[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121494-en":3,"doc-seo-121494-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"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},121494,1374391974585,"Genevieve","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",7,"Healthcare","Predicting Kidney Post-Transplantation Function from Optical Coherence Tomography Images Using Machine Learning Approaches - Thesis","This thesis develops machine learning approaches to predict kidney post-transplantation function using optical coherence tomography (OCT) image data. The work examines OCT-based texture feature extraction and integrates these features with KDPI scores to improve predictive performance. Methods address imbalanced learning through data balancing (SMOTE-Tomek) and cost-sensitive learning. Decision thresholds are optimized to enhance classification outcomes, and models are evaluated with metrics tailored for imbalanced datasets, including confusion-matrix–based assessment.","UNIVERSITY OF OKLAHOMA  \nGRADUATE COLLEGE  \nPREDICTING KIDNEY POST-TRANSPLANTATION FUNCTION FROM OPTICAL COHERENCE TOMOGRAPHY IMAGES USING MACHINE  \nLEARNING APPROACHES  \nA THESIS  \nSUBMITTED TO THE GRADUATE FACULTY in partial fulfillment of the requirements for the Degree of  \nMASTER OF SCIENCE  \nBy  \nBORNFACE MUTEMBEI  \nNorman, Oklahoma  \n2025  \nPREDICTING KIDNEY POST-TRANSPLANTATION FUNCTION FROM OPTICAL COHERENCE TOMOGRAPHY IMAGES USING MACHINE  \nLEARNING APPROACHES  \nA THESIS APPROVED FOR THE  \nSTEPHENSON SCHOOL OF BIOMEDICAL ENGINEERING  \nBY THE COMMITTEE CONSISTING OF  \nDr. Qinggong Tang (Chair)  \nDr. David Miller  \nDr. Handan Acar  \n© Copyright by BORNFACE MUTEMBEI 2025 All Rights Reserved.  \nAcknowledgments  \nFirst, I would like to express my sincere gratitude to my advisor, Dr. Qinggong Tang, for their invaluable support throughout my time in the Biophotonics Laboratory and for their mentorship during this project. Dr. Tang cultivated an environment that encouraged both independent work and collaboration.  \nI am also thankful to my committee members, Dr. Handan Acar and Dr. David Miller for their support during the revision of this work. In addition to serving on my committee, Dr. Acar has provided thoughtful guidance as a faculty mentor in matters of career development and professional growth.  \nA special thanks is extended to Dr. Theodore Trafalis from the School of Industrial Systems Engineering at the University of Oklahoma for insightful advice and mentorship throughout the course of this work.  \nLastly, I am grateful to my grandparents Samson and Harriet Muthamia whose unwavering support and investment in my education has made this journey possible.  \nTable of Contents  \nAcknowledgments iv  \nList Of Tables vii  \nList Of Figures viii  \nAbstract x  \n1 Introduction 1  \n1.1 Donor Categories and Expanded Criteria Donors ............. 1  \n1.2 Post-transplant categories, IGF versus DGF ............... 1  \n1.3 Utilizing OCT to assess renal graft function ............... 3  \n1.4 Texture Feature Extraction ........................ 4  \n1.5 The imbalanced data problem ....................... 5  \n1.5.1 Data-level solutions to imbalanced data .............. 6  \n1.5.2 Algorithm-level Solutions to imbalanced data .......... 7  \n2 Methods 8  \n2.1 Patient Demographic and Transplant Groups ............... 8  \n2.2 Data Processing ............................... 9  \n2.2.1 OCT Image Acquisition ...................... 9  \n2.2.2 OCT Image Segmentation ..................... 10  \n2.2.3 Texture Extraction ......................... 12  \n2.2.4 Feature Selection and dimension reduction ............ 13  \n2.3 Machine Learning Selection and Training ................. 16  \n2.4 Evaluation Metrics for Imbalanced Learning ............... 17  \n3 Results 20  \n3.1 Optimizing the decision threshold ..................... 20  \n3.2 Baseline: OCT-texture versus KDPI scores ................ 23  \n3.3 Combined OCT texture data and KDPI scores .............. 25  \n3.4 Balancing the data with SMOTE-Tomek ................. 27  \n3.5 Cost-sensitive learning ........................... 28  \n4 Conclusion and Future Work 30  \nReferences 34  \nAppendix 41  \n1 Appendix A   41  \nList Of Tables  \n2.1 Comparison of deep learning models for denoising OCT images ..... 12  \n2.2 Radiomics texture feature extraction from various image types ..... 13  \n2.3 2x2 Confusion Matrix for Binary Classification .............. 18  \n3.1 Evaluation of classifiers trained on KDPI scores alone, t = 0.5 ..... 21  \n3.2 Evaluation of classifiers trained on KDPI scores, t = toptimal ....... 21  \n3.3 Evaluation of classifiers trained on OCT texture data only, t = 0 .5 . . . 22  \n3.4 Evaluation of classifiers trained on OCT texture data only, t = toptimal . 23  \n3.5 Evaluation of classifiers combined OCT texture and KDPI data .... 26  \n3.6 Evaluation of classifiers after balancing data using SMOTE-Tomek ... 27  \n3.7 Evaluation of classifiers after implementing Cost-Sensitive Learning .. 29  \nA.1 Hy","cbCaisXOh2a1eDks","https://ap.wps.com/l/cbCaisXOh2a1eDks","pdf",4089814,1,52,"English","en",105,"# Introduction\n## Donor categories and expanded criteria\n## Post-transplant categories (IGF versus DGF)\n## Utilizing OCT to assess renal graft function\n## Texture feature extraction\n## The imbalanced data problem\n# Methods\n## Patient demographic and transplant groups\n## Data processing\n## Machine learning selection and training\n## Evaluation metrics for imbalanced learning\n# Results\n## Optimizing the decision threshold\n## Baseline: OCT-texture versus KDPI scores\n## Combined OCT texture data and KDPI scores\n## Balancing the data with SMOTE-Tomek\n## Cost-sensitive learning\n# Conclusion and Future Work","[{\"question\":\"How does the thesis use OCT images for predicting kidney post-transplantation function?\",\"answer\":\"It extracts texture features from OCT images after image acquisition and segmentation, then uses these features as inputs to machine learning models for classification of post-transplant outcomes.\"},{\"question\":\"What role do KDPI scores play in the prediction framework?\",\"answer\":\"The thesis compares OCT-texture-only baselines with KDPI-score-only baselines and evaluates improvements when OCT texture data are combined with KDPI scores.\"},{\"question\":\"How is the imbalanced data problem handled in the study?\",\"answer\":\"The work evaluates multiple strategies, including data-level balancing with SMOTE-Tomek and algorithm-level cost-sensitive learning, and uses evaluation metrics suited to imbalanced learning.\"}]","Predicting Kidney Post-Transplantation Function from Optical Coherence Tomography Images Using Machine Learning Approaches - