Cambridge, and the latest in ‘big data’ informatics, we will be able to do sophisticated analyses that could revolutionise our understanding of the brain – and how mental health disorders and dementias arise – as well of cancers and how we treat them

Ed Bullmore

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. For image use please see separate credits above.

Yes