{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/24518"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/24518","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Embedded hardware architecture for multi-parameter physiological signal monitoring","abstract":"Advancements in device technology along with development of robust and sophisticated signal processing algorithms have attracted considerable attention in realization of portable devices capable of performing robust patient health monitoring with highly accurate abnormal event detections. Significant challenges are faced when such biomedical monitoring techniques along with multi-dimensional signal processing capabilities are to be realized in portable platforms due to area, power, and real-time performance constraints in hardware platforms. This thesis focuses on achieving balanced and efficient hardware-based solutions for health monitoring techniques that provide reasonably accurate abnormal event detection results but at the same time are cost-effective from the hardware realization perspective. The first solution offered in this thesis presents design, implementation and evaluation of an efficient embedded hardware for accurate automated detection of epileptic seizures. Three field programmable gate array (FPGA)-based hardware configurations are proposed and evaluated in terms of accuracy of detection, utilization of hardware resources, and power consumption. The results show that a solution based on combination of the statistical function of variance (for feature extraction) and an artificial neural network (ANN) classifier allows one to achieve high detection accuracy (99.18%) with moderate hardware footprint (around 44% of the FPGA resources). Furthermore, use of algorithmic and architectural optimization techniques (reduction in precision of the fixed-point number representation and reuse of hardware components) allows reducing the hardware footprint by a factor of 4.4 and power consumption by a factor of 2.7 as compared with an un-optimized hardware configuration. The second approach proposes an architectural framework for patient-specific physiological abnormality detection utilizing concurrent analysis of multiple different sensor data streams. This design, termed as Multi-Parameter Signature-Based Health Monitoring Architecture (MSHMA), is capable of performing abnormality detection analysis in near real-time, is interactive and configurable by the user in run-time, and is simple enough to be suitable for hardware implementation. Initial evaluations show abnormal event detection true positive rate of up to 94.74% for arterial blood pressure (ABP) and heart rate (HR) signals using our Mean Analysis method and about 84.48% using the Error Analysis technique which also demonstrates possible early detection capability of about 6.21 minutes. A simplified version of the MSHMA is implemented in a FPGA-based platform for prototype purpose. This hardware implementation uses only 22% slice registers and 33% slice look-up table in the FPGA platform while consuming 154.22 mW of power. An effort to reduce the power consumption by lowering the system clock frequency shows power usage reduction by up to 2.96X compared to the original implementation.","abstract_html":"Advancements in device technology along with development of robust and sophisticated signal processing algorithms have attracted considerable attention in realization of portable devices capable of performing robust patient health monitoring with highly accurate abnormal event detections. Significant challenges are faced when such biomedical monitoring techniques along with multi-dimensional signal processing capabilities are to be realized in portable platforms due to area, power, and real-time performance constraints in hardware platforms. This thesis focuses on achieving balanced and efficient hardware-based solutions for health monitoring techniques that provide reasonably accurate abnormal event detection results but at the same time are cost-effective from the hardware realization perspective. The first solution offered in this thesis presents design, implementation and evaluation of an efficient embedded hardware for accurate automated detection of epileptic seizures. Three field programmable gate array (FPGA)-based hardware configurations are proposed and evaluated in terms of accuracy of detection, utilization of hardware resources, and power consumption. The results show that a solution based on combination of the statistical function of variance (for feature extraction) and an artificial neural network (ANN) classifier allows one to achieve high detection accuracy (99.18%) with moderate hardware footprint (around 44% of the FPGA resources). Furthermore, use of algorithmic and architectural optimization techniques (reduction in precision of the fixed-point number representation and reuse of hardware components) allows reducing the hardware footprint by a factor of 4.4 and power consumption by a factor of 2.7 as compared with an un-optimized hardware configuration. The second approach proposes an architectural framework for patient-specific physiological abnormality detection utilizing concurrent analysis of multiple different sensor data streams. This design, termed as Multi-Parameter Signature-Based Health Monitoring Architecture (MSHMA), is capable of performing abnormality detection analysis in near real-time, is interactive and configurable by the user in run-time, and is simple enough to be suitable for hardware implementation. Initial evaluations show abnormal event detection true