Siemens stopped trying to muffle the terrifying noise of an MRI machine and turned it into a robot's voice in a children's audiobook instead

问题

A frightened child who won't hold still ruins an MRI scan, and sedating them is slow, costly and carries its own risk

背景

换别人会怎么做

The standard hospital response to a frightening MRI is to work around the noise rather than change it: decorate the scanner room, hand the child noise-cancelling headphones, have a parent talk them through it, play a cartoon beforehand, or in the worst cases sedate the child outright. Every one of these treats the machine's sound as a problem to suppress or distract from, but none of them change what the child actually hears the instant the scan starts — the same unexplained banging and whirring that made them afraid in the first place, now happening despite the distraction rather than because of it.

他们看到了什么

Siemens and AREA 23 saw that MRI noise isn't random, it's mechanically predictable — the same scan type produces the same sequence of clunks, whirs, and beeps every time, precisely because the machine is following a fixed physical process. That predictability meant the noise itself could be scored like a piece of music and written into a story in advance, turning the exact sounds a child was afraid of into sound effects they were already listening for — a robot's footsteps, a spaceship engine — so the machine's real behavior became the plot instead of an unexplained threat interrupting a distraction.

那一手

Siemens Healthineers and agency AREA 23 built 'Magnetic Stories': audiobooks, written with bestselling children's authors, in which the MRI machine's actual mechanical sounds are woven directly into the plot as sound effects timed to the scanner's real noise sequence — the loud banging becomes a friendly robot's footsteps or a spaceship launching, so the child hears the same sounds the machine always made, but now as part of a story they're following rather than an unexplained threat.

为什么管用

By treating the MRI's mechanical sequence as fixed material to compose around rather than noise to mask, the team could map real recorded scan sounds (cranial, liver, spine, each with a different rhythm) to specific narrative beats, so the child's actual, real-time sensory experience became evidence that the story was true rather than a contradiction undermining a comforting lie. This is the crucial difference from noise-cancelling headphones or a cartoon playing beforehand: those methods leave a gap between what the child was told to expect and what they actually experience the moment the gradients start firing, and children are especially quick to distrust reassurance that gets contradicted by their own senses. Because the story and the noise are the same event, holding still and listening through the story is also holding still through the diagnostically necessary scan, aligning what the child wants to do (keep following an exciting story) with what the hospital needs them to do (stay still) — which is why radiology staff at the Portuguese pilot hospitals reported a drop in anxiety and reduced need to fall back on sedation, even without a published peer-reviewed clinical trial to confirm the effect size.

值了多少

The program launched as a pilot across Portugal's CUF private hospital network, where radiology staff reported a marked drop in child anxiety during scans and reduced need to fall back on sedation — Siemens has not published a peer-reviewed clinical figure for the reduction as of this writing, but the pilot's reception was strong enough that the company moved to expand Magnetic Stories into Spain, Germany, Italy and the UK, with plans for real-time AI-generated narration that can sync to any hospital's specific scan sequence.

什么时候会失灵

This mechanism depends on the unpleasant stimulus being mechanically predictable enough to score in advance — an MRI's fixed noise sequence per scan type is unusually well-suited to this, but a stimulus with irregular or patient-specific timing (a variable-length injection, an unpredictable procedural pain) can't be synced to a fixed narrative the same way, at least not without real-time adaptive narration keeping pace with the actual event as it unfolds. It also depends on the story genuinely holding the patient's attention through the full duration of the stimulus, since a bored or disengaged child returns to experiencing the noise as raw noise; a poorly written or age-mismatched story would fail exactly where a generic distraction already fails. And because this pilot's reported impact rests on radiology staff observation rather than a published clinical trial, this should be read as a strong candidate for study, not yet as clinically proven — a category error more efficacy-desperate imitators are prone to committing once impressed by a Cannes Grand Prix rather than by controlled evidence.

后来呢

Magnetic Stories won the Pharma Grand Prix at Cannes Lions 2024 along with multiple Clio Awards and a D&AD shortlist, and AREA 23 was named Cannes' Healthcare Agency of the Year on the strength of the campaign — it is now cited in pediatric-radiology and healthcare-marketing circles as a reference case for redesigning the patient experience around an unavoidable clinical stimulus rather than working around it.

资料来源

1. [1]Siemens wins Pharma Grand Prix with comforting children's audiobooks for scary MRIsAd Age, 2024adage.com
2. [2]Cannes Lions 2024: Pharma winnersContagious, 2024contagious.com

收下它

同一路聪明

• 2010 · American ExpressAmex didn't advertise its card on the biggest shopping weekend of the year — it invented a whole new day inside it
• 1962 · AvisAvis told customers outright it was only #2, and the confession made everything else it said believable
• 2025 · AXA France (with Publicis Conseil)AXA covered domestic violence like a house fire by changing three words in a contract millions of people already had