AI enhances low-cost MRI scanner: faster, more patient-friendly, and sustainable
Amsterdam UMC researcher Daisy van den Berg has developed an algorithm that reduces the scan time of low-field MRI scanners by 60 percent. This breakthrough was made possible through a public-private partnership between Amsterdam UMC and Esaote, an Italian manufacturer of orthopedic MRI systems.
IXA mediated between the researchers and the private party, thus representing the interests of both the hospital and the patient.
Low-field MRI scanners use a weaker magnetic field, making them more affordable and energy-efficient—an advantage for hospitals with limited resources. However, these scanners typically require longer scan times compared to high-field systems. Thanks to the new AI-powered algorithm, that drawback is now significantly reduced, making the technology faster, more patient-friendly, and more sustainable for clinical use.
Faster and improved image quality
Within Amsterdam UMC, research on speeding up MRI has been running for a long time in Gustav Strijkers’ group. ‘In recent years, we have developed an AI module that is very effective in speeding up MRI,’ says Matthan Caan, associate professor in the Department of Biomedical Engineering & Physics. ‘We have applied this widely to MRI scans at higher field strengths.’
To answer the manufacturer’s question, researcher Daisy van den Berg has now developed an algorithm that can reduce the scan time of an MRI scanner with low magnetic field strength by 60 per cent. “For example, a patient does not have to lie still for 5 minutes anymore, but now only 2 minutes per scan. Especially for low field scanners, our algorithm appears to work very well. Besides speeding up, in some cases an improvement in image quality can also be seen,” says Van den Berg. This specifically concerns MRI scanners used for orthopedic applications, for example to picture bone fractures of patients. Among other things, a modified design allows the knee or shoulder to be scanned in different positions, and the spine can be imaged both lying down and standing.
Public-private partnership
Ruben Boyd, who works at IXA-Amsterdam UMC, supports Amsterdam UMC researchers in finding out how their knowledge or invention can best be used more widely. The AI module to speed up MRI scan time had already been developed here. But Amsterdam UMC cannot market an MRI scanner itself to ensure that patients actually benefit from it. Therefore, a public-private partnership was set up with the manufacturer of low-field MRI scanners. IXA mediated between the researchers and the private party, thus representing the interests of both the hospital and the patient.