東華大學圖書館 |

語系: 繁體中文

說明(常見問題)

回圖書館首頁

手機版館藏查詢

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

FindBook

Google Book

Amazon

博客來

Machine Learning Approaches for Analyzing Human Mobility in Land and Sea Environments using Wearable Sensors.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Machine Learning Approaches for Analyzing Human Mobility in Land and Sea Environments using Wearable Sensors./
作者:	Choi, Jungyeon.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2022,
面頁冊數:	187 p.
附註:	Source: Dissertations Abstracts International, Volume: 83-12, Section: B.
Contained By:	Dissertations Abstracts International83-12B.
標題:	Information technology. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29165517
ISBN:	9798834000037

Machine Learning Approaches for Analyzing Human Mobility in Land and Sea Environments using Wearable Sensors.
Choi, Jungyeon.

Machine Learning Approaches for Analyzing Human Mobility in Land and Sea Environments using Wearable Sensors. - Ann Arbor : ProQuest Dissertations & Theses, 2022 - 187 p.

Source: Dissertations Abstracts International, Volume: 83-12, Section: B.

Thesis (Ph.D.)--University of Nebraska at Omaha, 2022.

This item must not be sold to any third party vendors.

With the rapid technological advances of recent years, many researchers in numerous scientific fields have effectively used these advances. Wearable sensors are widely used in the healthcare domain to collect physiological and mobile data, enabling health monitoring of various populations in real-world settings. However, it is necessary to validate applications in various environments in order to develop a robust system using wearable sensors. This dissertation used machine learning techniques to understand and analyze human mobility characteristics and to develop predictive models for assessing mobility ability and health status in different environments by finding optimal feature extraction/selection and modeling methods.Our first aim was to investigate a set of effective features for representing human mobility in different applications. Two case studies were conducted to detect the foot-side of walking steps in healthy individuals and assess walking environments in the elderly. We found that lateral features are essential for foot-side classification, whereas variability influences the classification of indoor and outdoor walking. These results could be applicable for tracking disease or rehabilitation progress by identifying a foot-side that affects asymmetrical gait and for understanding walking behaviors in outdoor walking in older adults. The second aim is to examine a practical sensor location to estimate clinical measures for assessing mobility ability in older adults. A pelvis sensor was more efficient at estimating the Timed Up and Go (TUG) Test and the Six Minute Walk Test. This finding allows us to monitor health conditions in older adults remotely. Lastly, we explored the effects of a ship's roll motion on the balance and stability of healthy individuals during walking in simulated sea conditions. The ship's roll motions significantly affected balance- and stability-related gait features, such as step time variability, the center of mass (COM), and the margin of stability (MOS).This dissertation contributes to our understanding of the characteristics of human mobility in various environments and suggests a machine learning-based robust model for monitoring human mobility by providing biomechanical and clinical assessments with wearable sensors.

ISBN: 9798834000037Subjects--Topical Terms:

532993
Information technology.
Subjects--Index Terms:

Deep learning

Machine Learning Approaches for Analyzing Human Mobility in Land and Sea Environments using Wearable Sensors.
LDR:03580nmm a2200421 4500 001 2352240
005 20221118093852.5
008 241004s2022 ||||||||||||||||| ||eng d
020 $a 9798834000037
035 $a (MiAaPQ)AAI29165517
035 $a AAI29165517
040 $a MiAaPQ $c MiAaPQ
100 1 $a Choi, Jungyeon. $3 3691860
245 1 0 $a Machine Learning Approaches for Analyzing Human Mobility in Land and Sea Environments using Wearable Sensors.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2022
300 $a 187 p.
500 $a Source: Dissertations Abstracts International, Volume: 83-12, Section: B.
500 $a Advisor: Youn, Jong-Hoon.
502 $a Thesis (Ph.D.)--University of Nebraska at Omaha, 2022.
506 $a This item must not be sold to any third party vendors.
520 $a With the rapid technological advances of recent years, many researchers in numerous scientific fields have effectively used these advances. Wearable sensors are widely used in the healthcare domain to collect physiological and mobile data, enabling health monitoring of various populations in real-world settings. However, it is necessary to validate applications in various environments in order to develop a robust system using wearable sensors. This dissertation used machine learning techniques to understand and analyze human mobility characteristics and to develop predictive models for assessing mobility ability and health status in different environments by finding optimal feature extraction/selection and modeling methods.Our first aim was to investigate a set of effective features for representing human mobility in different applications. Two case studies were conducted to detect the foot-side of walking steps in healthy individuals and assess walking environments in the elderly. We found that lateral features are essential for foot-side classification, whereas variability influences the classification of indoor and outdoor walking. These results could be applicable for tracking disease or rehabilitation progress by identifying a foot-side that affects asymmetrical gait and for understanding walking behaviors in outdoor walking in older adults. The second aim is to examine a practical sensor location to estimate clinical measures for assessing mobility ability in older adults. A pelvis sensor was more efficient at estimating the Timed Up and Go (TUG) Test and the Six Minute Walk Test. This finding allows us to monitor health conditions in older adults remotely. Lastly, we explored the effects of a ship's roll motion on the balance and stability of healthy individuals during walking in simulated sea conditions. The ship's roll motions significantly affected balance- and stability-related gait features, such as step time variability, the center of mass (COM), and the margin of stability (MOS).This dissertation contributes to our understanding of the characteristics of human mobility in various environments and suggests a machine learning-based robust model for monitoring human mobility by providing biomechanical and clinical assessments with wearable sensors.
590 $a School code: 1060.
650 4 $a Information technology. $3 532993
650 4 $a Physiology. $3 518431
650 4 $a Computer science. $3 523869
650 4 $a Artificial intelligence. $3 516317
650 4 $a Information science. $3 554358
650 4 $a Health sciences. $3 3168359
650 4 $a Kinesiology. $3 517627
653 $a Deep learning
653 $a Gait analysis
653 $a Human mobility
653 $a Machine learning
653 $a Wearable sensors
690 $a 0489
690 $a 0984
690 $a 0800
690 $a 0723
690 $a 0719
690 $a 0566
690 $a 0575
710 2 $a University of Nebraska at Omaha. $b Information Technology. $3 1043963
773 0 $t Dissertations Abstracts International $g 83-12B.
790 $a 1060
791 $a Ph.D.
792 $a 2022
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29165517