England’s latest imaging update combines three changes: additional CT and MRI machines, more community diagnostic scans and wider use of software that accelerates MRI acquisition. Each has a different denominator. Reading them as one measure would overstate what the figures establish. The government’s 3 November update
The software’s role is to help acquire images with less data. That is different from AI that searches an image for a possible abnormality.
Equipment and activity are separate counts
The government reports 125 additional CT and MRI scanners between March 2024 and March 2026, including 107 in community diagnostic centres. That is approximately 86% of the additional machines, calculated as 107 divided by 125.
The update also describes nearly 2.2 million CT and MRI scans at those centres in the year to March 2026, around a million more than in the year to March 2024. The first figure counts equipment; the second counts examinations. Neither identifies how many unique people were scanned.
Those counts can describe expansion without isolating which part of the activity change came from a new machine, software, staffing or another operational change.
Compatible scanners are the installation denominator
The release says acceleration software has been installed on 54% of compatible NHS MRI scanners. “Compatible” matters: this is not a claim about 54% of all imaging equipment or all NHS scans.
The government reports an average of 14% more scans in services using the technology. It also projects up to 162,000 additional annual checks from a full rollout, with trusts offered the opportunity to upgrade compatible equipment by 2029.
An observed average, an installation share and a future capacity estimate should retain their different meanings. The projection is not a count of additional checks already delivered.
Faster acquisition is not faster diagnosis by definition
A shorter acquisition can create room in a scanner schedule. The national announcement does not independently establish how quickly every image is reported or how long an individual waits for the next step in care.
Our prostate MRI AI report examines a separate detection task. This update concerns acquisition and capacity. Its figures are useful when their scope stays visible; they cannot provide a single causal estimate of AI’s effect on the diagnostic pathway.