- Kl 10.00: Prøveforelesning: Extraction of Health and Well-being Parameters from Wearable and Smart Home Sensors
- Kl 13.00: Disputas
Prøveforelesning og disputas er åpen for alle interesserte.
Det vil bli en sammenkomst i personalkantina Grevinnen umiddelbart etter disputasen.
PhD Frode Strisland, Sintef IKT
PhD Paul Monaghan, Novartis Pharmaceuticals UK
Førsteamanuensis Kristin Imenes, Institutt for mikrosystemer, HSN
Leder av disputas
Instituttleder Yngvar Berg
Professor Erik A. Johannessen, Institutt for mikrosystemer, HSN
Professor Philipp Häfliger, Institutt for informatikk, UiO
Førsteamanuensis Mehdi Azadmehr, Institutt for mikrosystemer, HSN
Engelskspråklig sammendrag av avhandlingen
Luis André L. Fernandes' thesis presents a novel hydration sensor that employs osmotic pressure to detect different levels of hydration in a proof of concept setting with the aim of implantation in an organism.
The hydration sensor described in this thesis offers an alternate solution to the methods used today for measuring hydration (sampling of body fluids, skin impedance analysis, body weight, etc.). This makes it less complex, portable, smaller, and a low power device capable of conducting continuous measurements. Its low power architecture, in combination with the advances made in microfabrication, make it a suitable candidate for further miniaturization and implantation in the human body.
A prototype (measuring 16 mm in diameter and 8 mm in height) was fabricated in an acetal copolymer acting as a sensor housing incorporating a 25.4 µm thick semi-permeable membrane (Nafion NR211), an electronic interface (TSMC 90 nm ASIC), a resistive pressure transducer (SW415-2, Sensonor, Norway) and a few discrete components assembled on a silicon substrate.
The pressure transducer is in contact with a cavity (reference chamber) where the osmotic pressure is generated from the physical separation of two solutions with different salt concentrations: (i) the solution inside the reference chamber and (ii) the solution of the external media.
The sensor was tested/validated for a ±20% hydration span, ranging from 220 mOsmL-1 (overhydration) to 340 mOsmL-1 (dehydration) and showing a resolution of approximately 2% with a response time of 5.5 hours to reach 100% signal. This result, corresponds to an accuracy of 5.71 bits for the sensor’s dynamic range.
The low power architecture implemented in the ASIC uses a low voltage supply (1 V) in the sub-threshold regime with bias currents of approximately 50 nA resulting in an average power consumption of only 39.4 µW. An Analog-to-Analog approach converts the voltage signal from the pressure transducer (due to a change in resistance) into a frequency output that, for every transition on the output, inverts the power to the transducer. This approach results in a resistance controlled oscillator (RCO), eliminating the use of a power-consuming clock to drive an Analog-to-Digital converter (ADC).
About Luis André L. Fernandes:
Luis André L. Fernandes received his BSc (Hons) degree in Physical Engineering from the University of Lisbon, Portugal, in 2005.
In 2009, he received his MSc degree from the University of Lisbon in collaboration with the Clinical Engineering Department, at the Royal Brompton Hospital NHS, and the International Centre for Circulatory Health (Imperial College), at St Mary’s Hospital NHS, London, UK.
He developed his Ph.D. work at the Department of Micro- and Nanosystems Technology at the University College of Southeast Norway in collaboration with the Department of Informatics at the University of Oslo.