Bioinstrumentation

Bioinstrumentation is the engineering discipline of measuring biological and physiological quantities. Examples include ECG, EEG, EMG, blood pressure, oxygen saturation, temperature, flow, respiratory motion, glucose concentration, force, gait, imaging signals, and implant telemetry. The measured variable may be electrical, mechanical, optical, chemical, thermal, or acoustic.

Engineering role

The central challenge is that biological signals are often small, variable, noisy, and patient-dependent. A measurement chain must convert a physiological event into a reliable signal without harming the patient, distorting the variable, or misrepresenting the clinical meaning. This requires careful design of sensors, electrodes, amplifiers, filters, isolation barriers, power supplies, software, user interfaces, and calibration procedures.

Signal chain

A typical bioinstrumentation system includes a transducer, analog front end, filtering, amplification, analog-to-digital conversion, digital processing, display or storage, and sometimes a communication link. Each stage has error sources: electrode motion, contact impedance, biological artefacts, thermal drift, optical scattering, electromagnetic interference, quantization, aliasing, and algorithmic assumptions.

Safety and validation

Patient-connected equipment must control leakage current, isolation, fault conditions, biocompatibility, sterilization effects, mechanical hazards, and usability risks. Validation should include calibration against traceable references, testing over the expected physiological range, artefact rejection checks, environmental testing, and clinical or simulated-use evidence where appropriate.

Design considerations

A high-quality design states the measurand, anatomical location, sampling rate, bandwidth, accuracy, resolution, response time, operating environment, and acceptable artefacts. The engineer must distinguish raw signal quality from clinical usefulness. A signal can be electrically clean but clinically irrelevant if sensor placement, population variability, or interpretation logic is wrong.

Common mistakes

Common mistakes include treating the human body as a simple signal source, ignoring motion artefacts, designing filters that remove clinically important content, and testing only on ideal bench signals. Another serious mistake is assuming that a consumer-grade measurement principle is automatically suitable for diagnosis, monitoring, or therapy without validation and risk analysis.