Design and Characterization of Soft Pressure-Transduction Sensors for Pulsatile Luminal Pressure Measurement

Sammanfattning

Continuous local blood pressure measurement can support cardiovascular monitoring and the control of implantable cardiac support devices. Extravascular sensors are promising because they can measure vessel-wall deformation without direct blood contact, reducing blood-exposure risks while still providing local pressure data. This thesis investigates soft pressure-transduction sensors for strain-based luminal pressure sensing in flexible vessels. Two designs were developed and tested in an in-vitro mock circulatory loop: an embedded design mainly responding to circumferential vessel-wall tension (Design I), and an A-frame design converting vessel expansion into radial compression on the sensor (Design II). Both designs were fabricated using Dragon Skin 30 silicone and integrated with a BMP384 MEMS pressure sensor. Their performance was evaluated through simulation, static step tests, frequency-sweep tests, held-out fixed-frequency validation, physiological aortic waveform reconstruction, and repeated assembly tests. A clinical pressure transducer was used as the reference. For pulsatile pressure reconstruction, a combined calibration method was developed using transfer-function-based dynamic compensation followed by Yeoh-based nonlinear pressure conversion. Static calibration alone was insufficient for pulsatile pressure measurement, especially at higher frequencies. The combined method reduced phase delay, frequency-dependent hysteresis, and nonlinear error. Both designs were able to reconstruct pulsatile pressure and physiological aortic wave forms with good accuracy. Design I showed more stable overall performance, while Design II showed strong waveform correlation but larger systolic and diastolic pressure errors, possibly due to its reduced usable output range under compression. Overall, the results show that soft pressure-transduction sensors can reconstruct pulsatile luminal pressure under mock-loop conditions. The study highlights the importance of dynamic calibration, sensor output range, and assembly reliability. Future work should improve packaging and fixation, reduce assembly sensitivity, assess long-term calibration stability, and validate the method in biological phantoms, ex vivo tissue, and in vivo conditions.