A hospital buys an imaging system the way it buys a building: as a capital asset with a long life, financed over years, justified by a utilisation forecast that runs a decade out. The machine is built to last that long and generally does.
The computer driving it is not built to last that long, and neither is the manufacturer's commitment to it. Support windows for the embedded operating system frequently expire well before the device reaches the end of its depreciation schedule, and at that point the hospital owns an expensive, functional, unpatchable machine.
Replacing early is a capital decision nobody can fund
The obvious answer is to replace the device, and the obvious answer collides with how hospitals are financed. Capital budgets are set years ahead and allocated against clinical need, so scrapping a working scanner because its software is out of support means either cancelling something else or explaining to a board why a machine that works must go.
Upgrading the software alone is usually not available. Medical devices are certified as a whole system, and modifying the operating environment can invalidate the certification, which turns a patching decision into a regulatory one. The manufacturer will often sell an upgrade path, and it is priced as a substantial fraction of a new device.
This is the same trap running through the machines on the factory floor, with the difference that the asset here is attached to a patient and sits on a network carrying clinical records. The mitigations are the same and the stakes are not: segment the device so it cannot reach the internet or the general network, control removable media, monitor what it talks to, and plan for recovery on the assumption that prevention will eventually fail.
The insurer is increasingly the party forcing the issue, which is the pattern now familiar wherever cyber insurers have become the de facto regulators of corporate security. Renewal questionnaires ask specifically about unsupported clinical devices, and the answer moves the premium — which converts an abstract risk into a line item a finance officer will act on, in a way that a security memo never did.
Where this lands hardest is where every capital constraint lands hardest. A large system can stagger replacement across a fleet and negotiate support extensions at volume. A small or rural hospital owns one of each, cannot spread the cost, and is choosing between an unsupported scanner and no scanner — which is the same arithmetic that decides what a rural hospital closes first.
