Summary of Dasar-dasar pengukuran gelombang otak dengan electroencephalography (EEG)

Summary of "Dasar-dasar pengukuran gelombang otak dengan Electroencephalography (EEG)"

This video provides a comprehensive introduction to the basics of measuring brain waves using Electroencephalography (EEG). It covers fundamental concepts, the technology behind EEG, brain wave types, data acquisition, noise/artifacts, and data analysis methods, concluding with practical tools and references for further study.

---

Main Ideas and Concepts

1. Introduction to EEG and Brain Waves

• EEG is a tool to measure brain waves by detecting electrical potentials generated by neurons.
• The brain produces several types of waves: Delta, Theta, Alpha, Beta, and Gamma.
  • Alpha waves are associated with relaxation or sleep.
  • Beta waves relate to active thinking.
  • Gamma waves relate to concentration.
• These classifications are simplified models and have underlying assumptions.

2. What is EEG?

• EEG records voltage differences on the scalp using metal Electrodes.
• Electrodes pick up electrical potentials generated by brain activity.
• EEG data (electroencephalogram) is represented graphically.
• EEG is a form of functional brain imaging, focusing on brain activity rather than structure.

3. Comparison with Other Brain Imaging Methods

• EEG has excellent temporal resolution (milliseconds to sub-milliseconds), allowing immediate detection of brain activity.
• fMRI has better spatial resolution (can localize activity to specific brain regions) but slower temporal resolution (seconds).
• EEG records signals from millions of synchronized neurons, mainly pyramidal cells in the cerebral cortex.
• Single-cell or synapse-level imaging requires invasive methods not applicable to humans.

4. Neural Basis of EEG Signals

• EEG measures postsynaptic potentials (PSPs) of pyramidal neurons, not individual action potentials.
• Pyramidal cells are oriented perpendicular to the cortical surface, influencing the direction of electrical and magnetic fields.
• EEG measures electrical potentials (aligned with cell orientation), while MEG measures magnetic fields (perpendicular to electrical fields).

5. Limitations of EEG

• Due to the folded brain structure (gyri and sulci), electrical signals recorded at an electrode may originate from distant brain areas.
• EEG cannot precisely localize brain activity sources.
• Analogy: recording crowd noise outside a stadium—you hear the noise but not individual speakers.

6. EEG Equipment and Setup

• Components: Electrodes (metal disks), Electrode caps, Amplifiers, and computers.
• Amplifiers boost tiny brain signals and convert analog signals into digital data.
• Sampling rate is crucial: higher sampling rates (e.g., 1000 Hz or more) better capture brain wave details and avoid aliasing errors.
• Electrodes types:
  • Wet Electrodes use conductive gel for better contact.
  • Dry Electrodes (comb-like) are easier to apply but may provide lower-quality data.
  • Passive Electrodes send raw signals to Amplifiers.
  • Active Electrodes have built-in Amplifiers for better signal quality.
• Electrode placement follows the 10-20 system, a standardized method using anatomical landmarks (nasion, inion, ear points) to ensure consistent electrode positioning.

7. Brain Wave Analysis

• Brain waves are complex mixtures of multiple sine waves at different frequencies and amplitudes.
• Raw EEG signals are a superposition of multiple frequencies (Delta to Gamma).
• To identify specific wave types, signal decomposition (e.g., Fourier Transform) is used.
• Types of analysis:
  • Time domain (e.g., event-related potentials or ERPs): looks at wave patterns relative to stimulus timing.
  • Frequency domain: identifies power in specific frequency bands (Alpha, Beta, etc.).
  • Time-frequency analysis: combines both aspects to show how frequencies change over time.
  • Machine learning can be applied to detect hidden features in EEG data.

8. Noise and Artifacts in EEG

• EEG signals are very small and easily contaminated by noise.
• Common noise sources:
  • Electrical noise from power lines (50 Hz in Indonesia, 60 Hz in the US).
  • Eye movements and blinks produce large artifacts detected near frontal Electrodes (EOG: electrooculogram).
  • Muscle activity (EMG), e.g., jaw clenching, swallowing, causes high-frequency noise.
  • Electrode movement or poor contact causes channel noise.
  • Alpha waves can be considered noise if participants are sleepy during experiments.
  • Cardiac signals (ECG) and blood pulsations can also contaminate EEG.
• Noise mitigation strategies:
  • Minimize electrical devices in the room.
  • Use Faraday cages to shield external electromagnetic interference.
  • Ensure proper electrode placement and gel application.
  • Instruct participants to minimize movement and blinking.
  • Data cleaning and filtering during analysis.

9. Data Processing and Interpretation

Raw EEG

Notable Quotes

— 02:09 — « Today, the weather was ok. »

— 03:02 — « Dog treats are the greatest invention ever. »

— 21:32 — « The analogy that is often used is like we are recording the sound of a stadium from outside. You can hear the cheers but you can't tell who is sitting where or what each person is saying. »

— 64:36 — « How can we make our data noise-free? That's impossible, right? There will always be the simplest noise, blinking, heartbeats, or electrical interference from the computer stimulus. »

— 65:30 — « No substitute for clean data, meaning if your data is dirty, it can't be fixed or manipulated easily. Clean data is essential. »

Category

Educational

Video