Using AI to safely detect cancer from patient data – Artificial intelligence and robotics news

A new way to use artificial intelligence to predict cancer from patient data without putting personal information at risk has been developed by a team including medical scientists from the University of Leeds.

Artificial intelligence (AI) can analyze large amounts of data, such as images or test results, and can identify patterns often undetectable by humans, making it very useful for speeding up detection, diagnosis and treatment of diseases.

However, the use of technology in medical settings is controversial due to the risk of inadvertent data disclosure and many systems are owned and controlled by private companies, giving them access to confidential patient data – and the responsibility for keeping it. protect.

The researchers sought to find out if a form of AI, called swarm learning, could be used to help computers predict cancer in medical images of patient tissue samples, without disclosing hospital data.

Swarm learning trains AI algorithms to detect patterns in data from a local hospital or university, such as genetic changes in images of human tissue. The swarm learning system then sends this newly trained algorithm – but mostly no local data or patient information – to a central computer. There it is combined with algorithms generated by other hospitals in the same way to create an optimized algorithm. This is then sent back to the local hospital, where it is reapplied to the original data, improving the detection of genetic changes through its more sensitive detection capabilities.

By undertaking this multiple times, the algorithm can be improved and create one that works on all data sets. This means that the technique can be applied without the need to disclose data to third-party companies or send it between hospitals or across international borders.

The team trained AI algorithms on study data from three groups of patients from Northern Ireland, Germany and the United States. The algorithms were tested on two large sets of data images generated in Leeds and found to have successfully learned how to predict the presence of different cancer subtypes in the images.

The research was led by Jakob Nikolas Kather, visiting associate professor at the University of Leeds Medical School and researcher at RWTH Aachen University Hospital. The team included Professors Heike Grabsch and Phil Quirke, as well as Dr Nick West from the University of Leeds Medical School.

Dr Kather said: “Based on data from over 5,000 patients, we were able to show that AI models trained with swarm learning can predict clinically relevant genetic changes directly from images of tissues from colon tumors. »

Phil Quirke, Professor of Pathology at the University of Leeds Medical School, said: “We have shown that swarm learning can be used in medicine to train independent AI algorithms for any analytical task. pictures. This means that the need for data transfer can be overcome without institutions having to give up secure control of their data.

“Creating an AI system that can perform this task improves our ability to apply AI in the future. »

Source of the story:

Materials provided by University of Leeds. Note: Content may be edited for style and length.

We would love to give thanks to the author of this article for this incredible content

Using AI to safely detect cancer from patient data – Artificial intelligence and robotics news

Explore our social media accounts and other pages related to them