2.4 Physiological Measurements and Instrumentation

Physiological Measurements in Biomedical Engineering

Physiological measurements let engineers and clinicians quantify what's happening inside the body, from the electrical impulses of the heart to the oxygen level in your blood. These measurements form the foundation of modern diagnostics, patient monitoring, and medical research.

This section covers the core principles behind acquiring biological signals, the signal processing chain that makes raw data usable, vital signs monitoring, electrical activity assessment (ECG and EMG), and the major medical imaging modalities.

Principles and Applications

Physiological measurements involve quantifying biological signals and parameters to assess how the body is functioning. They're essential for diagnosing diseases, monitoring patients, and evaluating whether treatments are working.

Biomedical engineers design the sensors, transducers, and instrumentation that capture this data. A sensor detects a physical quantity (like voltage or pressure), while a transducer converts one form of energy into another (usually into an electrical signal that can be recorded). Some examples:

• Electrodes detect the tiny voltages produced by cardiac or muscle cells (used in ECG)
• Force sensors measure ground reaction forces during walking (used in gait analysis)
• Thermistors change their electrical resistance in response to temperature shifts

Once a signal is acquired, it needs to be cleaned up and made usable through signal processing:

• Filtering removes unwanted noise and artifacts
• Amplification boosts weak biological signals so they can be analyzed
• Digitization converts the continuous analog signal into a digital format for computer processing

These measurements show up across many fields: continuous monitoring in ICUs, studying disease mechanisms in research labs, optimizing performance in sports medicine, and restoring function in rehabilitation engineering.

Signal Processing and Data Analysis

Raw physiological signals are almost always contaminated with noise, motion artifacts, and electrical interference. Signal processing is how you turn messy raw data into something clinically meaningful.

Filtering removes specific frequency components from a signal. The type of filter you choose depends on what kind of noise you're dealing with:

• Low-pass filters remove high-frequency noise (e.g., muscle artifacts contaminating an ECG signal)
• High-pass filters remove low-frequency drift (e.g., slow baseline wander from breathing movements in an ECG)
• Band-pass filters keep only a specific frequency range and attenuate everything outside it

Amplification boosts signal strength. Many biological signals are extremely small (EMG signals are often in the microvolt range), so amplification is critical. Differential amplifiers are commonly used because they amplify the difference between two electrode inputs while rejecting noise that appears equally on both inputs (common-mode rejection).

Analog-to-digital conversion (ADC) turns the continuous analog signal into discrete digital samples. Two key ADC parameters determine signal quality:

• Sampling rate: how many samples per second are captured. A higher sampling rate preserves more detail. By the Nyquist theorem, the sampling rate must be at least twice the highest frequency in the signal.
• Resolution (bit depth): how finely each sample's amplitude is represented. More bits means smaller measurable voltage differences.

Averaging techniques improve the signal-to-noise ratio (SNR) by reducing random noise:

• Signal averaging collects multiple recordings of the same event and averages them together. Random noise cancels out while the consistent signal remains.
• Ensemble averaging works the same way but is applied to periodic signals, averaging across many cycles.

For more complex signals, advanced methods come into play. Wavelet transforms provide time-frequency localization, which is useful for non-stationary signals like EEG where frequency content changes over time. Adaptive filters automatically adjust their parameters based on the signal characteristics, optimizing noise reduction on the fly.

Vital Signs Measurement Techniques

Blood Pressure and Heart Rate Monitoring

Blood pressure can be measured invasively or non-invasively:

• Arterial line monitoring (invasive): A catheter is inserted directly into an artery, giving continuous, beat-by-beat pressure readings. This is the gold standard for accuracy but carries infection and bleeding risks.
• Auscultatory method (non-invasive): A sphygmomanometer cuff is inflated around the arm, then slowly deflated while a clinician listens with a stethoscope for Korotkoff sounds. The pressure at which sounds first appear is systolic; the pressure at which they disappear is diastolic.
• Oscillometric method (non-invasive): The cuff detects small oscillations in cuff pressure caused by arterial pulsations. An algorithm estimates systolic and diastolic values from the oscillation pattern. This is the method used by most automatic blood pressure monitors.

