#363 ‒ A new frontier in neurosurgery: restoring brain function with brain-computer interfaces, advancing glioblastoma care, and new hope for devastating brain diseases | Edward Chang, M.D.
Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
Podcast Summary
In this fascinating episode, Dr. Edward Chang takes listeners inside the cutting-edge world of modern neurosurgery and brain-computer interfaces. From Harvey Cushing's revolutionary techniques a century ago to today's game-changing awake brain surgeries, Chang reveals how neurosurgery has evolved into a precision discipline that maps neural circuits in real-time (19:38). The conversation explores groundbreaking research with patients like Ann, who after 18 years of paralysis from a brainstem stroke, achieved remarkable communication through brain-computer interfaces that decode speech intentions at 80 words per minute (61:30). Chang outlines his vision for 2030 and beyond: fully implantable wireless brain devices that could transform conditions like ALS, spinal cord injuries, and aggressive brain tumors into manageable chronic illnesses, representing a fundamental shift from traditional pharmaceutical approaches to direct neural engineering solutions (94:54).
Speakers
Dr. Edward Chang
Chair of neurosurgery at UCSF and leading innovator in functional neurosurgery and brain-computer interfaces. His groundbreaking work bridges the operating room, research lab, and engineering bench to restore speech and movement for patients with paralysis and neurological conditions.
Peter Attia (Host)
Physician and host of The Drive podcast, focusing on translating longevity science into accessible content. Former surgical resident with expertise in analyzing complex medical interventions and their implications for human health and performance.
Key Takeaways
Master the Art of Direct Recording
Move beyond surface-level EEG to electrocorticography (ECOG) for 1,000x better resolution. Place sensors directly on the brain cortex to extract meaningful signals from 86 billion neurons. (55:56) Surface recordings through skull lose critical fidelity - breakthrough insights require getting as close to the source as possible.
Engineer Biology as Your Next Technology Stack
The future isn't just smaller electrodes - it's engineered cells interfacing with the brain instead of metal. Biology has already solved scaling problems with specialized cells from identical genetic programming. (92:18) Think organoids, stem cell therapies, and biological computing that operates at the scale of 86 billion neurons.
Build AI-Native Decoding Algorithms
Create systems that translate brain activity to speech units in 10-20 millisecond chunks, using statistical phonemes derived from Meta's speech recognition algorithms. (80:45) Combine this with language models for probabilistic inference - achieving 80 words per minute from thought to text in real patients.
Embrace Volitional Intent as Your Core Signal
The breakthrough isn't reading thoughts - it's capturing the brain's attempt to execute motor commands. Ann had to actively try to speak for 18 years post-stroke to generate decodable signals. (73:39) Focus on the motor cortex areas controlling lips, jaw, larynx - not the reading or inner monologue regions.
Solve for Chronic, Not Perfect
Transform devastating conditions like glioblastoma from 18-month death sentences into chronic manageable diseases. Use genetic profiling to target specific mutations rather than generic chemotherapy approaches. (12:33) Optimize for years of meaningful life rather than pursuing perfect cures that may never come.
Statistics & Facts
- 86,000,000,000 (86 billion) neurons exist in the human brain, a precise count that highlights the complexity of the organ where brain-computer interfaces must operate. (01:48)
- Ann achieved approximately 80 words per minute speech output through ECOG brain-computer interface technology, compared to the normal conversational rate of 150-160 words per minute between speakers. (85:53)
- 253 ECOG sensors were implanted in a credit card-sized array, spaced 3 millimeters apart with each sensor measuring 1 millimeter in diameter, to decode speech from Ann's motor cortex. (78:33)
