{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etdr-2714"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etdr-2714","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"Knowledge discovery on the integrative analysis of electrical and mechanical dyssynchrony to improve cardiac resynchronization therapy","abstract":"Cardiac resynchronization therapy (CRT) is a standard method of treating heart failure by coordinating the function of the left and right ventricles. However, up to 40% of CRT recipients do not experience clinical symptoms or cardiac function improvements. The main reasons for CRT non-response include: (1) suboptimal patient selection based on electrical dyssynchrony measured by electrocardiogram (ECG) in current guidelines; (2) mechanical dyssynchrony has been shown to be effective but has not been fully explored; and (3) inappropriate placement of the CRT left ventricular (LV) lead in a significant number of patients. In terms of mechanical dyssynchrony, we utilize an autoencoder to extract new predictive features from nuclear medicine images, characterizing local mechanical dyssynchrony and improving the CRT response rate. Although machine learning can identify complex patterns and make accurate predictions from large datasets, the low interpretability of these \"black box\" methods makes it difficult to integrate them with clinical decisions made by physicians in the healthcare setting. Therefore, we use visualization techniques to enable physicians to understand the physical meaning of new features and the reasoning behind the clinical decisions made by the artificial intelligent model. For electrical dyssynchrony, we use short-time Fourier transform (STFT) to transform one-dimensional waveforms into two-dimensional frequency-time spectra. And transfer learning is used to leverage the knowledge learned from a large arrhythmia ECG dataset of related medical conditions to improve patient selection for CRT with limited data. This improves prediction accuracy, reduces the time and resources required, and potentially leads to better patient outcomes. Furthermore, an innovative approach is proposed for using three-dimensional spatial VCG information to describe the characteristics of electrical dyssynchrony, locate the latest activation site, and combine it with the latest mechanical contraction site to select the optimal LV lead position. In addition, we apply deep reinforcement learning to the decision-making problem of CRT patients. We investigate discrete state space/specific action space models to find the best treatment strategy, improve the reward equation based on the physician's experience, and learn the approximation of the best action-value function that can improve the treatment policy used by clinicians and provide interpretability.","abstract_html":"Cardiac resynchronization therapy (CRT) is a standard method of treating heart failure by coordinating the function of the left and right ventricles. However, up to 40% of CRT recipients do not experience clinical symptoms or cardiac function improvements. The main reasons for CRT non-response include: (1) suboptimal patient selection based on electrical dyssynchrony measured by electrocardiogram (ECG) in current guidelines; (2) mechanical dyssynchrony has been shown to be effective but has not been fully explored; and (3) inappropriate placement of the CRT left ventricular (LV) lead in a significant number of patients. In terms of mechanical dyssynchrony, we utilize an autoencoder to extract new predictive features from nuclear medicine images, characterizing local mechanical dyssynchrony and improving the CRT response rate. Although machine learning can identify complex patterns and make accurate predictions from large datasets, the low interpretability of these &quot;black box&quot; methods makes it difficult to integrate them with clinical decisions made by physicians in the healthcare setting. Therefore, we use visualization techniques to enable physicians to understand the physical meaning of new features and the reasoning behind the clinical decisions made by the artificial intelligent model. For electrical dyssynchrony, we use short-time Fourier transform (STFT) to transform one-dimensional waveforms into two-dimensional frequency-time spectra. And transfer learning is used to leverage the knowledge learned from a large arrhythmia ECG dataset of related medical conditions to improve patient selection for CRT with limited data. This improves prediction accuracy, reduces the time and resources required, and potentially leads to better patient outcomes. Furthermore, an innovative approach is proposed for using three-dimensional spatial VCG information to describe the characteristics of electrical dyssynchrony, locate the latest activation site, and combine it with the latest mechanical contraction site to select the optimal LV lead position. In addition, we apply deep reinforcement learning to the decision-making problem of CRT patients. We investigate discrete state space/specific action space models to find the best treatment strategy, improve the reward equation based on the physician&#x27;s experience, and learn the approximation of the best action-value function that can improve the treatment policy used by clinicians and provide interpretability.","abstract_has_math":false,"creators":["He, Zhuo"],"institution":null,"degree_name":"Doctor of Philosophy in Computational Science and Engineering (PhD)","degree_level":"Open Access Dissertation","degree_discipline":"Department of Applied Computing","degree_department":null,"school":null,"contributors":["Weihua Zhou"],"advisors":[],"committee_chairs":[],"committee_members":["Guy C. Hembroff","Jinshan Tang","Qinghui Chen"],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:18:53Z","subjects":["Applied Statistics","Artificial Intelligence and Robotics","Biomedical Informatics","Biostatistics","Categorical Data Analysis","Databases and Information Systems","Data Science"],"languages":[],"rights":[],"rights_urls":["http://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etdr/1639"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etdr/1639","href":"https://digitalcommons.mtu.edu/etdr/1639","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etdr/1639","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etdr/1639","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.mtu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.mtu.edu%3Aetdr-2714","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Weihua Zhou"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Guy C. Hembroff","Jinshan Tang","Qinghui Chen"]},{"key":"dc:creator","label":"Author","values":["He, Zhuo"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.created","label":"Dc Date Created","values":["2023-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Applied