Back to search

IKTPLUSS-IKT og digital innovasjon

MEDICAL SENSING, LOCALIZATION, AND COMMUNICATION USING ULTRA WIDEBAND TECHNOLOGY II

Awarded: NOK 9.3 mill.

Medical Sensing, Localization, and Communications using Ultra Wideband Technology (MELODY II) investigated using the ultra wideband (UWB) frequencies to develop novel wireless health technology, for both improved network communication and improved and novel medical applications. Two major research directions were pursued, namely short-range sensing and imaging and wireless channel characterization and signal processing for communications, where the developed prototypes were tested both in phantoms and living animal experiments. The research on UWB sensing/imaging of objects and tissues inside the human body by the use of electromagnetic waves. The full microwave region was investigated to determine tradeoffs between high penetrations into human tissue at lower frequencies and narrow focus of beams at higher frequencies. The sensing focused on measuring vital signs parameters such as pulse, heart rate, and blood pressure using different techniques ranging from simple continuous wave radars to complex imaging systems. Novel implant sensor communications considered developing methods for in-body channel models and detailed simulation studies. Moreover, three experiments performed on animal subjects, where they have tested the transceivers and in-body antennas placed in different depths and regions (e.g. chest, below heart, lungs, and stomach). The animal study was approved by the animal welfare authorities in Norway and has been performed by experienced surgeons and staff at the Oslo University Hospital. The new methods to wirelessly communicate with implants have patented with commercial activities are planned with the industry. The main contribution has been to extract the vital signs in noninvasive ways. For instance, this can be applicable to heart rate estimation from a camera while recording a person?s face or detecting cancers using capillary density evaluation (under investigation). These applications lead to mathematical problems. This is investigated and two approaches were proposed in "Extracting remote photoplethysmogram signal from endoscopy videos for vessel and capillary density recognition" and "Accurate Heart Rate Estimation from Camera Recording via MUSIC Algorithm". The project was headed by Oslo University Hospital (OUS) and had Norwegian Defense Research Establishment (FFI) and Norwegian University of Science and Technology (NTNU) as national partners. Furthermore, there were two company partners and four international academic partners. The four-year project received funding from the Research Council of Norway (RCN) for 1 PhD fellowship (3 man years), 3 man years of Postdoc fellowships, and 4 guest researcher fellowships. Moreover, the project received 4 man years of ERCIM Postdoc fellowships from NTNU, 2 man years of PhD scholarship from FFI, and 1 man year of Postdoc fellowship from OUS as their own contribution. The total man-years equal 17. The project manager, three principle scientists and the project coordinator have been the core project team supported by the steering group. All partners have given access to their laboratories to design and develop prototypes, where testing and evaluation have been performed as part of animal experiments at the Oslo University Hospital. The scientific publications account 29 journal papers, 3 book chapters, 46 peer-reviewed conference papers, and 10 abstracts (publication list is available at http://www.melody-project.info). Thanks to international collaboration and guest researchers we have a larger number of publications. The project has been granted one patent and has submitted a second patent application. As part of the research and development collaboration between OUS and OmniVision Technologies AS two industrial PhD fellows have been recruited. A spin-off project based on new research questions arising from the MELODY was approved for funding by the European Commission under the H2020-MARIE Skodowska-CURIE ACTIONS (MSCA-ITN-2015) in May 2015. The project Wireless In-Body Environment Communications (WiBEC) is coordinated by the OUS and has 8 partners from Norway, France, Germany and Spain. The project has 16 PhD students and will run for 4 years from 2016. Our PhD and Postdoc researchers have had research visits spanning from 1 to 9 months aboard while we have received 7 guest researchers from abroad spending 1-14 months in Norway. The project has co-organized special sessions in major international conferences and was very successful in communicating the results to public through newspaper articles and social media. http://www.melody-project.info

The project addresses the use of ultra wideband (UWB) technology for improved wireless sensor communication from in-body to on-body devices and short range sensing of vital signs and tissue imaging. In-body channel models proposed in the MELODY I will be revised and improved using physical measurements performed in animal subjects. A simplified transceiver design will be considered for high quality video transmission incorporating the region-of-interest coding to reduce the transmission cost. The region- of-interest will be estimated using reflected visible light from the camera source in the capsule video endoscope, where we anticipate to use a combination of independent component analysis and Green color component from the video stream - "photo-plethys mography". Video frames with region-of-interest will be coded differentially while other frames will not be transmitted or compressed heavily. A low complexity video encoder will be optimized together with decoders incorporating frame interpolation and p ost processing for relevant video material. Based on transmission measurements of the abdominal channel, practical orthogonal transmission schemes will be selected to avoid, or simplify channel equalization while minimizing the transmission energy per bit and keeping the computational cost and power consumption as small as possible. The detection is executed by several cooperating on-body receivers. A simple return channel will be included for power control and parameter selection. Remote sensing for vita l signs and imaging tissues will be performed in combination of multichannel switched antenna array and angular and spatial depths. Imaging radar will use advanced signal processing for measuring specific tissue characteristics using micro Doppler analysi s of a given resolution cell in a sequence of images. Practical validations will be done in phantoms and animal experiments.

Funding scheme:

IKTPLUSS-IKT og digital innovasjon