Technical Background

James D. Geyer, MD

Paul R. Carney, MD

Introduction

Polysomnography is the recording of multiple physiologic functions during sleep. Standard polysomnography usually includes the following variables:

Electroencephalogram (EEG)
Electrooculogram (EOG)
Electrocardiogram (EKG)
Electromyography (EMG)
Pulse oximetry
Airflow (nasal and oral)
Respiratory effort (thoracic and abdominal)
Limb movements
Snore sensors

Special studies may include the following:

Expanded EEG
Esophageal manometry
CO₂ monitoring
CPAP/BiPAP
Esophageal pH

The EEG, EMG, and EOG data are used to identify the sleep stage. Accurate sleep staging is necessary for the diagnosis and management of sleep disorders. In addition to these channels, the other variables listed above are used to identify specific sleeprelated disorders.

Signal Processing

Differential amplifier: A differential amplifier amplifies the difference between two input signals. Any potentials shared between the two signals are removed leaving only the difference between the signals.

Common mode rejection ratio (CMRR): This ratio refers to the ability of an amplifier to reject inphase potentials and amplify outof-phase potentials. The CMRR is measured by connecting both input channels of an amplifier to the same signal source. Ideally, the output would be zero (CMRR would be infinity). Good amplifiers have a CMRR between 1,000 and 10,000.

Polarity: Standard EEG polarity convention is as follows:

If input 1 is negative compared to input 2, an upward signal is displayed.

If input 1 is positive compared to input 2, a downward signal is displayed.

If input 2 is negative compared to input 1, a downward signal is displayed.

If input 2 is positive compared to input 1, an upward signal is displayed.

Filters

Filters allow the technologist and polysomnographer an opportunity to attenuate artifacts.

Low-frequency filter: Ideally, the low-frequency filter attenuates all frequencies below the cutoff frequency of the filter and allows all frequencies above the cutoff frequency to pass unchanged. In reality, the cutoff does not occur at a distinct single frequency but over a range of frequencies with a variable attenuation. For EEG, the cutoff frequency is defined as the frequency at which the output is reduced by 30%. In digital recording, little filtering is performed prior to digitization. Subsequent digital filtering can be performed as needed.

High-frequency filter: Ideally, the high-frequency filter attenuates all frequencies above the cutoff frequency of the filter and allows all frequencies below the cutoff frequency to pass unchanged. In reality, the cutoff does not occur at a distinct single frequency but over a range of frequencies with a variable attenuation. For EEG, the cutoff frequency is defined as the frequency at which the output is reduced by 30%. In digital recording, little filtering is performed prior to digitization. Subsequent digital filtering can be performed as needed.

60 Hz notch filter: Ideally, the 60-Hz notch frequency filter removes 60-Hz noise from electrical sources without affecting other frequencies. In reality, the 60-Hz filter attenuates a range of frequencies around 60 Hz. When 60-Hz noise is attenuated, all EEG activity in that range is also attenuated. Given these and other problems associated with the use of the 60-Hz notch filter, it should be used only when absolutely necessary.

Phase Shift: Shift of a waveform either earlier or later in time caused by filtering.

Only gold members can continue reading. Log In or Register to continue