Computing"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Open Access Dissertation"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy in Computational Science and Engineering (PhD)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Applied Statistics","Artificial Intelligence and Robotics","Biomedical Informatics","Biostatistics","Categorical Data Analysis","Databases and Information Systems","Data Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:rights.license","label":"Dc Rights License","values":["http://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etdr/1639","https://digitalcommons.mtu.edu/etdr/1639"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Cardiac resynchronization therapy (CRT) is a standard method of treating heart failure by coordinating the function of the left and right ventricles. However, up to 40% of CRT recipients do not experience clinical symptoms or cardiac function improvements. The main reasons for CRT non-response include: (1) suboptimal patient selection based on electrical dyssynchrony measured by electrocardiogram (ECG) in current guidelines; (2) mechanical dyssynchrony has been shown to be effective but has not been fully explored; and (3) inappropriate placement of the CRT left ventricular (LV) lead in a significant number of patients. In terms of mechanical dyssynchrony, we utilize an autoencoder to extract new predictive features from nuclear medicine images, characterizing local mechanical dyssynchrony and improving the CRT response rate. Although machine learning can identify complex patterns and make accurate predictions from large datasets, the low interpretability of these \"black box\" methods makes it difficult to integrate them with clinical decisions made by physicians in the healthcare setting. Therefore, we use visualization techniques to enable physicians to understand the physical meaning of new features and the reasoning behind the clinical decisions made by the artificial intelligent model. For electrical dyssynchrony, we use short-time Fourier transform (STFT) to transform one-dimensional waveforms into two-dimensional frequency-time spectra. And transfer learning is used to leverage the knowledge learned from a large arrhythmia ECG dataset of related medical conditions to improve patient selection for CRT with limited data. This improves prediction accuracy, reduces the time and resources required, and potentially leads to better patient outcomes. Furthermore, an innovative approach is proposed for using three-dimensional spatial VCG information to describe the characteristics of electrical dyssynchrony, locate the latest activation site, and combine it with the latest mechanical contraction site to select the optimal LV lead position. In addition, we apply deep reinforcement learning to the decision-making problem of CRT patients. We investigate discrete state space/specific action space models to find the best treatment strategy, improve the reward equation based on the physician's experience, and learn the approximation of the best action-value function that can improve the treatment policy used by clinicians and provide interpretability."]},{"key":"dc:title","label":"Title","values":["Knowledge discovery on the integrative analysis of electrical and mechanical dyssynchrony to improve cardiac resynchronization therapy"]}]}],"canonical_facts":{"dc:contributor":["Weihua Zhou"],"dc:contributor.committeemember":["Guy C. Hembroff","Jinshan Tang","Qinghui Chen"],"dc:creator":["He, Zhuo"],"dc:date.created":["2023-01-01T08:00:00Z"],"dc:description.abstract":["Cardiac resynchronization therapy (CRT) is a standard method of treating heart failure by coordinating the function of the left and right ventricles. However, up to 40% of CRT recipients do not experience clinical symptoms or cardiac function improvements. The main reasons for CRT non-response include: (1) suboptimal patient selection based on electrical dyssynchrony measured by electrocardiogram (ECG) in current guidelines; (2) mechanical dyssynchrony has been shown to be effective but has not been fully explored; and (3) inappropriate placement of the CRT left ventricular (LV) lead in a significant number of patients. In terms of mechanical dyssynchrony, we utilize an autoencoder to extract new predictive features from nuclear medicine images, characterizing local mechanical dyssynchrony and improving the CRT response rate. Although machine learning can identify complex patterns and make accurate predictions from large datasets, the low interpretability of these \"black box\" methods makes it difficult to integrate them with clinical decisions made by physicians in the healthcare setting. Therefore, we use visualization techniques to enable physicians to understand the physical meaning of new features and the reasoning behind the clinical decisions made by the artificial intelligent model. For electrical dyssynchrony, we use short-time Fourier transform (STFT) to transform one-dimensional waveforms into two-dimensional frequency-time spectra. And transfer learning is used to leverage the knowledge learned from a large arrhythmia ECG dataset of related medical conditions to improve patient selection for CRT with limited data. This improves prediction accuracy, reduces the time and resources required, and potentially leads to better patient outcomes. Furthermore, an innovative approach is proposed for using three-dimensional spatial VCG information to describe the characteristics of electrical dyssynchrony, locate the latest activation site, and combine it with the latest mechanical contraction site to select the optimal LV lead position. In addition, we apply deep reinforcement learning to the decision-making problem of CRT patients. We investigate discrete state space/specific action space models to find the best treatment strategy, improve the reward equation based on the physician's experience, and learn the approximation of the best action-value function that can improve the treatment policy used by clinicians and provide interpretability."],"dc:identifier":["10.37099/mtu.dc.etdr/1639","https://digitalcommons.mtu.edu/etdr/1639"],"dc:rights.license":["http://creativecommons.org/licenses/by/4.0/"],"dc:subject":["Applied Statistics","Artificial Intelligence and Robotics","Biomedical Informatics","Biostatistics","Categorical Data Analysis","Databases and Information Systems","Data Science"],"dc:title":["Knowledge discovery on the integrative analysis of electrical and mechanical dyssynchrony to improve cardiac resynchronization therapy"],"thesis:degree_discipline":["Department of Applied Computing"],"thesis:degree_level":["Open Access Dissertation"],"thesis:degree_name":["Doctor of Philosophy in Computational Science and Engineering (PhD)"]},"updated_at":"2026-11-19T08:18:53Z"}