04-Bio-A8 · May 2015
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Paper format: National Exams, May 2015 — 04-Bio-A8 Biophysical Measurements. Three hours, open book, any non-communicating calculator. Seven questions of equal value (20 marks each); five constitute a complete paper and only the first five appearing in the answer book are marked. All seven are solved here, because this set is a study resource rather than an examination script. Every question is qualitative/descriptive — there is no numerical data to compute — so each answer follows flowing prose with instrumentation block diagrams where the question explicitly asks for one. Questions 2 and 4 cover nerve-fibre stimulate/record instrumentation and the Einthoven-triangle/Lead I ECG.
Reference texts (the books an open-book candidate should have on the desk for this subject):
Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.
The analysis is restated in full below.
The scalp-recorded EEG does not arise from individual neuronal action potentials, which are too brief and too poorly synchronized across the millions of neurons under an electrode to summate into a measurable field. Instead it originates from the summed extracellular field potentials produced by postsynaptic potentials (excitatory and inhibitory, EPSPs and IPSPs) in the large, geometrically aligned pyramidal neurons of the cerebral cortex: because these neurons lie in parallel columns oriented perpendicular to the cortical surface, their individual dipole fields add constructively when many are active synchronously, producing a field large enough to be detected through the skull and scalp. The surface EEG has an amplitude of roughly 10–100 μV (occasionally up to 200 μV in certain states or pathologies) and a bandwidth of approximately 0.5–100 Hz, conventionally subdivided into the delta (0.5–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), beta (13–30 Hz), and gamma (>30 Hz) bands, each associated with different states of arousal, sleep stage, or cognitive activity.
The click stimulator delivers brief (about 100 μs), repetitive audio clicks into the ear at a rate of roughly 10–20 per second and simultaneously issues a synchronization trigger, which is essential because it defines the time-zero from which the response is time-locked and later averaged. Electrodes are placed at the vertex (Cz) and on the ipsilateral earlobe/mastoid, feeding a differential preamplifier with high input impedance and high common-mode rejection ratio, which amplifies the tiny vertex-minus-ear potential difference while rejecting common-mode interference picked up equally by both electrodes. The signal then passes through a bandpass filter restricted to the 150–3000 Hz band specified for this potential, which both matches the known spectral content of the BAEP and rejects the much larger, slower ongoing EEG background that would otherwise dominate. The core of the system is the signal averager: because a single BAEP sweep (amplitude <1 μV) is completely buried in background EEG and noise many times larger, hundreds to a couple of thousand stimulus-locked sweeps are digitized and averaged together, synchronously with the stimulus trigger, so the time-locked evoked response reinforces while the uncorrelated background noise averages toward zero. Finally, the averaged waveform is passed to a display and waveform-recording block, where the characteristic sequence of peaks (waves I–V, reflecting sequential brainstem auditory relay stations) is identified and their latencies measured.
The dominant "noise" is the ongoing background EEG itself, which at 10–100 μV is one to two orders of magnitude larger than the sub-microvolt evoked response; because it is not time-locked to the stimulus, while the evoked potential is, synchronous (coherent) averaging over $N$ stimulus-locked sweeps improves the signal-to-noise ratio by a factor of $\sqrt{N}$, which is why hundreds to thousands of sweeps are used. Power-line interference at 60 Hz (mains hum picked up by the electrode leads) is removed by the differential amplifier's common-mode rejection together with, if necessary, a notch filter at 60 Hz, and is reduced at the source by shielding the cables and using a driven-right-leg-style reference. Myogenic (EMG) noise from scalp and neck muscle contraction is broadband and can be large; it is minimized by having the patient remain relaxed (or sedated in infants), and is partly attenuated by the 150–3000 Hz bandpass filter, which excludes some of the EMG spectrum outside that band. Electrode motion and contact-impedance noise, caused by variable skin–electrode contact, is minimized by careful skin preparation (abrasion, conductive gel) to achieve low, stable electrode impedance, and is further rejected by the differential amplifier's high input impedance and CMRR. Finally, amplifier input (thermal/flicker) noise, unavoidable in any front-end electronics, is minimized by using a low-noise instrumentation amplifier as the first gain stage, placed as close to the electrodes as practical to avoid adding cable-pickup noise before the signal is amplified.
A neonatal hearing-screening BAEP test would use exactly this chain — vertex/ear electrodes, a 150–3000 Hz bandpass, and averaging of roughly 1500–2000 click-locked sweeps — to reliably resolve a sub-microvolt wave V response despite the infant's much larger background EEG and any residual movement artifact, confirming intact auditory pathway conduction from cochlea to brainstem.
