A consortium of technology companies comprising Cloudwick, MirrorWeb, and Amazon Web Services EMEA SARL (AWS), is working with PanSurg, an Imperial College London COVID-19 surgical network made up of healthcare professionals and academics from the Department of Surgery and Cancer and the Institute of Global Health Innovation.
The idea is to build a global knowledge platform to help provide the healthcare community with guidance on delivering high quality care to COVID-19 patients.
Dubbed REDASA (REaltime Data Analysis and Synthesis), the platform will pool global data on COVID-19 from over half a million sources and use machine learning and artificial intelligence combined with human curation to extract the most important insights in real-time for clinicians and policymakers.
James Kinross, Clinical Senior Lecturer and lead for PanSurg said that the pandemic has given rise to a huge mass of material on COVID-19 scattered across a vast range of sources. Yet, there is limited guidance and consensus on how to best deliver care for coronavirus patients. REDASA was launched to address these issues.
“Healthcare professionals are facing huge volumes of academic literature, public information and noise on COVID-19, making it challenging to extract key insights and translate these into best clinical practice”, he said.
“We are excited to collaborate with Cloudwick, MirrorWeb and Amazon Web Services to create a reliable, accurate information source with REDASA, for healthcare professionals seeking guidance during the pandemic.”
REDASA is using MirrorWeb’s website capture technology to collect large volumes of data at speed. The organisation, which specialises in archiving and monitoring web channels, applies machine learning models to ensure accurate and credible information is gathered from public data sources such as medical journals, healthcare literature and news sources on a daily basis. The information is then stored and processed on Cloudwick’s data and analytics platform, Amorphic, which facilitates the use of advanced data science techniques to help automate insight generation and provides a secure access model. Both the MirrorWeb technology and the Amorphic platform by Cloudwick are built on AWS.
Using a combination of AWS machine learning services, including Natural Language search and data labelling built on Amazon Kendra and Amazon SageMaker Ground Truth, together with human curation, REDASA performs deep data analysis and quickly extracts the most important insights, which helps teams to make sense of the vast amounts of information pouring in. The platform enables a ‘live systematic review’ so that the information is continuously updated and analysed. The information is provided to private and public sector healthcare organisations and physicians via a secure portal through which they can analyse pandemic data to help develop better treatment plans and accelerate R&D.
Harshdeep Singh, Director of Operations – EMEA, at Cloudwick Technologies UK said: “COVID-19 has made it clear just how challenging it is to find critical information quickly and easily in an ‘infodemic’ situation, with the noise of millions of articles to sift through,” said Dr Matthew Howard, International Healthcare Data Science Lead at Amazon Web Services EMEA SARL. “This solution we are developing with PanSurg, and AWS Partner Network Partners, Cloudwick and MirrorWeb, combines the best of expert human review with AWS machine learning technologies. Our aim is to provide a new approach that will put the most accurate information possible in the hands of healthcare professionals, help improve medical knowledge, and develop more effective methods of patient care that will make a difference to frontline healthcare workers.”
With the help of Cloudwick, MirrorWeb, and AWS, PanSurg has been able to quickly build their testing platform. The team has built a minimum viable product that will launch in the coming months. This will become an important legacy project with applications beyond COVID-19.
The present focus is to support the healthcare community on COVID-19 through REDASA. But if this is successful, Mr Kinross believes this model could instigate sharing healthcare information in this manner more widely, for example in other disease areas such as Oncology.
