Medical device giant Philips made a slew of announcements this week during the Radiological Society of North America’s annual conference in Chicago — most notably the launch of a new CT machine and the development of a new helium-free MRI system.
Shez Partovi, Philips’ chief innovation and strategy officer, said that the company’s new CT system, named Verida, is the world’s first spectral CT system to be fully powered by AI. The CT scanner uses spectral detectors that capture multiple X-ray energy levels in every scan, and that AI is built directly into the imaging chain to improve clarity and reduce noise, he explained.
In his view, the key innovation is that AI is built directly into the CT detector hardware, not just applied to images after they’re formed.
“As soon as the X-ray beam goes to the patient, the AI is saying, ‘How do I optimize the signal and take away the noise and the graininess?’ That is the heart of this machine, which is why we say its AI-powered end-to-end,” Partovi remarked.
This end-to-end design means AI is involved in every step — from signal capture to patient positioning to correcting for motion, such as a beating heart — he added.
Philips plans to make the new machine commercially available in select markets in 2026.
One of the other notable machines the company was displaying during the conference was the BlueSeal Horizon MRI system, an upgrade from last year’s BlueSeal MRI model. Unlike the 1.5 T version unveiled in 2024, Horizon is a 3 T system, which means it offers double the magnetic field strength. This means it can produce more detailed images, as well as enable scanning that is up to three times faster, Partovi said.
He also noted that this model — which is still in development — is helium-free, meaning its magnet is permanently sealed and does not require costly helium refills or a quench pipe. This reduces operating costs, as well as the complexity of installing the scanner in a hospital.
Traditional MRI magnets typically require nearly 400 gallons of liquid helium to effectively cool the magnet’s coils and therefore allow them to maintain superconductivity. Partovi pointed out that this makes MRI machines very heavy — which is why most of them are on the first floor or basement of the hospital.
“Now they can be anywhere,” he declared.
And just like the AI in the Verida CT machine, BlueSeal Horizon embeds AI directly into the signal acquisition stage of the MRI, not just the image. For better image quality, the AI is designed to optimize raw signals in real-time as the magnets generate signals from the body, Partovi explained.
By integrating AI at the hardware level and tackling operational barriers like helium dependency, Philips is not only seeking to improve image quality but also aiming to make high-quality imaging more accessible and practical for hospitals, he stated.
Photo: Milos Ruzicka, Getty Images
