3 Biosignals

• The CNS (Central Nervous System) → consists of the brain and spinal cord.

• The PNS (Peripheral Nervous System) → consists of the nerves that branch out from the CNS and extend to the rest of the body.

  • The PNS connects the CNS to the external environment.

Human activities are behaviors concerning the body or the environment.

• The three types are:

  • Body Position (e.g., sitting, standing),

  • Body Action (e.g., catching, throwing, pulling),

  • Body Status (e.g., brain activity, heart activity).

Human activities are mapped using the "Where/What/How" framework:

• Body Position (Where) → indoor positioning, hand tracking, social distancing.

• Body Action (What) → fall detection, gait analysis, gesture recognition, HCI.

• Body Status (How) → emotion sensing, stress sensing, heart rate detection.

A sensor is a device that produces a measurable output change to a known input stimulus.

• The input can be physical (e.g., temperature, pressure) or chemical/biochemical.

• Sensors are transducers — they convert an incoming property into a signal.

They can be classified by:

• Active vs. Passive — active sensors need an external signal or power; passive sensors generate outputs directly without external signal.

• Means of detection used in the sensor — e.g., electric, biological, chemical, radioactive…

• Type of conversion phenomenon — input and output: e.g., photoelectric, thermoelectric, electrochemical, electromagnetic…

A signal is a function that conveys information about a phenomenon in the physical world.

• They represent an observable change in quantity over space or time.

• There are:

  • analog signals (time-continuous → they vary smoothly and constantly)

  • digital signals (time-discrete → measured only at specific intervals)

• Sampling is the registration of measured values at discrete points in time. It converts a time-continuous (analog) signal into a time-discrete (digital) signal.

• The sampling rate is the frequency at which a time-continous function is sampled (in Hz) → number of samples per second.

• Biosignals are autonomous signals produced by a living organism measured in physical quantities.

• The human body generates biosignals through cellular activity, they are measurable physically (e.g. temperature).

• Kinematic biosignals — capture motion of points or objects.

• Optical biosignals — measure light or changes in light.

• Chemical biosignals — reflect chemical composition and its temporal changes in body solids, liquids, and gases.

• Electrical biosignals — electric fields are generated in cells (nerve/muscle) and organs.

• Acoustic biosignals — describe the acoustic sounds produced by the body (vibrations and motions).

• Thermal biosignals — temperature measurements show physical and biochemical processes in the organism.

Wearable technologies are small electronic and mobile devices, incorporated into gadgets, accessories, or clothes that can be worn on the human body.

Their two key capabilities are:

• providing monitoring and scanning features via embedded sensor technology

• tracking, analyzing, and transmitting personal data.

• Wrist/hand (e.g., smartwatches, smart rings) → most widely adopted

• Head (e.g., AR/VR glasses, audio headsets)

• Upper-body (e.g., near-body e-patches, on-body smart clothes, in-body implants)

• Lower-body (e.g., smart shoes, smart insoles).

• Data processing & communication challenges — (handling the volume and complexity of sensor data)

• Privacy and security challenges — (sensitive personal data is vulnerable to cyber threats)

• Hardware challenges — (battery life, miniaturization, durability)

• Interoperability challenges — (different devices use different APIs and data formats, making integration difficult)

• Adoption challenges — (getting users to actually use and continue using wearables)

1. Sensor → capture a signal

2. Recording → amplify, filter, and convert the raw signal digitally

3. Processing → remove artifacts and detect events

4. Modeling → analyze events, extract features, recognize and classify patterns

They are both sensors to measure heart activity.

• ECG (electrocardiogram) is an electrical sensor that captures the heart's electrical activity — its key feature is the R-R interval.

• PPG (photoplethysmogram) is an optical sensor that measures changes in blood volume in vessels (blood volume pulse) — its key feature is the inter-beat interval.

• A heartbeat is one complete pulsation of the heart as it sends blood around the body.

  • The heart rate (HR) is the number of heartbeats per minute (BPM).

• The pulse is the regular movement of blood through the body (caused by the beating heart).

  • The pulse rate measures the rate of blood pressure changes in the vessels.

• The Interbeat Interval (IBI) is the time in milliseconds between two individual heartbeats.

  • If a person's average heart rate is 60 BPM, the average IBI would be 1000 ms. But it can vary, some are 800ms or 1200ms apart.

HR (BPM) = 60,000 / IBI (ms)

→ So when the IBI shortens, the heart rate increases.

Heart rate is how fast the heart beats (BPM).

Heart rate variability (HRV) is about how consistently it beats.

• The time between beats is constantly being nudged up or down by the brain via the autonomic nervous system (ANS). → HRV captures exactly that variation.

• IBI (Inter-Beat Interval) = the time gap between two heartbeats

• HRV = how much those IBI values vary from beat to beat

• A low HRV (heart is beating rigidly) is often a sign of stress or fatigue

• Time-domain: quantify the amount of variability in IBI measurements (statistics)

• Frequency-domain: count the amount of low and high frequency beats that occur.

• Non-linear metrics: quantify the unpredictability of the IBI time series

Using those methods to analyze the data, HRV can be measured over 24h, short-term (~5 min), or ultra-short-term (<5 min) windows.

Eye tracking works by pointing a video camera at the eye and use image processing to determine the position.

• The eye tracker needs to identify 2 landmarks:

  • the pupil (black circle)

  • and the corneal reflection (small white circle somewhere on the iris)

• There are 2 techniques: dark pupil and bright pupil techniques

Steps:

• The eye tracker sends out near-infrared light onto the user's eyes

• The light is reflected in the eyes

• The reflections are picked up by the tracker's cameras

• Through filtering and calculations, the eye tracker can capture where one is looking

AOIs are tools in eye tracking research, that lead to quantitative metrics.

• In practice, the user draws a boundary around a feature or element of the eye tracking stimulus.

• Eye tracking metrics are typically extracted based on areas of interest.

They are different measures that can be calculated from the recorded gaze data.

• Fixation → is the maintaining of the visual gaze on a single location (average fixation duration is between 100 and 200ms).

• Saccades → are rapid eye movements between fixation points. They bring new objects of focus to the retina (typical duration is 10ms to 100ms).

Pupillometry is the measurement of pupil size changes. There are 3 types of responses:

• Pupil light response (reflex to brightness changes)

• Pupil near response (focusing on near objects)

• Psychosensory pupil response (driven by cognitive/emotional states)

Recent research shows that pupil size reflects cognitive states such as visual awareness, visual attention, and mental workload. A larger pupil = more dilation, often tied to higher cognitive load or arousal.