Skip to content

Synchron Stentrode

Synchron Stentrode is the minimally invasive alternative pathway for BCI — placing electrodes through blood vessels and avoiding craniotomy. By 2024 it has been operating long-term in 6 patients and is the safest and fastest-to-market BCI approach. Synchron's partnerships with OpenAI and Apple foreshadow a future in which consumer-grade speech/handwriting BCI may reach the market before Neuralink.

1. The Stentrode Device

Structure

  • A vascular stent serves as the electrode carrier
  • 16 electrodes are anchored on the mesh structure
  • Placed in the superior sagittal sinus, adjacent to M1
  • Inserted via the jugular vein

Surgery

  • Endovascular procedure: jugular catheter → superior sagittal sinus
  • No craniotomy required
  • 1–2 hours, local anesthesia
  • Rapid postoperative recovery

Wireless Data

  • BrainOS wireless module implanted in the chest
  • Bluetooth transmission
  • No exposed cabling

2. Why the Endovascular Route

Anatomy

  • The brain surface is covered by a layer of venous sinuses
  • The superior sagittal sinus lies directly above M1 / SMA
  • Intravascular signal ≈ epidural ECoG

Advantages

  • No brain-tissue injury
  • Surgical safety approaches that of stent placement (already routine)
  • Interfaces naturally with vascular surgery

Disadvantages

  • Lower signal quality (farther from neurons)
  • Few channels (16)
  • Coarse spatial resolution

3. The COMMAND Clinical Trial

Launched in 2021

  • FDA IDE approval (2021)
  • Multi-center (Mount Sinai, U Pittsburgh, Sydney)
  • Target: 6 patients, 1-year follow-up

Inclusion

  • ALS, spinal cord injury
  • Loss of arm function
  • Cognition intact

First Patient: Philip O'Keefe (Australia 2021)

  • ALS patient
  • Sent the first "brain tweet" in 2021
  • World's first BCI social-media post
  • Has continued to use it for 3+ years

First U.S. Patient (2022): Rodney Gorham

  • Also ALS
  • Mount Sinai
  • Day-to-day computer control

2024: All 6 Patients Enrolled

  • UK, USA, Australia
  • Integrated results pending publication

4. Capability Demonstrations

Basic Control

  • Mouse cursor
  • Clicks (imagined hand movements)
  • Texting, email
  • Web browsing

Speed

  • ~10–15 clicks per minute
  • Slower than Neuralink but usable day-to-day

Reliability

  • Stable long-term (3+ years)
  • This is Stentrode's strongest advantage
  • Neuralink has not yet reached this level of longitudinal validation

5. Synchron × OpenAI Collaboration

2023 Announcement

  • Synchron integrated ChatGPT into the BCI
  • Brain control + LLM dialogue augmentation
  • The LLM helps complete what the user wants to say

Implementation

  • BCI decoding → intent
  • Intent → OpenAI API → generated text
  • User confirms by selection

Significance

  • LLMs fill the bandwidth gap of low-throughput BCIs
  • Stentrode's 16 channels are too few for fast typing, but enough to select intents
  • See LLM post-processing fusion

6. Synchron × Apple Collaboration

2024 News

  • Synchron integrating with Apple Vision Pro
  • BCI control of visionOS
  • AR / VR made accessible to paralyzed patients

Goal

  • Consumer AR + BCI
  • Medical first, then expansion

7. Synchron's Commercial Strategy

Medical Focus

  • Avoids Musk-style hype
  • Quiet progress with physicians and patients
  • FDA PMA expected 2027

Cooperate Rather than Monopolize

  • Partnering with Apple and OpenAI
  • "BCI as interface" rather than "BCI as operating system"
  • Positioning analogous to a USB protocol

Financing

  • Raised ~$145 M to date
  • Gates Foundation, Khosla, Bezos among investors
  • Valuation ~$1B

8. Technical Comparison

Neuralink Synchron
Channels 1024 16
Invasion Intra-cortical Intravascular
Surgery Complex robotic Endovascular
Signal Single-unit LFP-like
Speed Fast Slow
Safety Research Close to approval bar
Commercial 2027+ 2026+

vs Precision

See Precision_Paradromics_Blackrock.

9. Future Roadmap

1. More Channels

  • Next-generation Stentrode: 32 → 64 channels
  • Multi-stent concurrent implants
  • Still far below Neuralink

2. Stimulation Capability

  • Read-only at present
  • Bidirectional versions under study (ICMS from inside the vessel?)

3. Consumer Grade

  • Combine with LLMs → low channel count can still deliver fast intent
  • Consumer partners such as Apple
  • After medical approval → expansion into consumer health

4. Global Expansion

  • European and Asian markets
  • Expected price $25,000–50,000 (cheaper than surgical BCIs)

10. Indication Expansion

Near term (2026–2028)

  • ALS
  • Spinal cord injury
  • Locked-in syndrome

Medium term (2028–2030)

  • Stroke (partial motor function)
  • Multiple sclerosis
  • Severe frailty in the elderly

Long term (2030+)

  • Consumer-grade healthy users
  • AR / VR interfaces
  • Workplace assistance

11. Limitations

1. Signal Quality

  • Intravascular placement is several millimeters from neurons
  • Good temporal resolution, poor spatial resolution
  • Complex tasks (high-DOF motor, speech) lag behind cortical recordings

2. Vascular Constraints

  • Electrode location is constrained by vascular anatomy
  • Cannot precisely target a chosen brain region
  • Some regions are unreachable

3. Thrombosis Risk

  • A foreign body inside a vessel → thrombosis
  • Long-term anticoagulant therapy
  • Drug side effects

4. Scale

  • Only 6 patients today — small
  • Neuralink expanded faster in 2024

12. Logic Chain

  1. Stentrode is implanted via blood vessels, skipping craniotomy — the safest BCI pathway.
  2. 16 channels but long-term stability — validated 3+ years without issue.
  3. The COMMAND trial has enrolled all 6 patients across the U.S., UK, and Australia.
  4. Partnership with OpenAI uses LLMs to compensate low bandwidth → practical utility.
  5. Partnership with Apple points to AR/VR + BCI consumer directions.
  6. Commercialization expected 2026+, ahead of Neuralink.
  7. Limitations: signal quality, vascular constraints, thrombosis risk — but the safety advantage is overwhelming.

References

  • Mitchell et al. (2023). Assessment of safety of a fully implanted endovascular brain-computer interface for severe paralysis in 4 patients: the Stentrode with Thought-Controlled Digital Switch (SWITCH) study. JAMA Neurology.
  • Oxley et al. (2021). Motor neuroprosthesis implanted with neurointerventional surgery improves capacity for activities of daily living tasks in severe paralysis: first in-human experience. J NeuroInterv Surg.
  • Synchron (2023). Press release on ChatGPT integration.
  • Opie et al. (2018). Focal stimulation of the sheep motor cortex with a chronically implanted minimally invasive electrode array mounted on an endovascular stent. Nat Biomed Eng.
  • clinicaltrials.gov: NCT03834857 (SWITCH), NCT05035082 (COMMAND).

评论 #