HongShan Leads Angel Round in Tianfukang, a Peripheral-Nerve BCI Startup Targeting Stroke Recovery
The deal shows peripheral-nerve BCI can secure major VC backing without published human clinical results
HongShan — the firm formerly known as Sequoia China — has led an angel-round financing in Tianfukang Medical Technology (天复康医疗科技), a Beijing-based startup building a wireless closed-loop neuromodulation system for post-stroke upper-limb rehabilitation. The round, also joined by Yuanming Brain Science Fund, was disclosed in late September 2026. HongShan holds a stake in Zhiran Medical, a separately funded invasive cortical BCI company, making the firm one of a small number of Chinese institutional investors now holding positions across more than one BCI architectural paradigm. That dual position suggests institutional investors are beginning to hedge across competing BCI design approaches rather than converging entirely on the cortical implant model that has dominated China's sector headlines.
Tianfukang Targets Peripheral Nerves While China's Other BCI Startups Go Cortical
The majority of China's funded BCI companies — including NeuroXess, StairMed, and Zhiran Medical — place electrode arrays directly on or into motor cortex tissue, intercepting motor commands at the planning stage. Tianfukang intercepts those commands at the execution stage, targeting peripheral nerve bundles between the spinal cord and muscle. Its hardware uses flexible self-coiling electrodes designed to conform to the cylindrical geometry of nerve fascicles, in contrast to the rigid Utah-array probes that cortical systems typically require.
The system operates as a genuine closed loop. A proprietary BCI processor decodes residual motor intent from the peripheral nerve electrophysiology remaining in stroke survivors' affected limbs, then triggers targeted stimulation timed to reinforce the attempted movement. This design draws on the central-peripheral-central (CPC) rehabilitation model, which proposes that stroke recovery depends on co-activating both brain-level motor intention and peripheral execution pathways simultaneously, potentially driving neuroplasticity through reinforcement of simultaneous neural activity. Non-human primate testing has validated the core closed-loop chain from motor intent to peripheral nerve stimulation to movement output; human clinical trials have not yet begun.
The peripheral-nerve approach trades signal resolution for access. Cortical recordings from individual neurons carry richer motor-intent data; peripheral-nerve recordings aggregate activity across many fibers and capture lower-dimensional information. For applications requiring fine multi-finger reconstruction, cortical implants hold an advantage. For restoring functional grasping and reaching in stroke rehabilitation, peripheral-nerve stimulation may be sufficient — and the elimination of craniotomy risk shifts the risk-benefit calculation toward Tianfukang's less invasive design. Tianfukang has also established a joint laboratory with Beijing Tiantan Hospital, one of China's leading neurology institutions, to support product definition and clinical research design.
China's BCI Sector Has Crossed Commercial Approval and Is Moving Into Volume Funding
China's National Medical Products Administration (NMPA) granted the world's first commercial approval for an implantable BCI medical device in March 2026 — the NEO system developed by Neuracle Technology, cleared for patients with cervical spinal cord injury — establishing that the Class III regulatory pathway for invasive BCIs is navigable. Funding has followed at an accelerating pace.
According to IT Juzi data cited by multiple outlets, between 2013 and May 2026, 43 Chinese BCI companies completed 104 financing rounds, raising a total of 9.37 billion yuan. In the first four months of 2026 alone, sector financing already exceeded the full-year total for 2025. The Tianfukang angel round is part of this broader acceleration, not an outlier within it.
China's 15th Five-Year Plan (2026–2030) formally designates brain-computer interfaces as one of six national strategic "future industries," alongside quantum technology, biomanufacturing, hydrogen and fusion energy, embodied intelligence, and 6G. That policy designation provides state-level regulatory priority, R&D support, and capital mobilization infrastructure that reinforces private investment in the sector.
No Clinical Data Yet, and Key Technical Risks Remain
The angel round precedes any published clinical results. Tianfukang has disclosed its architecture, its non-human primate validation results, and its clinical partnership, but no human decoding accuracy figures, stimulation parameters, rehabilitation outcomes, or independent assessments. The round amount was not disclosed. Chronic peripheral-nerve implants also face a tissue-response challenge that cortical BCIs share: electrode encapsulation by scar tissue degrades signal quality over time. Tianfukang's flexible self-coiling design is intended to reduce mechanical mismatch with soft nerve tissue, but long-term human data for this electrode class remains limited relative to the intracortical implant literature.
HongShan's investment in both Zhiran Medical and Tianfukang — two different BCI architectures within a single portfolio — reflects a deliberate hedge rather than architectural conviction. The peripheral-nerve rehabilitation sub-segment now has enough technical credibility to attract the same institutional backing as cortical implants, even without published efficacy data. Whether Tianfukang's closed-loop approach produces measurable rehabilitation gains beyond standard physical therapy will depend on first-in-human safety studies and subsequent trials under NMPA oversight. The more immediate benchmark for China's BCI sector is how quickly the agency processes new implantable neurostimulation applications now that one commercial precedent exists.