An international team led by professionals University of Cambridge in the UK and Huazhong University of Science and Technology in China Second hand A technique called federal learning to create a new model that would allow the presence of the COVID-19 virus to be detected through artificial intelligence (AI) practices.With federal learning, an AI model can be applied in one hospital or country, and you can independently verify with a dataset from another hospital or country without sharing any information.
The researchers based their model on more than 9,000 CT scans from nearly 3,300 patients in 23 hospitals in the UK and China. Your search resultsPublished in the magazine The Intelligence of the Machine of NaturAnd They provide a framework in which AI technologies can be more reliable and accurate, especially in areas like medical diagnostics where privacy is vital.
Artificial intelligence diagnosis has provided a promising solution to accelerate the outcome of COVID-19 and future public health crises. but still, Safety and reliability concerns prevent the large-scale collection of representative medical data, Which is a challenge to train a model that can be used worldwide.
In the early days of the pandemic, many AI researchers worked to develop models that could diagnose the disease. However, many of them were created with poor quality data, incomplete or incomplete datasets, and a lack of information from clinicians. Several researchers in the current study note that these older models were not suitable for clinical use during the 2021 pandemic season.
“Artificial intelligence has many limitations in diagnosing COVID-19, and we have to carefully analyze and select the data until we end up with a model that works and is reliable.”First co-author Hanshin Wang from Cambridge from the Department of Engineering at the University of Cambridge explained. And added: “When previous models were based on arbitrary open source data, we worked with a large team of NHS radiologists and the Wuhan Tongji Hospital Group to identify the data, so we started from a strong position.”.
The researchers used two selected, reasonably large sets of external validation information to test their model and ensure that it worked well with data sets from different hospitals or countries. “Before COVID-19, people didn’t know how much data they needed to collect to create medical AI applications,” noted co-author Michael Roberts of AstraZeneca, responsible for Cambridge’s Department of Applied Mathematics and Theoretical Physics.
“Different hospitals, different countries, they all have their own procedures, so the data sets need to be as large as possible to provide useful follow-up for all kinds of doctors.”
The researchers based their framework on 3D computed tomography rather than 2D images. CT scans offer a much higher level of detail, which results in a better model.They used 9,573 CT scans from 3,336 patients from 23 hospitals in China and the UK.
What’s more They needed to mitigate bias from different data sets and used standardized learning to train a more generalized AI model while keeping each data center private in a collaborative environment. For a fair comparison The researchers validated all models with the same data without disrupting the training data. The team had a panel of radiologists who made diagnostic predictions based on the same CT scans and compared the accuracy of the AI models with that obtained from the specialists.
The researchers say their model is useful not just for COVID-19, but for any other disease that can be diagnosed with a CT scan. “Next time there’s a pandemic, and there is every reason to believe there will be, we’ll be in a much better position to use AI technology quickly so we can understand new diseases faster” said Wang.
“We have shown that medical information can be encrypted so that we can create and use these tools while protecting patient privacy across internal and external boundaries. By working with other countries, we can do a lot more than we can alone, ”added Roberts.
Researchers are currently working with the newly established WHO Center for Epidemiological Intelligence and Epidemiology to explore the potential for privacy enhancing digital health frameworks.
You are developing a global system with artificial intelligence to identify COVID-19 variants
They were able to identify COVID-19 cases using AI applied to speech analysis
Artificial intelligence and COVID-19: can machines save us?