363 ‒ A new frontier in neurosurgery: brain-computer interfaces, new hope for brain diseases, & more

Following Cushing, pioneers like Wilder Penfield advanced the field by mapping functional areas of the brain and developing epilepsy surgery methods. Penfield's characterization of the brain's homunculus—a map representing the body's muscle control over the brain's surface—laid the groundwork for awake brain surgery. Awake procedures, now a standard in certain tumor and epilepsy surgeries, allow surgeons to identify and preserve critical language and motor areas by interacting with patients during the operation. The conversation sets the stage for appreciating how historical insights continue to influence today's cutting-edge practices.

Progress in Neurosurgical Techniques

Chang highlights that although some surgical methods, like the craniotomy, remain largely similar in concept, significant technical and procedural advances have redefined patient care. Procedures that once required large incisions and extensive bone removal are now often replaced by minimally invasive approaches such as laser ablation and focused ultrasound. These innovations allow precise targeting of deep brain structures with minimal trauma, drastically reducing recovery times and complications.

In vascular neurosurgery, the shift from open microscopic dissections for aneurysms to catheter-based endovascular coil embolization exemplifies this transformation. Today, many strokes caused by large vessel clots are treated through mechanical thrombectomy via catheter insertion in the groin, a method that enables patients to potentially recover quickly enough to leave the hospital the next day. This evolution has essentially reclassified strokes as "brain attacks," akin to heart attacks, and intervention is increasingly happening in cath labs alongside cardiac emergencies. Chang articulates that the overall trajectory in surgery favors reducing invasiveness while improving outcomes, reflecting broader trends in medicine.

Glioblastoma Multiforme

The discussion turns poignant when the topic of glioblastoma multiforme (GBM) arises. Chang clarifies that GBMs, malignant tumors originating from glial support cells in the brain, are notoriously aggressive and lethal. Histologically, they are characterized by multifocal patterns and necrotic regions due to rapid outgrowth exceeding vascular supply. Despite being devastating, neurosurgeons still aim for maximal tumor resection, as the extent of surgical removal directly correlates with prolonged patient survival, though it remains far from curative.

The future, Chang asserts, holds promise due to advances in molecular genetics. Unlike a decade ago, today's diagnostics include detailed genetic profiling of GBMs, enabling treatment plans tailored to individual mutation patterns. Additionally, immunotherapy research is focusing on overcoming the tumor's immune evasion mechanisms, attempting to "uncloak" GBMs so immune cells can recognize and attack them. Innovative drug delivery techniques such as focused ultrasound are under exploration to transiently disrupt the blood-brain barrier, facilitating targeted chemotherapy access to the tumor. While significant strides are underway, Chang acknowledges that progress must accelerate to meaningfully alter the grim prognosis of GBM.

Awake Brain Surgery

One of the most intriguing elements addressed is the nature and necessity of awake neurosurgery. Chang explains that the brain itself lacks pain receptors, a fact that enables surgeons to operate on cortical areas while patients remain conscious. The pain is predominantly from scalp and dural incisions, which are anesthetized locally. Patients may be lightly sedated but must remain awake and responsive during critical phases to ensure essential brain functions, like speech and movement, are preserved.

This approach requires delicate balancing: resecting as much tissue as possible to remove tumors or epileptogenic zones while avoiding damage that could cause paralysis or language deficits such as aphasia. Real-time brain mapping involves electrically stimulating cortical surface points to observe temporary disruptions in function, guiding the surgeon in sparing "precious real estate." This interplay of maximal removal and functional preservation encapsulates the complexity of neurosurgical decision-making in these procedures, underscoring the interplay between anatomy and function.

Brain Plasticity

Chang elaborates on how the brain can accommodate injury through plasticity, a process whereby functions shift between hemispheres or within regions over time. For example, patients with slowly growing lesions in the frontal lobe can experience significant cortical reorganization, enabling them to tolerate resections that might otherwise cause major deficits. This adaptability is supported by synaptic plasticity—the dynamic remodeling of neuronal connections and strengths based on function and experience.

The corpus callosum, a heavily myelinated fiber tract connecting the brain's hemispheres, facilitates communication and functional redistribution. Severing this structure, as is done in some epilepsy surgeries to prevent seizure spread, reveals fascinating dissociation syndromes with distinct cognitive and sensory consequences. Chang reflects that while the philosophical implications about consciousness splitting remain mysterious, clinically, the procedure helps control debilitating seizures. This understanding emphasizes the intricate balance between brain structure, function, and adaptive capacity.

