Osteoarthritis (OA) is the leading cause of disability among the aging population in the United States and is frequently treated by replacing deteriorated joints with metal and plastic components. Developing better quantitative measures of movement quality to track patients longitudinally in their own homes would enable personalized treatment plans and hasten the advancement of promising new interventions. Wearable sensors and machine learning used to quantify patient movement could revolutionize the diagnosis and treatment of movement disorders. The purpose of this dissertation was to overcome technical challenges associated with the use of wearable sensors, specifically Inertial Measurement Units (IMUs), as a diagnostic tool for osteoarthritic (OA) and total knee replacement patients (TKR) through a detailed biomechanical assessment and development of machine learning algorithms. Specifically, the first study developed a relevant dataset consisting of IMU and associated biomechanical parameters of OA and TKR patients performing various activities, created a machine learning-based framework to accurately estimate spatiotemporal movement characteristics from IMU during level ground walking, and defined optimum sensor configuration associated with the patient population and activity. The second study designed a framework to generate synthetic kinematic and associated IMU data as well as investigated the influence of adding synthetic data into training-measured data on deep learning model performance. The third study investigated the kinematic variation between two patient’s population across various activities: stair ascent, stair descent, and gait using principle component analysis PCA. Additionally, PCA-based autoencoders were developed to generate synthetic kinematics data for each patient population and activity. The fourth study investigated the potential use of a universal deep learning model for the estimation of lower extremities’ kinematics across various activities. Therefore, this model can be used as a global model for transfer learning methods in future research. This line of study resulted in a machine-learning framework that can be used to estimate biomechanical movements based on a stream of signals emitted from low-cost and portable IMUs. Eventually, this could lead to a simple clinical tool for tracking patients' movements in their own homes and translating those movements into diagnostic metrics that clinicians will be able to use to tailor treatment to each patient's needs in the future.
","abstract_html":"<p>Osteoarthritis (OA) is the leading cause of disability among the aging population in the United States and is frequently treated by replacing deteriorated joints with metal and plastic components. Developing better quantitative measures of movement quality to track patients longitudinally in their own homes would enable personalized treatment plans and hasten the advancement of promising new interventions. Wearable sensors and machine learning used to quantify patient movement could revolutionize the diagnosis and treatment of movement disorders. The purpose of this dissertation was to overcome technical challenges associated with the use of wearable sensors, specifically Inertial Measurement Units (IMUs), as a diagnostic tool for osteoarthritic (OA) and total knee replacement patients (TKR) through a detailed biomechanical assessment and development of machine learning algorithms. Specifically, the first study developed a relevant dataset consisting of IMU and associated biomechanical parameters of OA and TKR patients performing various activities, created a machine learning-based framework to accurately estimate spatiotemporal movement characteristics from IMU during level ground walking, and defined optimum sensor configuration associated with the patient population and activity. The second study designed a framework to generate synthetic kinematic and associated IMU data as well as investigated the influence of adding synthetic data into training-measured data on deep learning model performance. The third study investigated the kinematic variation between two patient’s population across various activities: stair ascent, stair descent, and gait using principle component analysis PCA. Additionally, PCA-based autoencoders were developed to generate synthetic kinematics data for each patient population and activity. The fourth study investigated the potential use of a universal deep learning model for the estimation of lower extremities’ kinematics across various activities. Therefore, this model can be used as a global model for transfer learning methods in future research. This line of study resulted in a machine-learning framework that can be used to estimate biomechanical movements based on a stream of signals emitted from low-cost and portable IMUs. Eventually, this could lead to a simple clinical tool for tracking patients' movements in their own homes and translating those movements into diagnostic metrics that clinicians will be able to use to tailor treatment to each patient's needs in the future.</p>","abstract_has_math":false,"creators":["Sharifi Renani, Mohsen"],"institution":null,"degree_name":"Ph.D.","degree_level":"Dissertation","degree_discipline":null,"degree_department":null,"school":null,"contributors":["Chadd W. Clary"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023-06-01T07:00:00Z","date_published":"2023-06-01T07:00:00Z","updated_at":"2026-07-24T02:02:03Z","subjects":["Artificial intelligence","Biomechanics","Deep learning","Osteoarthritic (OA)","Total knee replacement (TKR)","Wearables","Artificial Intelligence and Robotics","Biomechanical Engineering","Biomechanics and Biotransport","Biomedical Devices and Instrumentation","Biomedical Engineering and Bioengineering","Computer Sciences","Engineering"],"languages":["en"],"rights":["Copyright is held by the author. User is responsible for all copyright compliance.
"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.du.edu/etd/2215","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Chadd W. Clary"]},{"key":"dc:creator","label":"Author","values":["Sharifi Renani, Mohsen"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Ph.D."]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial intelligence","Biomechanics","Deep learning","Osteoarthritic (OA)","Total knee replacement (TKR)","Wearables","Artificial Intelligence and Robotics","Biomechanical Engineering","Biomechanics and Biotransport","Biomedical Devices and Instrumentation","Biomedical Engineering and Bioengineering","Computer Sciences","Engineering"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright is held by the author. User is responsible for all copyright compliance.
