Low Power Circuits for Smart Flexible ECG Sensors
| dc.contributor.advisor | Lian, Yong Peter | |
| dc.contributor.author | Zhao, Yang | |
| dc.date.accessioned | 2020-05-11T12:35:46Z | |
| dc.date.available | 2020-05-11T12:35:46Z | |
| dc.date.copyright | 2019-08 | |
| dc.date.issued | 2020-05-11 | |
| dc.date.updated | 2020-05-11T12:35:46Z | |
| dc.degree.discipline | Computer Science | |
| dc.degree.level | Doctoral | |
| dc.degree.name | PhD - Doctor of Philosophy | |
| dc.description.abstract | Cardiovascular diseases (CVDs) are the world leading cause of death. In-home heart condition monitoring effectively reduced the CVD patient hospitalization rate. Flexible electrocardiogram (ECG) sensor provides an affordable, convenient and comfortable in-home monitoring solution. The three critical building blocks of the ECG sensor i.e., analog frontend (AFE), QRS detector, and cardiac arrhythmia classifier (CAC), are studied in this research. A fully differential difference amplifier (FDDA) based AFE that employs DC-coupled input stage increases the input impedance and improves CMRR. A parasitic capacitor reuse technique is proposed to improve the noise/area efficiency and CMRR. An on-body DC bias scheme is introduced to deal with the input DC offset. Implemented in 0.35m CMOS process with an area of 0.405mm2, the proposed AFE consumes 0.9W at 1.8V and shows excellent noise effective factor of 2.55, and CMRR of 76dB. Experiment shows the proposed AFE not only picks up clean ECG signal with electrodes placed as close as 2cm under both resting and walking conditions, but also obtains the distinct -wave after eye blink from EEG recording. A personalized QRS detection algorithm is proposed to achieve an average positive prediction rate of 99.39% and sensitivity rate of 99.21%. The user-specific template avoids the complicate models and parameters used in existing algorithms while covers most situations for practical applications. The detection is based on the comparison of the correlation coefficient of the user-specific template with the ECG segment under detection. The proposed one-target clustering reduced the required loops. A continuous-in-time discrete-in-amplitude (CTDA) artificial neural network (ANN) based CAC is proposed for the smart ECG sensor. The proposed CAC achieves over 98% classification accuracy for 4 types of beats defined by AAMI (Association for the Advancement of Medical Instrumentation). The CTDA scheme significantly reduces the input sample numbers and simplifies the sample representation to one bit. Thus, the number of arithmetic operations and the ANN structure are greatly simplified. The proposed CAC is verified by FPGA and implemented in 0.18m CMOS process. Simulation results show it can operate at clock frequencies from 10KHz to 50MHz. Average power for the patient with 75bpm heart rate is 13.34W. | |
| dc.identifier.uri | https://hdl.handle.net/10315/37346 | |
| dc.language | en | |
| dc.rights | Author owns copyright, except where explicitly noted. Please contact the author directly with licensing requests. | |
| dc.subject | Biomedical engineering | |
| dc.subject.keywords | Electrocardiogram (ECG) | |
| dc.subject.keywords | Analog Front-end | |
| dc.subject.keywords | QRS detector | |
| dc.subject.keywords | Cardiac Arrhythmia Classifier | |
| dc.subject.keywords | Low power | |
| dc.subject.keywords | Common mode rejection ratio (CMRR) | |
| dc.subject.keywords | Noise efficiency factor (NEF) | |
| dc.subject.keywords | Application specific integrated circuits (ASIC) | |
| dc.subject.keywords | DC-coupled | |
| dc.subject.keywords | Patient-specific | |
| dc.subject.keywords | Classification accuracy | |
| dc.subject.keywords | Low noise | |
| dc.subject.keywords | Flexible ECG sensor | |
| dc.subject.keywords | Sensor interface circuits | |
| dc.title | Low Power Circuits for Smart Flexible ECG Sensors | |
| dc.type | Electronic Thesis or Dissertation |
Files
Original bundle
1 - 1 of 1