Biosignal

Biosignals are time-varying signals produced by living organisms that can be measured and analyzed to infer physiological state or function. They originate from electrical, mechanical, chemical, or optical processes and are captured using a range of sensors. Common examples include electrocardiography (ECG), electroencephalography (EEG), electromyography (EMG), photoplethysmography (PPG), respiration signals, skin conductance, and body temperature. The precise signal and measurement method depend on the biological source, anatomical location, and the intended application.

Measurement relies on electrodes, probes, or optical sensors placed on the body, with data acquisition systems that sample at frequencies appropriate to the signal (for example, thousands of hertz for EEG, tens or hundreds for ECG). Data quality is affected by movement, sensor impedance, ambient noise, and device limitations. Preprocessing steps such as filtering, artifact removal, detrending, and normalization are commonly applied before analysis.

Analysis and processing range from simple time-domain statistics and frequency analysis to advanced machine learning and neuroscience methods. Features may include heart rate, heart rate variability, spectral power, event-related potentials, or EMG burst characteristics. In many contexts, biosignals are interpreted in conjunction with clinical data or other modalities.

Applications span medical diagnosis and monitoring (critical care, cardiology, neurology), rehabilitation, human–computer interaction, sports science, sleep research, and consumer wearables. They enable screening, movement assessment, brain–computer interfaces, and personalized health feedback.

Challenges include noise and artifacts, inter- and intra-individual variability, device drift, and data privacy concerns. Standards and interoperability efforts address sensor calibration, sampling rates, and data formats to facilitate sharing and reproducibility.