"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.du.edu/etd/2215"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Osteoarthritis (OA) is the leading cause of disability among the aging population in the United States and is frequently treated by replacing deteriorated joints with metal and plastic components. Developing better quantitative measures of movement quality to track patients longitudinally in their own homes would enable personalized treatment plans and hasten the advancement of promising new interventions. Wearable sensors and machine learning used to quantify patient movement could revolutionize the diagnosis and treatment of movement disorders. The purpose of this dissertation was to overcome technical challenges associated with the use of wearable sensors, specifically Inertial Measurement Units (IMUs), as a diagnostic tool for osteoarthritic (OA) and total knee replacement patients (TKR) through a detailed biomechanical assessment and development of machine learning algorithms. Specifically, the first study developed a relevant dataset consisting of IMU and associated biomechanical parameters of OA and TKR patients performing various activities, created a machine learning-based framework to accurately estimate spatiotemporal movement characteristics from IMU during level ground walking, and defined optimum sensor configuration associated with the patient population and activity. The second study designed a framework to generate synthetic kinematic and associated IMU data as well as investigated the influence of adding synthetic data into training-measured data on deep learning model performance. The third study investigated the kinematic variation between two patient’s population across various activities: stair ascent, stair descent, and gait using principle component analysis PCA. Additionally, PCA-based autoencoders were developed to generate synthetic kinematics data for each patient population and activity. The fourth study investigated the potential use of a universal deep learning model for the estimation of lower extremities’ kinematics across various activities. Therefore, this model can be used as a global model for transfer learning methods in future research. This line of study resulted in a machine-learning framework that can be used to estimate biomechanical movements based on a stream of signals emitted from low-cost and portable IMUs. Eventually, this could lead to a simple clinical tool for tracking patients' movements in their own homes and translating those movements into diagnostic metrics that clinicians will be able to use to tailor treatment to each patient's needs in the future.
"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Patient Movement Monitoring Based on IMU and Deep Learning"]}]}],"canonical_facts":{"dc:contributor":["Chadd W. Clary"],"dc:creator":["Sharifi Renani, Mohsen"],"dc:description.abstract":["Osteoarthritis (OA) is the leading cause of disability among the aging population in the United States and is frequently treated by replacing deteriorated joints with metal and plastic components. Developing better quantitative measures of movement quality to track patients longitudinally in their own homes would enable personalized treatment plans and hasten the advancement of promising new interventions. Wearable sensors and machine learning used to quantify patient movement could revolutionize the diagnosis and treatment of movement disorders. The purpose of this dissertation was to overcome technical challenges associated with the use of wearable sensors, specifically Inertial Measurement Units (IMUs), as a diagnostic tool for osteoarthritic (OA) and total knee replacement patients (TKR) through a detailed biomechanical assessment and development of machine learning algorithms. Specifically, the first study developed a relevant dataset consisting of IMU and associated biomechanical parameters of OA and TKR patients performing various activities, created a machine learning-based framework to accurately estimate spatiotemporal movement characteristics from IMU during level ground walking, and defined optimum sensor configuration associated with the patient population and activity. The second study designed a framework to generate synthetic kinematic and associated IMU data as well as investigated the influence of adding synthetic data into training-measured data on deep learning model performance. The third study investigated the kinematic variation between two patient’s population across various activities: stair ascent, stair descent, and gait using principle component analysis PCA. Additionally, PCA-based autoencoders were developed to generate synthetic kinematics data for each patient population and activity. The fourth study investigated the potential use of a universal deep learning model for the estimation of lower extremities’ kinematics across various activities. Therefore, this model can be used as a global model for transfer learning methods in future research. This line of study resulted in a machine-learning framework that can be used to estimate biomechanical movements based on a stream of signals emitted from low-cost and portable IMUs. Eventually, this could lead to a simple clinical tool for tracking patients' movements in their own homes and translating those movements into diagnostic metrics that clinicians will be able to use to tailor treatment to each patient's needs in the future.
"],"dc:format":["application/pdf"],"dc:identifier":["https://digitalcommons.du.edu/etd/2215"],"dc:language":["en"],"dc:rights":["Copyright is held by the author. User is responsible for all copyright compliance.
"],"dc:subject":["Artificial intelligence","Biomechanics","Deep learning","Osteoarthritic (OA)","Total knee replacement (TKR)","Wearables","Artificial Intelligence and Robotics","Biomechanical Engineering","Biomechanics and Biotransport","Biomedical Devices and Instrumentation","Biomedical Engineering and Bioengineering","Computer Sciences","Engineering"],"dc:title":["Patient Movement Monitoring Based on IMU and Deep Learning"],"thesis:degree_level":["Dissertation"],"thesis:degree_name":["Ph.D."]},"updated_at":"2026-07-24T02:02:03Z"}