Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/164024
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSeenivasan Sindhu Barathien_US
dc.date.accessioned2023-01-04T01:40:01Z-
dc.date.available2023-01-04T01:40:01Z-
dc.date.issued2022-
dc.identifier.citationSeenivasan Sindhu Barathi (2022). Development of machine learning models for vital signs monitoring. Master's thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/164024en_US
dc.identifier.urihttps://hdl.handle.net/10356/164024-
dc.description.abstractBlood pressure is an important clinical vital sign nowadays and it is recommended to monitor the blood pressure measurements daily to reduce the risks of high blood pressure. Generally, Blood pressure measurements are made using contact and non-contact methods. This study proposes a non-contact blood pressure measurement which uses two frequency modulated continuous wave (FMCW) mm-wave radars to detect the heart rate, and breathing rate, and a cuff-based OMRON device for blood pressure prediction. Databases were created and collected from 53 subjects using mm-wave radar that can extract the chest and neck pulse waveforms. In this experiment, the systolic blood pressure (SBP) and diastolic blood pressure (DBP) data were collected from the subject using an OMRON device. The datasets have been pre-processed with signal processing and a total of 54 features have been extracted. The processed data is trained with a deep neural network model with the Bonobo optimization algorithm for accurate prediction of both SBP and DBP. Moreover, this study compares the other machine learning models and achieves an RMSE value of SBP of 0.3068 and an RMSE value of DBP of 0.2282. The model has been tested for 15 subjects with 5 pulse waveforms (150 seconds) and 1 pulse waveform (30 seconds) at a distance of 0.5-1 metre using radar and blood pressure results are compared with the OMRON device. The tested results achieve a (Mean Absolute error ± Standard deviation) of SBP as 2.267±1.340 mmHg and DBP as 2.433±1.616 mmHg, which meets the AAMI requirements. Hence, while comparing with other models, our proposed study outperforms the results and achieves an overall accuracy of 93%.en_US
dc.language.isoenen_US
dc.publisherNanyang Technological Universityen_US
dc.subjectEngineering::Electrical and electronic engineeringen_US
dc.titleDevelopment of machine learning models for vital signs monitoringen_US
dc.typeThesis-Master by Courseworken_US
dc.contributor.supervisorMuhammad Faeyz Karimen_US
dc.contributor.schoolSchool of Electrical and Electronic Engineeringen_US
dc.description.degreeMaster of Science (Computer Control and Automation)en_US
dc.contributor.researchSingapore Centre for Environmental Life Sciences and Engineering (SCELSE)en_US
dc.contributor.supervisoremailfaeyz@ntu.edu.sgen_US
item.grantfulltextembargo_restricted_20250103-
item.fulltextWith Fulltext-
Appears in Collections:EEE Theses
Files in This Item:
File Description SizeFormat 
SEENIVASAN SINDHU BARATHI_DISSERTATION_final.pdf
  Until 2025-01-03
3.08 MBAdobe PDFUnder embargo until Jan 03, 2025

Page view(s)

174
Updated on May 23, 2024

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.