Wu, Yu; (2019) Advances in Integrated Circuits and Systems for Wearable Biomedical Electrical Impedance Tomography. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Advances in Integrated Circuits and Systems for Wearable Biomedical Electrical Impedance Tomography.pdf - Accepted Version Download (7MB) | Preview

Abstract

Electrical impedance tomography (EIT) is an impedance mapping technique that can be used to image the inner impedance distribution of the subject under test. It is non-invasive, inexpensive and radiation-free, while at the same time it can facilitate long-term and real-time dynamic monitoring. Thus, EIT lends itself particularly well to the development of a bio-signal monitoring/imaging system in the form of wearable technology. This work focuses on EIT system hardware advancement using complementary metal oxide semiconductor (CMOS) technology. It presents the design and testing of application specific integrated circuit (ASIC) and their successful use in two bio-medical applications, namely, neonatal lung function monitoring and human-machine interface (HMI) for prosthetic hand control. Each year fifteen million babies are born prematurely, and up to 30% suffer from lung disease. Although respiratory support, especially mechanical ventilation, can improve their survival, it also can cause injury to their vulnerable lungs resulting in severe and chronic pulmonary morbidity lasting into adulthood, thus an integrated wearable EIT system for neonatal lung function monitoring is urgently needed. In this work, two wearable belt systems are presented. The first belt features a miniaturized active electrode module built around an analog front-end ASIC which is fabricated with 0.35-µm high-voltage process technology with ±9 V power supplies and occupies a total die area of 3.9 mm². The ASIC offers a high power active current driver capable of up to 6 mAp-p output, and wideband active buffer for EIT recording as well as contact impedance monitoring. The belt has a bandwidth of 500 kHz, and an image frame rate of 107 frame/s. To further improve the system, the active electrode module is integrated into one ASIC. It contains a fully differential current driver, a current feedback instrumentation amplifier (IA), a digital controller and multiplexors with a total die area of 9.6 mm². Compared to the conventional active electrode architecture employed in the first EIT belt, the second belt features a new architecture. It allows programmable flexible electrode current drive and voltage sense patterns under simple digital control. It has intimate connections to the electrodes for the current drive and to the IA for direct differential voltage measurement providing superior common-mode rejection ratio (CMRR) up to 74 dB, and with active gain, the noise level can be reduced by a factor of √3 using the adjacent scan. The second belt has a wider operating bandwidth of 1 MHz and multi-frequency operation. The image frame rate is 122 frame/s, the fastest wearable EIT reported to date. It measures impedance with 98% accuracy and has less than 0.5 Ω and 1° variation across all channels. In addition the ASIC facilitates several other functionalities to provide supplementary clinical information at the bedside. With the advancement of technology and the ever-increasing fusion of computer and machine into daily life, a seamless HMI system that can recognize hand gestures and motions and allow the control of robotic machines or prostheses to perform dexterous tasks, is a target of research. Originally developed as an imaging technique, EIT can be used with a machine learning technique to track bones and muscles movement towards understanding the human user’s intentions and ultimately controlling prosthetic hand applications. For this application, an analog front-end ASIC is designed using 0.35-µm standard process technology with ±1.65 V power supplies. It comprises a current driver capable of differential drive and a low noise (9μVrms) IA with a CMRR of 80 dB. The function modules occupy an area of 0.07 mm². Using the ASIC, a complete HMI system based on the EIT principle for hand prosthesis control has been presented, and the user’s forearm inner bio-impedance redistribution is assessed. Using artificial neural networks, bio-impedance redistribution can be learned so as to recognise the user’s intention in real-time for prosthesis operation. In this work, eleven hand motions are designed for prosthesis operation. Experiments with five subjects show that the system can achieve an overall recognition accuracy of 95.8%.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as CSVXMLJSON

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item