Heart rate can be measured through several approaches:

• ECG detects the heart's electrical activity via skin electrodes (more detail below)
• Photoplethysmography (PPG) uses an optical sensor to detect changes in blood volume in the microvasculature. Each heartbeat causes a small pulse of blood that changes how much light is absorbed or reflected.
• Pulse oximetry measures both heart rate and oxygen saturation by analyzing light absorption at two wavelengths (red and infrared)

In critical care settings like ICUs, blood pressure and heart rate are monitored continuously. Bedside monitors display real-time waveforms and numerical values, with alarms that trigger when readings cross predefined thresholds.

Respiratory Rate and Other Vital Signs

Respiratory rate is measured in several ways:

• Visual observation: counting chest wall movements per minute (simple but labor-intensive and intermittent)
• Impedance pneumography: measures changes in electrical impedance across the chest as the lungs expand and contract during breathing
• Capnography: an infrared sensor measures the concentration of $CO_2$ in exhaled air, producing both a waveform and a numerical respiratory rate. This is widely used in anesthesia and critical care.

Body temperature measurement depends on the clinical context:

• Oral and rectal thermometers use mercury or digital sensors. Rectal measurement is closest to core body temperature.
• Tympanic thermometers detect infrared radiation from the eardrum (tympanic membrane), giving a rapid reading that approximates core temperature.
• Skin surface thermometers use thermistors or infrared sensors but measure peripheral temperature, which can differ from core temperature.

Oxygen saturation ($SpO_2$) is measured by pulse oximetry. A probe placed on the finger, toe, or earlobe emits red and infrared light. Oxygenated hemoglobin ($HbO_2$) and deoxygenated hemoglobin ($Hb$) absorb these wavelengths differently. The ratio of absorption at the two wavelengths is used to calculate the percentage of hemoglobin that is oxygenated. Normal $SpO_2$ is typically 95–100%.

Integrated monitoring systems display all vital signs on a single screen, allowing clinicians to spot trends and detect deterioration early.

Electrical Activity Assessment with ECG and EMG

Electrocardiography (ECG)

ECG is a non-invasive recording of the heart's electrical activity using electrodes on the skin. It's one of the most commonly performed diagnostic tests in medicine.

The standard ECG waveform has three main components:

• P wave: atrial depolarization (the atria contract)
• QRS complex: ventricular depolarization (the ventricles contract). This is the largest deflection because the ventricles have much more muscle mass than the atria.
• T wave: ventricular repolarization (the ventricles reset electrically before the next beat)

ECG is used to diagnose a range of cardiac conditions:

• Arrhythmias: abnormal heart rhythms such as atrial fibrillation (irregular, rapid atrial activity) or ventricular tachycardia (dangerously fast ventricular rhythm)
• Myocardial infarction (heart attack): identified by ST-segment elevation or depression on the ECG tracing
• Conduction disorders: conditions like bundle branch blocks show up as abnormally shaped or widened QRS complexes

ECG signal processing follows a standard chain:

1. High-pass filtering removes baseline wander from respiration or patient movement
2. Low-pass filtering removes high-frequency noise such as muscle artifacts
3. Notch filtering eliminates 50/60 Hz power line interference
4. Amplification boosts the signal for display and analysis
5. Digitization converts the analog signal for computer processing

Two advanced ECG techniques worth knowing:

• Heart rate variability (HRV) analysis examines the beat-to-beat variation in time intervals between heartbeats. This variation reflects autonomic nervous system function. Reduced HRV is associated with various cardiac and non-cardiac conditions.
• Signal-averaged ECG (SAECG) averages many heartbeat cycles together to reveal very small signals called late potentials, which indicate increased risk for ventricular arrhythmias.

Electromyography (EMG)

EMG measures the electrical activity of skeletal muscles. The signal represents the summation of motor unit action potentials (MUAPs) generated by muscle fibers during contraction.

There are two main types:

• Surface EMG: electrodes placed on the skin record the overall activity of a muscle or muscle group. This is non-invasive and commonly used in biomechanics and rehabilitation research.
• Needle EMG: a fine wire electrode is inserted directly into the muscle to record individual motor unit activity. This provides much more specific information and is used clinically to diagnose neuromuscular disorders.

Clinical and research applications of EMG include:

• Diagnosing neuromuscular disorders such as myopathies (muscle diseases) and neuropathies (nerve diseases) by analyzing signal characteristics
• Assessing muscle function by measuring signal amplitude and frequency content at different contraction levels
• Studying biomechanics by mapping muscle activation patterns and coordination during movement

EMG signal processing involves several steps:

1. Amplification: EMG signals are very small (typically in the microvolt range), so significant amplification is needed
2. Band-pass filtering: removes low-frequency motion artifacts and high-frequency noise
3. Rectification: converts the bipolar signal (which swings positive and negative) into a unipolar signal by taking the absolute value or squaring the signal. This makes it easier to assess overall activation level.