Brain-Computer Interfaces

The heart of the episode delves into Bruce-Computer Interfaces (BCI), a rapidly emerging field blending engineering and neuroscience. Chang describes BCIs as systems that record neural signals either non-invasively (via scalp EEG) or invasively (via electrocorticography or intracortical electrodes) and translate them through computational algorithms into actionable outputs. While non-invasive methods offer safety and ease, invasive approaches provide vastly superior spatial and temporal resolution essential for complex decoding.

Chang's lab has pioneered ECOG-based approaches where arrays of electrodes placed on the cortical surface decode neural activity related to speech production. He highlights a groundbreaking clinical trial involving "An," a woman rendered unable to speak by a brainstem stroke, who achieved remarkably high-accuracy speech decoding through such an implant. The breakthrough utilized machine learning and natural language processing tools, adapting AI algorithms that convert raw brain signals into rapid, intelligible text and synthesized speech. In this way, BCIs hold transformative potential to restore communication for paralyzed individuals and represent a new "language" for interfacing the brain with machines.

Advances in Neural Decoding

Chang details the intricate engineering hurdles facing BCIs, focusing on signal resolution and device stability. Intracortical electrodes provide single-neuron resolution but face biological challenges like immune reactions and signal degradation over time. Surface ECOG arrays, while less precise at the individual neuron level, maintain stable long-term recordings without penetrating brain tissue. Chang estimates that ECOG offers approximately a thousand-fold better resolution than EEG and intracortical devices provide an order of magnitude beyond ECOG, though with greater technical complexity.

He also explains novel decoding strategies involving breaking down speech into small acoustic units (phonemes or computationally derived "speech units") then using language models to reconstruct meaningful sequences. This layered approach mirrors advances in AI speech recognition but is adapted to operate directly on neural signals. Despite remaining challenges, particularly around downtime recalibration and individual variability, these engineering innovations have rapidly improved achievable words-per-minute rates, advancing real-world feasibility.

Neurological Restoration Beyond Communication

The conversation touches on broader aspirations beyond speech restoration, such as regaining motor functions and even autonomic processes like breathing. In diseases like ALS, the degeneration of motor neuron pathways disrupts communication with muscles and respiratory muscles, leading to severe disability and death. Chang is optimistic about coupling BCIs with functional electrical stimulation (FES) to bypass damaged pathways entirely, potentially restoring controlled muscle movements through direct muscle activation with implanted electrodes.

This integrative effort embodies an intersection of fields—neurology, neurosurgery, electrical and biomedical engineering, computer science, and material science—all converging to build neuroprosthetic systems. The complexity lies not only in signal decoding but also in creating compact, biocompatible, and safe implantable technologies capable of chronic function. Chang emphasizes that collaboration across disciplines is paramount, and future solutions will meld biological insights with engineered systems.

Biological and Cellular Therapies in Neuroscience

Chang reflects on cellular therapies and their promise in neurological diseases, particularly Parkinson's disease. Stem cell transplants aiming to replace lost dopaminergic neurons in the substantia nigra have shown some early clinical success, though challenges remain in controlling dopamine release to avoid side effects like dyskinesias. He notes that prior fetal graft attempts had mixed results but that newer cell models and delivery techniques may improve outcomes.

In a broader vision, Chang highlights the growing field of bioengineering and organoids—miniature lab-grown brain models—that advance understanding and treatment development. He envisions a future where biology itself becomes the platform for new technologies, such as biologically engineered neural interfaces and synthetic cells customized to restore or augment neurological functions, blending living systems with engineered constructs.

The Future of Neurosurgery

Looking ahead to 2040, Chang expresses cautious optimism that many neurological diseases once considered untreatable will become manageable chronic conditions or even curable. Decoding the molecular and genetic basis of conditions like GBM provides a pathway to precise molecular therapies. Similarly, early diagnosis and intervention in neurodegenerative diseases, including Alzheimer's, may revolutionize outcomes.

Chang envisions brain computer interfaces becoming widely available beyond research settings, with more durable, less invasive, high-channel count implantable devices offering complex neural decoding for a variety of disabilities. In neurosurgery itself, while some traditional techniques like craniotomy persist, emerging technology will continue to minimize invasiveness and maximize precision. Ultimately, understanding and interfacing with the brain's circuits will redefine treatment paradigms in the decades to come.