Embedded Files
Comparison
Comparison
EEG is one of the only techniques that measures the brain's electrical activity directly. Most other neuroimaging techniques rely on indirect markers such as blood flow or metabolic processes.
For example, PET and fMRI provide information about the absorption of specific substances in brain regions or about metabolic rates and glucose consumption. Compared to EEG, these techniques enable excellent spatial resolution, as they detect metabolites with precision, but they have very low temporal resolution.
In contrast, EEG measures the brain's electrical activity arising from the postsynaptic potentials of neurons. It offers a noninvasive, low-cost method to explore the relationships between cortical sites and the time evolution of brain processes.
Example of an EEG machine attached to the subject whilst sleeping
Limits
Limits
EEG has different limitations, especially in terms of spatial resolution. While it provides detailed information on "when" brain activity occurs, it is difficult to understand "where" exactly the activity originates within the brain (low spatial resolution). This is because EEG measures electrical activity, which is a gradient and can't be precisely localised.
Furthermore, the signals are primarily influenced by cortical regions, as deeper brain structures are less accessible due to signal attenuation (skull and tissue resistance).
Another challenge is the sensitivity of EEG to noise and artifacts. Muscle movements, blinking, or external electrical interference can distort the recorded data, requiring careful preprocessing to ensure accurate interpretation. Usually, while measuring EEG, one also measures heart rate and puts two electrodes near the eyes to measure eye movements. This helps to remove artifacts, but how the noise is cancelled is still arbitrary and one is never sure if they are also deleting important information.
How to improve
How to improve
Efforts to enhance EEG often focus on combining it with other imaging techniques, such as fMRI, in what is known as multimodal neuroimaging. This approach seeks to integrate the temporal resolution of EEG with the high spatial resolution of fMRI, providing a more comprehensive picture of brain activity. While promising, multimodal approaches introduce complexities in data synchronization, analysis, and cost.
Other ways to improve EEG include:
High-density EEG (HD-EEG): Using systems with more electrodes (up to 256) can improve spatial resolution by capturing finer details of electrical activity distribution. However, this is often more time consuming, uncomfortable for the patient and still limited compared to other techniques.
Source Localization Algorithms: Advanced computational methods, such as beamforming or independent component analysis (ICA), can better estimate the location of neural activity sources and remove artifacts.
Artifact Removal Techniques: Developing more robust preprocessing pipelines to eliminate noise from muscle movements, eye blinks, or external interference.
Wearable Technology: Creating portable, wireless EEG systems can make the technique more accessible for clinical and everyday applications. For example, research studies in sleep have started using portable EEGs, so that the patients can sleep at home creating a more close to reality environment.
Another option is a portable EEG like the EMOTIV headset, which contains only 14 electrodes, making it more comfortable and used in every day settings like video games, to move wheelchairs for paraplegic patients or just 2-channel headphones that give you real time brain data. Whether ethically this is a good idea, is left to the reader, who is invited to also to reflect on the consequences of having people read how focused or productive you are during your work.
Integration with Machine Learning: Machine learning models can improve the interpretation of EEG data, making it possible to detect patterns or predict brain states with higher accuracy.
Why does EEG have low spatial resolution?
Why does EEG have low spatial resolution?
Here the EEG activity is compared to a stadium, one just hears the noise the stadium makes (high temporal resolution), but it is not possible to locate where the noise is coming from (low spatial resolution).
"Imagine you are a journalist equipped with a hand-held microphone, which will here be analogous to a recording electrode. You are reporting from a soccer match. If you are standing next to the coach, you can interview her and comprehend her voice despite the noise throughout the stadium. This is akin to recording action potentials of individual neurons. If you are in the press box, you will not be able to record the ongoing conversations between the coach and players on the field, but instead can capture the general commentary of other reporters inside the press box as well as the hum of the audience outside. This is akin to recording local field potentials, where there are contributions of both proximal and relatively distal events. Finally, from your hotel balcony, having lost your press credentials, you may nonetheless be able to hear the joyful cry in unison of the team’s supporters from within the stadium when a goal is scored. This is analogous to EEG recordings."[1]
references[1] https://www.sciencedirect.com/science/article/pii/S096098221831551300 - Electroencephalography
Page updated
Google Sites
Report abuse