Powering intracranial pressure sensor implants via piezoelectric energy harvesting
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Program
Biomedical engineering (MPBME), MSc
Publicerad
2023
Författare
Erndin Dahlberg, Charlie
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Intracranial pressure monitoring is conventionally done via an external
ventricular drain or an implant probe where a wire protrudes through the skull.
Such methods are cumbersome, costly, highly invasive and they constitute a
significant risk of infection with each procedure. For long-term intracranial
pressure monitoring, new types of implants which can be left inside of the
patient are required in order to minimize the amount of surgery. Such implants
can not be powered by batteries as batteries need to be replaced when they
have been discharged. This thesis instead turns towards energy harvesting,
and investigates whether energy can be harvested from the human brain by
the means of a piezoelectric cantilever harvester, in order to power a pressure
sensor implant. A literature review was conducted in order to examine the
current state-of-the-art of energy harvesting inside the human body. A PZT
(lead zirconate titanate) cantilever harvester was simulated together with a
simple brain model in order to see how much power can be generated by
placing the harvester on the brain surface. Results show that a simple 7x7 mm2
thin plate bimorph PZT harvester can generate up to 2.6 nW. Additionally, a
phantom brain test rig was constructed and pressurized in order to mimic the
pressure dynamics inside the human brain. The system power requirements for
a sensor implant with a transceiver would however amount to around 300 µW.
Thus, a wake-up receiver system should be used where measurements are only
taken sparsely for short durations. The power loss of supercapacitor leakage
currents sets the lower limit of required harvesting output to roughly 1 µW.
Meanwhile, the upper limit of what can be produced by a PZT brain harvester
is 1.4 µW. Thus, powering a sensor implant by piezoelectric brain harvesting
is theoretically possible, but the harvester must be very well optimized for the
task.