A patient settles into a reclined chair in a Wuhan outpatient room. A curved coil is pressed against the left side of the scalp, a technician taps a screen, and the room fills with a sound like a woodpecker hitting a hollow log. Twenty minutes later the patient stands up, books the next session, and leaves. No anaesthetic, no incision, no recovery room.
The coil doing the knocking is not imported. It was designed a few kilometres away, by a company that has been building magnets for Chinese hospitals since before most doctors here had heard the phrase brain–computer interface (脑机接口).
Key takeaways
- In 2025, Wuhan Yiruide's transcranial magnetic stimulation (经颅磁刺激, TMS) device became the first in China to hold a class III medical device registration that names depression as its treated indication.
- The company says it has more than 10,000 of its CCY-type stimulators in service and equipment in over 5,000 hospitals, and it claims the leading domestic share for years running.
- In 2026 its TMS became the first Chinese device of its kind to receive an EU MDR CE certificate, and its newer closed-loop rehab system began treating patients at a Wuhan teaching hospital.
What the licence actually certifies
TMS is not new. A coil fires brief, strong magnetic pulses through the skull and induces a small current in the cortex below; repeated high-frequency pulses are thought to nudge a sluggish mood-regulating network back toward normal activity. The US Food and Drug Administration cleared repetitive TMS for major depression in 2008, and Chinese hospitals have used the machines for over a decade.
That history is why the paperwork matters more than the physics. Yiruide put its first domestically developed stimulator on the Chinese market in 2009, with a registration that allowed clinical use. A class III certificate that explicitly names depression is a different animal: it means the regulator reviewed a multi-centre, randomised controlled trial designed around that one indication. Yiruide says it ran that trial with several tertiary hospitals.
The certificate number is public — 国械注准20253090871 — and the approval was announced at a brain-science and neuromodulation conference in May 2025. What is not public is the trial's effect size. A registration tells you the regulator accepted the evidence; it does not tell you how many patients improved, or by how much.
Two decades, one magnet at a time
Yiruide's origin is academic. The technology came out of Tongji Hospital at Huazhong University of Science and Technology, and the company was set up in 2007 to industrialise it. Early on, hospitals preferred imported machines, and imported machines were expensive to service.
The company's answer was to keep widening the product line — paired-pulse stimulators, deep coils, navigated robotic arms, magnetically induced seizure therapy — until the catalogue covered more of the clinic than any single imported brand did. It says that breadth, plus a service network inside China, pushed its installed base past 10,000 units.
The quieter milestone: a European certificate
The certificate that may matter most commercially arrived in 2026, when the company's TMS received an EU Medical Device Regulation CE mark. According to Hubei's science and technology department, it is the first such certificate issued to a Chinese TMS device.
MDR is not a marketing badge. It requires a clinical evaluation report, a quality system audit and a post-market surveillance plan, and it is the gate for the EU plus a long list of markets that accept CE marking by default — parts of Southeast Asia, the Middle East and Latin America among them. For a category Chinese firms have sold mainly at home, that redraws the map.
From a coil to a loop
The company's newer product is where the brain-computer interface label gets earned rather than borrowed. Its 汉脑·知行 system, launched in March 2026, pairs an EEG cap and near-infrared brain imaging with stimulation hardware and a rehab protocol: the system reads what the brain is doing, decodes intent or state, adjusts stimulation, and feeds the result back to the patient during training.
The system's first clinical deployment came in late May 2026 at Zhongnan Hospital of Wuhan University, in patients recovering from stroke and spinal cord injury. By the end of June, the company says, more than ten patients had been treated, with rehabilitation time cut by more than half compared with conventional passive training.
That is a case series, not a trial. It is also the shape of things to come: the commercial argument for non-invasive neurotech in China is shifting from "we can stimulate" to "we can measure, decide and adjust in real time" — a loop that sounds like an implant story but requires no surgery.
The significance of a domestic TMS supply chain is less about any single certificate than about who supplies the hardware when Chinese hospitals start buying closed-loop rehab systems in volume — a procurement question that will be settled province by province, not by a marquee approval.
The part the certificate cannot fix
China has an estimated 95 million people living with depression, according to figures cited by Hubei Daily. A course of TMS is typically twenty to thirty sessions, each requiring a clinic visit, a trained operator and a working machine — a capacity problem that a licence does not solve.
Response is also uneven. A meaningful share of patients do not improve on TMS, and the treatment is usually positioned alongside medication and therapy rather than instead of them. For a device maker, that is the gap between a certificate and a queue of patients who can actually reach the chair.
What readers can do now
- If you or a family member is considering TMS, ask whether the clinic's device carries a registration naming your indication, and ask what the treatment protocol and session count will be before you start.
- Treat "first in China" claims as a starting point, not a conclusion: check the registration number in the national device database rather than a press release.
- Watch the closed-loop rehab systems, not just the magnets. The interesting question for the next two years is whether measuring the brain in real time actually changes outcomes, or only the sales pitch.
Honest limitations
This article relies on Yiruide's own disclosures, Hubei Daily and Hubei provincial science department reporting, and conference announcements; we did not see the underlying trial report, and the effect size behind the 2025 registration is not public. The installed base of "more than 10,000 units", the "5,000-plus hospitals" and the claim of sustained first-place market share are company figures we could not audit against procurement records. The EU MDR certificate is confirmed by provincial government reporting and the company, but its scope — which models, which indications — was not independently reviewed here. The 汉脑·知行 results cover roughly a dozen patients with no control group, and the "more than half" cut in rehab time is a company characterisation. The 95 million depression estimate is a widely cited prevalence figure, not a measured patient count.