Invasive Brain-Computer Interface (iBCI)
- nikitha sangepu
- Sep 30, 2025
- 5 min read
What Is an Invasive Brain–Computer Interface (BCI)?
Invasive brain–computer interfaces (BCIs) represent the cutting edge of neurotechnology, offering unmatched precision in decoding brain signals to control external devices. Unlike noninvasive systems that rely on scalp recordings, invasive BCIs involve implanting electrodes directly into or near brain tissue, allowing researchers and clinicians to tap into the brain’s most intimate electrical conversations.
What Does an Invasive BCI Capture?
To understand what invasive BCIs record, we must first explore the structure of a neuron
Neurons communicate via action potentials which are brief electrical impulses that travel from cell body to synaptic terminals, triggering the release of neurotransmitters. These spikes are generated by ion movement across the neuronal membrane and are the fundamental units of brain signaling.
Invasive BCIs capture three main types of signals:
1. Single Unit Activity (SUA):
These are records of spikes/action potentials originating from a single neuron. They offer the highest resolutions as they directly capture the changes at cellular level, and hence are considered as the gold standard.
The challenge experienced while recording single unit activity lies with prevention of electrode drift from the implanted site and the immune response mounted by the body against these foreign objects.
2.Multi-Unit Activity (MUA):
Neurons that respond similarly to a given stimulus often fire action potentials simultaneously. Multi-unit activity (MUA) captures the collective spiking behavior of such nearby neurons, whereas single-unit activity (SUA) isolates and records the firing patterns of an individual neuron.
This helps in easier tracking of the activity but is less selective compared to SUA.
From a BCI perspective, the exact source of information, whether it is MUA or SUA, is not significant as long as it is selective for the end task.
3. Local Field Potentials (LFPs):
As the action potential travels down the neuron, transmembrane potential
differences generate electrical currents in the surrounding medium. This electric field arises from the superposition of all ionic processes occurring in the brain, ranging from rapid action potentials and synaptic currents to the slower fluctuations produced by glial cells. At any given point in space, these overlapping currents combine to produce electric potentials, which are recorded as local field potentials (LFPs).
It is hypothesized that synaptic currents play a significant role in shaping LFPs, as they are relatively slower and more sustained. This temporal persistence allows extracellular currents from multiple neuronal compartments to overlap in time, resulting in a measurable and coherent signal.
What are the components of an Invasive BCI?
Brain signal sensors | Electrodes to capture the brains signals |
Decoder | Algorithms used to translate the signals to computer commands |
Effector | The end object that performs the task |
1: Brain Signal Sensors
Electrodes are surgically implanted to detect neural signals. Types include:
A. Penetrating Microelectrodes:
Thin and small electrodes grouped in the form of an array that pierce the cortical surface to a depth of 0.5 to 1.5 mm. These require careful placement to avoid damaging blood vessels on the brain surface.
B. Stereoencephalographic (sEEG) Electrodes:
Long electrodes inserted deep into brain matter via burr holes. They can detect the signals over a large surface along their longitudinal length. Burr hole surgery carries less surgical burden compared to placement of microelectrode arrays where a part of the skull is removed. It offers faster recovery as well.
C. Endovascular Neural Implants (stentrode):
Electrodes are placed inside cerebral blood vessels of the desired area using a vascular stent-like delivery. they are less invasive but are limited to the accessible vasculature only.
2: Decoder
Once signals are captured, they must be translated into machine-readable commands. This is done using decoding algorithms, such as:
A. Population Vector Analysis:
It relies on the observation that primary motor cortex M1 neurons often exhibit a “preferred direction” of firing related to movement direction, firing maximally for movements in one direction and declining in a cosine manner for movements in directions other than their preferred direction. By summing the weighted contributions of multiple neurons, a population vector can be constructed to predict the intended movement direction.
B. Continuous Linear Filtering:
This technique uses regression-based models to continuously decode motor intentions from neural signals. By applying linear filters to the input data (e.g., spike rates or local field potentials), it predicts the corresponding output (e.g., trajectory of a cursor or robotic arm) in real time.
3: Effector
The decoded signals control external devices, such as computer cursors, robotic limbs, Functional Electrical Stimulation (FES).
Most patients prefer regaining control over their own limbs rather than relying on a robotic arm. In such conditions, FES can be useful. It operates on the principle that electrodes implanted in muscles can evoke movement in response to the movements imagined by the user (imagined motor commands). This technique is particularly beneficial for patients with spinal cord injuries with intact brain and muscle function.
Why Invasive BCIs Are More Accurate?
High Signal-to-Noise Ratio:
Non-invasive brain-computer interfaces are susceptible to various sources of noise, such as eye blinks, electromagnetic interference from nearby cables, or magnetic fields in the environment. These artifacts can distort the signal and compromise data quality. To extract meaningful commands for specific tasks, the raw data must undergo signal extraction and feature vector construction.
In contrast, Invasive BCIs collect brain signals from the closest possible location—directly from neurons and the surrounding medium. This proximity minimizes environmental interference and yields high-resolution data with superior signal fidelity.
Specific Neuronal Activity Tracking:
Invasive BCIs enable precise monitoring of individual neurons. This granularity allows the recorded signals to be directly correlated with specific aspects of movement, enhancing the accuracy of motor decoding.
Access to Deep Brain Structures
Using stereoelectroencephalography (SEEG) electrodes, invasive BCIs can access deeper brain regions that are typically unreachable with surface-level techniques. This expands the scope of neural data acquisition beyond cortical areas.
Enhanced Degrees of Freedom in Movement
High signal fidelity, neuron-level specificity, and access to deep structures collectively enable invasive BCIs to achieve multiple degrees of freedom in movement control. Such precision is often limited in non-invasive methods like EEG, which suffer from low spatial resolution and signal degradation.
What are the risks of Invasive BCI?
Despite their promise, invasive BCIs face several hurdles:
Biocompatibility: Any foreign object implanted into the body could mount an Immune response in the form of inflammation and scar tissue formation around the object. This results in the degradation of signal quality over time.
Surgical Risks: Neurosurgery carries inherent complications, including anesthesia-related risks, infections, bleeding, etc.
Longevity: Electrodes may shift or degrade after implantation, affecting signal stability.
Hardware Maintenance: The system, once being implanted, is difficult to fix hardware problems or to update any component, therefore requiring much higher reliability and reducing a certain degree of flexibility.
Affordability: due to the complexity of the surgical procedure itself and the necessary care afterwards, invasive BCI is expensive, which needs to be addressed in order to increase its accessibility.
Final Thoughts:
Invasive BCIs offer a powerful window into the brain, enabling life-changing applications in neuroprosthetics, rehabilitation, and even communication for locked-in patients. While challenges remain, ongoing innovations in electrode design, decoding algorithms, and biocompatibility are steadily pushing the boundaries of what’s possible.
Whether you're a clinician, researcher, or curious mind, invasive BCIs represent one of the most exciting frontiers in neuroscience and human–machine interfacing.
Comments