[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-118670-en":3,"doc-seo-118670-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},118670,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",7,"Healthcare","Machine Learning Applications in Cardiology - PhD Thesis","Cardiovascular diseases remain the leading cause of death globally, creating strong demand for advanced prevention and management strategies. This PhD thesis evaluates the transformative potential of artificial intelligence in cardiology by applying modern machine learning and deep learning methods to clinically relevant tasks. It covers ECG-based atrial fibrillation prediction, multi-modal survival modeling, unsupervised clustering for phenotype identification in dilated cardiomyopathy, and a survival-clustering framework for treatment response assessment. It also develops an automated deep learning tool for coronary computed tomography angiography and diagnostic models for transthyretin amyloid cardiomyopathy, emphasizing rigorous study design and performance evaluation aligned with clinical objectives.","university of trieste  \ndepartment of mathematics, informatics and geosciences  \nphd in applied data science and artificial intelligence  \nmachine learning applications in  \ncardiology  \ngiovanni baj Matr. PHD1400007  \nthesis supervisors:  \nProf. Giulia Barbati Prof. Luca Bortolussi Arjuna Scagnetto  \nacademic year 2023-2024  \nCardiovascular diseases remain the leading cause of death globally and impose significant economic burdens, emphasizing the need for advanced prevention and management strategies. This thesis investigates the transformative potential of artificial intelligence in cardiology, employing state-of-the-art machine learning and deep learning methodologies to address key challenges in cardiovascular care.  \nUsing electrocardiogram data, the first part of this work evaluates machine learning models for atrial fibrillation prediction, demonstrating the impact of class imbalance corrections on calibration and the importance of adequate sample sizes for deep learning, explored through learning curve techniques. Continuing within the context of atrial fibrillation prediction, multi-modal approaches integrating electrocardiogram and tabular data in survival frameworks are compared to single-modality models. Unsupervised clustering is applied to phenotype dilated cardiomyopathy patients using electrocardiogram, demographic, and clinical data, identifying clusters with distinct genetic backgrounds and outcomes. Additionally, a novel method that combines survival neural networks with functional clustering is developed to assess treatment response, capturing time-dependent effects and feature interactions for personalized care. In the extended framework of integrating also images, a fully automated deep learning tool for coronary computed tomography angiography was developed, demonstrating high accuracy in detecting rare congenital heart diseases. Finally, machine learning-based diagnostic models effectively identified transthyretin amyloid cardiomyopathy in patients with severe aortic stenosis, with computed tomography strain emerging asthe most accurate modality.  \nThe findings of this thesis highlight the potential of artificial intelligence tools to enhance cardiovascular diagnostics and patient management, emphasizing the critical role of rigorous study design and the careful evaluation of performance measures aligned with clinical objectives  \npublished  \n• Giovanni Baj et al.“Comparison of discrimination and calibration performance of ECG-based machine learning models for prediction of new-onset atrial fibrillation.” In: BMC Medical Research Methodology 23.1 (July 2023), p. 169. doi: 10. 1186/s12874023- 01989-3  \n• Giovanni Baj et al. “Deep Learning Survival Model to Predict Atrial Fibrillation From ECGs and EHR Data.” In: Progress in Artificial Intelligence. Springer Nature Switzerland, 2023, pp. 222–233. doi: 10.1007/978-3-031-49011-8_ 18  \n• Isaac Shiri, Sebastian Balzer, Giovanni Baj et al.“Multi-modality artificial intelligence-based transthyretin amyloid cardiomyopathy detection in patients with severe aortic stenosis.” In: European Journal of Nuclear Medicine and Molecular Imaging (Sept. 2024) . doi: 10.1007/s00259-024-06922-4  \nunder review  \n• Ilaria Gandin, Maria Perotto, Alessia Paldino, Giovanni Baj et al.“Clustering in Dilated Cardiomyopathy at Initial Evaluation: An Effective Tool for Clinical Stratification”. Submitted to European Journal of Heart Failure.  \n• Isaac Shiri, Giovanni Baj et al.“Artificial Intelligence Based Detection and Classification of Anomalous Aortic Origin of Coronary Arteries in Coronary CT Angiography: A Multi-Center Development, Testing and Clinical Evaluation Study”. Under review in Nature Communications.  \n• Daniela Pacella, Emanuele Giusti, Annamaria Porreca, Giovanni Baj, Ilaria Gandin, Giulia Barbati.“Sample size determination via learning curves for AI models: an application to deep learning algorithms for diagnostic and prediction tasks”. Submitted to Artificial Intell","cbCaiu0el3MENU3H","https://ap.wps.com/l/cbCaiu0el3MENU3H","pdf",7235635,1,185,"English","en",105,"# Introduction\n## Background\n## Research questions\n## Data modalities\n# Atrial fibrillation prediction\n## Effects of class imbalance corrections on ML models performance\n## Learning curve for DL\n## Survival multi-modal model for AF prediction\n# Clustering for patient phenotyping\n## Dilated Cardiomyopathy phenotyping","[{\"question\":\"How does the thesis address atrial fibrillation prediction from ECG data?\",\"answer\":\"It evaluates machine learning models using ECG data and analyzes how class imbalance corrections influence calibration. It also studies adequate sample sizes for deep learning through learning-curve techniques.\"},{\"question\":\"What multi-modal approaches are compared for atrial fibrillation risk modeling?\",\"answer\":\"It compares single-modality ECG approaches against multi-modal frameworks that integrate electrocardiogram and tabular data within survival modeling settings.\"},{\"question\":\"How are patient phenotypes and treatment response assessed in dilated cardiomyopathy?\",\"answer\":\"Unsupervised clustering is applied to phenotype patients using ECG, demographic, and clinical data, separating clusters with distinct genetic backgrounds and outcomes. A method combining survival neural networks with functional clustering then evaluates treatment response with time-dependent effects and feature interactions.\"}]","Machine Learning Applications in Cardiology - PhD Thesis | PDF",1785684823,466,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-applications-in-cardiology-phd-thesis","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/healthcare/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-applications-in-cardiology-phd-thesis/118670/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-02",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"How does the thesis address atrial fibrillation prediction from ECG data?","Question",{"text":76,"@type":77},"It evaluates machine learning models using ECG data and analyzes how class imbalance corrections influence calibration. It also studies adequate sample sizes for deep learning through learning-curve techniques.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What multi-modal approaches are compared for atrial fibrillation risk modeling?",{"text":81,"@type":77},"It compares single-modality ECG approaches against multi-modal frameworks that integrate electrocardiogram and tabular data within survival modeling settings.",{"name":83,"@type":74,"acceptedAnswer":84},"How are patient phenotypes and treatment response assessed in dilated cardiomyopathy?",{"text":85,"@type":77},"Unsupervised clustering is applied to phenotype patients using ECG, demographic, and clinical data, separating clusters with distinct genetic backgrounds and outcomes. A method combining survival neural networks with functional clustering then evaluates treatment response with time-dependent effects and feature interactions.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,119,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":117,"slug":118},40,"healthcare",{"id":120,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":122,"slug":123},8,"Research & Report",30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]