Advanced EMG analysis falls into a few categories:

• Time-domain analysis: parameters like root mean square ($RMS$) amplitude, integrated EMG ($iEMG$), and zero crossings quantify the signal's magnitude and activity level
• Frequency-domain analysis: power spectral density ($PSD$) estimation reveals the frequency content of the signal. A shift toward lower frequencies during sustained contraction is a classic indicator of muscle fatigue.
• Muscle synergy analysis: decomposes multi-muscle EMG recordings into underlying activation patterns, helping researchers understand how the nervous system coordinates groups of muscles

Imaging Techniques for Anatomical Visualization

X-ray and Computed Tomography (CT)

X-ray imaging uses ionizing radiation to produce 2D projection images of internal structures. X-rays are generated by an X-ray tube, pass through the body, and are attenuated (absorbed or scattered) differently depending on tissue density. The result is a shadowgram: dense structures like bone appear white, air-filled spaces appear black, and soft tissues appear in shades of gray.

Common uses include chest X-rays (assessing lungs, heart, and major vessels) and skeletal imaging for fractures.

Computed tomography (CT) extends X-ray imaging into three dimensions. Here's how it works:

1. A rotating X-ray tube and detector array circle around the patient
2. Multiple projection images are acquired from many different angles
3. Mathematical reconstruction algorithms (such as filtered back projection) combine these projections into cross-sectional slices
4. The slices can be stacked to create a full 3D image

CT provides high spatial resolution and excellent soft tissue contrast compared to plain X-ray. It's widely used for detailed imaging of the brain, chest, abdomen, and pelvis, and for guiding interventional procedures like biopsies.

Radiation dose is a real concern with both X-ray and CT, since ionizing radiation exposure increases cancer risk over time. Strategies to minimize dose include:

• Low-dose CT scan protocols
• Iterative reconstruction algorithms (which produce quality images from less radiation data)
• Lead shielding and beam collimation to limit exposure to the area of interest

Magnetic Resonance Imaging (MRI) and Ultrasound

MRI creates high-resolution images of soft tissues using strong magnetic fields and radio waves, with no ionizing radiation involved.

The basic MRI process:

1. A powerful magnet (typically 1.5 or 3 Tesla) aligns hydrogen nuclei (protons) in the body
2. Radiofrequency (RF) pulses are applied, which tip the protons out of alignment and cause them to emit detectable signals
3. Receiver coils detect these signals
4. Fourier transform and other mathematical techniques reconstruct the signals into images

Different MRI sequences highlight different tissue properties:

• T1-weighted images provide good anatomical detail (fat appears bright)
• T2-weighted images highlight fluid (useful for detecting edema or inflammation)
• Diffusion-weighted imaging detects the movement of water molecules, which is critical for early stroke detection

Functional MRI (fMRI) measures the blood oxygenation level-dependent (BOLD) signal. When a brain region becomes active, local blood flow increases and the ratio of oxygenated to deoxygenated hemoglobin changes. fMRI detects this change, allowing researchers to map brain activity during tasks. MR spectroscopy (MRS) can measure concentrations of specific metabolites in tissue.

MRI is particularly valuable for imaging the brain, spinal cord, joints, and soft tissues where CT contrast is limited.

Ultrasound uses high-frequency sound waves (typically 2–18 MHz) to visualize soft tissues in real time. A transducer emits sound waves that penetrate the body and reflect off tissue boundaries. The reflected echoes are processed into images.

• Real-time imaging makes ultrasound ideal for monitoring fetal development and guiding needle placement during procedures
• Doppler ultrasound detects the frequency shift of reflected sound waves caused by moving blood cells, allowing measurement of blood flow velocity and direction

Both MRI and ultrasound are non-invasive and radiation-free, making them safe for repeated use. Ultrasound is portable and inexpensive compared to MRI, but MRI provides superior soft tissue contrast and resolution.

Biomedical engineers contribute to advancing these modalities through techniques like parallel imaging and compressed sensing (to reduce MRI scan times), machine learning for automated image analysis, and development of image segmentation, registration, and visualization tools that help clinicians interpret complex imaging data.