positive rate of up to 94.74% for arterial blood pressure (ABP) and heart rate (HR) signals using our Mean Analysis method and about 84.48% using the Error Analysis technique which also demonstrates possible early detection capability of about 6.21 minutes. A simplified version of the MSHMA is implemented in a FPGA-based platform for prototype purpose. This hardware implementation uses only 22% slice registers and 33% slice look-up table in the FPGA platform while consuming 154.22 mW of power. An effort to reduce the power consumption by lowering the system clock frequency shows power usage reduction by up to 2.96X compared to the original implementation.","abstract_has_math":false,"creators":["Saleheen, Mushfiq U."],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Electrical & Computer Engr","degree_department":null,"school":null,"contributors":["Iyer, Ravishankar K."],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2011,"date_issued":"2011-05-25T14:26:33Z","date_published":"2011-05-25T14:26:33Z","updated_at":"2026-07-22T22:25:23Z","subjects":["Physiological Monitoring","Embedded Hardware","Multi-Parameter","Statistical Signal Processing"],"languages":["en"],"rights":["Copyright 2011 Mushfiq U. Saleheen"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/24518","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Iyer, Ravishankar K."]},{"key":"dc:creator","label":"Author","values":["Saleheen, Mushfiq U."]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2011-05-25T14:26:33Z","2013-05-26T10:00:21Z","2011-05"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engr"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Physiological Monitoring","Embedded Hardware","Multi-Parameter","Statistical Signal Processing"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2011 Mushfiq U. Saleheen"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/24518"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Advancements in device technology along with development of robust and sophisticated signal processing algorithms have attracted considerable attention in realization of portable devices capable of performing robust patient health monitoring with highly accurate abnormal event detections. Significant challenges are faced when such biomedical monitoring techniques along with multi-dimensional signal processing capabilities are to be realized in portable platforms due to area, power, and real-time performance constraints in hardware platforms. This thesis focuses on achieving balanced and efficient hardware-based solutions for health monitoring techniques that provide reasonably accurate abnormal event detection results but at the same time are cost-effective from the hardware realization perspective. The first solution offered in this thesis presents design, implementation and evaluation of an efficient embedded hardware for accurate automated detection of epileptic seizures. Three field programmable gate array (FPGA)-based hardware configurations are proposed and evaluated in terms of accuracy of detection, utilization of hardware resources, and power consumption. The results show that a solution based on combination of the statistical function of variance (for feature extraction) and an artificial neural network (ANN) classifier allows one to achieve high detection accuracy (99.18%) with moderate hardware footprint (around 44% of the FPGA resources). Furthermore, use of algorithmic and architectural optimization techniques (reduction in precision of the fixed-point number representation and reuse of hardware components) allows reducing the hardware footprint by a factor of 4.4 and power consumption by a factor of 2.7 as compared with an un-optimized hardware configuration. The second approach proposes an architectural framework for patient-specific physiological abnormality detection utilizing concurrent analysis of multiple different sensor data streams. This design, termed as Multi-Parameter Signature-Based Health Monitoring Architecture (MSHMA), is capable of performing abnormality detection analysis in near real-time, is interactive and configurable by the user in run-time, and is simple enough to be suitable for hardware implementation. Initial evaluations show abnormal event detection true positive rate of up to 94.74% for arterial blood pressure (ABP) and heart rate (HR) signals using our Mean Analysis method and about 84.48% using the Error Analysis technique which also demonstrates possible early detection capability of about 6.21 minutes. A simplified version of the MSHMA is implemented in a FPGA-based platform for prototype purpose. This hardware implementation uses only 22% slice registers and 33% slice look-up table in the FPGA platform while consuming 154.22 mW of power. An effort to reduce the power consumption by lowering the system clock frequency shows power usage reduction by up to 2.96X compared to the original implementation.","Item withdrawn by Mark Zulauf (zulauf@illinois.edu) on 2011-03-03T21:11:21Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 1 Saleheen_Mushfiq.pdf: 1281954 bytes, checksum: b517a9bdbcb0134300d3783dc35fa397 (MD5)","Made available in DSpace on 2011-05-25T14:26:33Z (GMT). No. of bitstreams: 2 Saleheen_Mushfiq.pdf: 1281954 bytes, checksum: b517a9bdbcb0134300d3783dc35fa397 (MD5) license.txt: 4066 bytes, checksum: 8475749b1fc6fb15efeec576755988c6 (MD5)","Item marked as restricted to the 'Administrator' Group (id=1) by William Ingram (wingram2@illinois.edu) on 2011-05-25T14:30:14Z Item is restricted until 2013-05-25T14:29:35Z","Item reinstated by Sarah Shreeves (sshreeve@illinois.edu) on 