The artificial intelligence-based innovations that CERN initially developed to improve the maintenance of its particle accelerator have revolutionary applications in the field of health. Reducing the size of radiotherapy machines and optimizing them to facilitate their use – especially in countries with fewer resources -, designing an intelligent breast cancer prevention program or improving the monitoring of stroke patients are some of the promising projects in the which the European nuclear research laboratory works, in collaboration with European hospitals, including the Vall d’Hebron in Barcelona.
CERN offers hospitals its experience in managing enormous amounts of data in a secure and decentralized way, key to guaranteeing the privacy and security of patients’ private information that is used to feed the algorithm. This institution, whose main field is particle physics, works on its data with a system that avoids the transfer of information to central storage, performing the processing locally to guarantee privacy and optimize resources when different hospitals collaboratively create reliable models. AI-based disease analysis and prediction.
“It is a new paradigm. Before, there was a large amount of data that was centralized and stored. Now we do the processing at the place where the data is acquired, for example a hospital. If we guarantee privacy, data protection and the robustness of the model, we have something that is of great interest for medical applications,” explains CERN scientist Luigi Serio during a visit to the laboratory facilities organized by the International Union for Cancer Control (UICC) on the occasion of the World Cancer Congress in Geneva (Switzerland).
Luigi Serio, responsible for artificial intelligence developments applied to the field of health, admits that the laboratory that discovered antimatter—or where the web was born—is not the only one that uses this information processing model: “But we offer the robustness , the powers of CERN and the fact that we can provide the service guaranteed by the good name and non-profit as a selfless organization that we are,” Serio clarifies.
This is how AI helps improve stroke monitoring
One of the applications based on this system is Truckstroke, which already makes it possible to improve stroke treatment with artificial intelligence in some 10,000 patients in hospitals in Germany and Belgium and in the Vall d’Hebron Stroke Unit in Barcelona.
By comparing the brain images of the patient affected by a stroke with the models trained by CERN in the so-called “Truststroke Project”, the algorithm predicts how the patient could evolve, what therapy should be administered and the follow-up required. More importantly, the tool predicts the risk of recurrence.
Every year 1.1 million people in Europe suffer a stroke, half a million die and there are also almost 10 million survivors who doctors have to care for in the long term. “Professionals are overwhelmed by stroke patients and need increasingly new tools to support their work,” clarifies Luigi Serio.
Hospitals have all the data configured locally, but by exchanging the parameters with the main server they obtain prediction models capable of measuring the severity of the stroke. “The doctor can use these models to decide what type of therapy to administer to the patient. You also know the probable outcome and the follow-up required, how long you should stay in the hospital, when you can be discharged, etc.,” explains the researcher.
The algorithm knows who should undergo a mammogram
CERN plans to have a cancer detection program completed next year that promises to be 50% more accurate than the screening model currently used, the GAIL. In addition to age factors and clinical history, the CERN model will determine the risks of having breast cancer by combining multiple factors, for example the consumption of certain foods or alcohol, lifestyle and physical activity, the age of the woman at her first pregnancy or menopause, among other parameters.
The current screening system does not have an approach based on all risk factors. “The idea is to have a tool that can analyze several factors beyond those currently taken into account to decide whether a mammogram should be done earlier, including whether it can be delayed and for what reasons,” explains Luigi Serio.
The data to train the tool comes from EPIC (European Prospective Study on Diet, Cancer and Health)which contains information collected over more than 20 years. Once the model is finalized next year, it will have to be tested and regulated, so there are still steps left for the promising breast cancer screening system to replace the current protocol.
CERN wants to improve radiotherapy linear accelerators (LINAC) with artificial intelligence to simplify their use and adapt them to the environments of low- and middle-income countries, where it is difficult to access a machine not only because of its high cost but also because there is a lack of people. that they know how to use them. “The machines are difficult to acquire, install, operate and maintain,” explains Luigi Serio. The use of artificial intelligence in this field could provide efficient quality of care because, he explains, the machine can be used and the diagnosis made even if there is no expert.
With new AI-based software, failures can be predicted, maintenance streamlined and also guide those who use the machines, in addition to reducing the downtime of radiotherapy facilities, which are now sometimes out of use due to lack of of the staff who know how to use them. This model could even open the door to automating treatment planning.
The project, called STELLA, is initially intended to improve radiotherapy treatment in some African countries, where there is one radiotherapy device for every 3.5 million people, compared to one for every 80,000 to 100,000 people in Africa. USA and most European countries.
Predict the evolution of tumors or Alzheimer’s
A medical application developed by CERN is capable of determining the defects, anomalies or pathologies that the brain has and indicating to doctors at what exact point a pathology, for example a tumor, could be developing, thanks to a complex system based on which This institution was created to prevent failures in the operation of the particle accelerator.
“Interestingly, the brain is a complex system that can be modeled as a graph. You have neurons in different parts of the brain that are connected to each other, and you can establish a matrix of nodes and vectors that connect the different parts,” explains the CERN researcher. By processing brain images obtained through MRI, the algorithm is able to detect with a certain precision where there could be pathology.
“The algorithm would extract the image saying there is some irregularity, and it can even actually predict where the anomaly is and where it propagates,” he explains. This technology is being clinically tested at Kapodistrian University Hospital (Greece). For the moment, explains Luigi Serio, they have been used for the segmentation of tumors or strokes, but CERN plans to use this system to monitor the evolution of Alzheimer’s or dementia.