2013-05-26T10:00:21Z Item was in collections: University of Illinois Dissertations and Theses (ID: 204) Dissertations and Theses - Electrical and Computer Engineering (ID: 446) No. of bitstreams: 3 Saleheen_Mushfiq.pdf.txt: 147254 bytes, checksum: 5e1ae1dbfad3ea574fddc1b262e856f5 (MD5) Saleheen_Mushfiq.pdf: 1281954 bytes, checksum: b517a9bdbcb0134300d3783dc35fa397 (MD5) license.txt: 4066 bytes, checksum: 8475749b1fc6fb15efeec576755988c6 (MD5)","Item released from any restrictions by Sarah Shreeves (sshreeve@illinois.edu) on 2013-05-26T10:00:21Z"]},{"key":"dc:title","label":"Title","values":["Embedded hardware architecture for multi-parameter physiological signal monitoring"]}]}],"canonical_facts":{"dc:contributor":["Iyer, Ravishankar K."],"dc:creator":["Saleheen, Mushfiq U."],"dc:date":["2011-05-25T14:26:33Z","2013-05-26T10:00:21Z","2011-05"],"dc:description":["Advancements in device technology along with development of robust and sophisticated signal processing algorithms have attracted considerable attention in realization of portable devices capable of performing robust patient health monitoring with highly accurate abnormal event detections. Significant challenges are faced when such biomedical monitoring techniques along with multi-dimensional signal processing capabilities are to be realized in portable platforms due to area, power, and real-time performance constraints in hardware platforms. This thesis focuses on achieving balanced and efficient hardware-based solutions for health monitoring techniques that provide reasonably accurate abnormal event detection results but at the same time are cost-effective from the hardware realization perspective. The first solution offered in this thesis presents design, implementation and evaluation of an efficient embedded hardware for accurate automated detection of epileptic seizures. Three field programmable gate array (FPGA)-based hardware configurations are proposed and evaluated in terms of accuracy of detection, utilization of hardware resources, and power consumption. The results show that a solution based on combination of the statistical function of variance (for feature extraction) and an artificial neural network (ANN) classifier allows one to achieve high detection accuracy (99.18%) with moderate hardware footprint (around 44% of the FPGA resources). Furthermore, use of algorithmic and architectural optimization techniques (reduction in precision of the fixed-point number representation and reuse of hardware components) allows reducing the hardware footprint by a factor of 4.4 and power consumption by a factor of 2.7 as compared with an un-optimized hardware configuration. The second approach proposes an architectural framework for patient-specific physiological abnormality detection utilizing concurrent analysis of multiple different sensor data streams. This design, termed as Multi-Parameter Signature-Based Health Monitoring Architecture (MSHMA), is capable of performing abnormality detection analysis in near real-time, is interactive and configurable by the user in run-time, and is simple enough to be suitable for hardware implementation. Initial evaluations show abnormal event detection true positive rate of up to 94.74% for arterial blood pressure (ABP) and heart rate (HR) signals using our Mean Analysis method and about 84.48% using the Error Analysis technique which also demonstrates possible early detection capability of about 6.21 minutes. A simplified version of the MSHMA is implemented in a FPGA-based platform for prototype purpose. This hardware implementation uses only 22% slice registers and 33% slice look-up table in the FPGA platform while consuming 154.22 mW of power. An effort to reduce the power consumption by lowering the system clock frequency shows power usage reduction by up to 2.96X compared to the original implementation.","Item withdrawn by Mark Zulauf (zulauf@illinois.edu) on 2011-03-03T21:11:21Z Item was in collections: University of Illinois Theses & Dissertations (ID: 1) No. of bitstreams: 1 Saleheen_Mushfiq.pdf: 1281954 bytes, checksum: b517a9bdbcb0134300d3783dc35fa397 (MD5)","Made available in DSpace on 2011-05-25T14:26:33Z (GMT). No. of bitstreams: 2 Saleheen_Mushfiq.pdf: 1281954 bytes, checksum: b517a9bdbcb0134300d3783dc35fa397 (MD5) license.txt: 4066 bytes, checksum: 8475749b1fc6fb15efeec576755988c6 (MD5)","Item marked as restricted to the 'Administrator' Group (id=1) by William Ingram (wingram2@illinois.edu) on 2011-05-25T14:30:14Z Item is restricted until 2013-05-25T14:29:35Z","Item reinstated by Sarah Shreeves (sshreeve@illinois.edu) on 2013-05-26T10:00:21Z Item was in collections: University of Illinois Dissertations and Theses (ID: 204) Dissertations and Theses - Electrical and Computer Engineering (ID: 446) No. of bitstreams: 3 Saleheen_Mushfiq.pdf.txt: 147254 bytes, checksum: 5e1ae1dbfad3ea574fddc1b262e856f5 (MD5) Saleheen_Mushfiq.pdf: 1281954 bytes, checksum: b517a9bdbcb0134300d3783dc35fa397 (MD5) license.txt: 4066 bytes, checksum: 8475749b1fc6fb15efeec576755988c6 (MD5)","Item released from any restrictions by Sarah Shreeves (sshreeve@illinois.edu) on 2013-05-26T10:00:21Z"],"dc:identifier":["http://hdl.handle.net/2142/24518"],"dc:language":["en"],"dc:rights":["Copyright 2011 Mushfiq U. Saleheen"],"dc:subject":["Physiological Monitoring","Embedded Hardware","Multi-Parameter","Statistical Signal Processing"],"dc:title":["Embedded hardware architecture for multi-parameter physiological signal monitoring"],"thesis:degree_discipline":["Electrical & Computer Engr"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:25:23Z"}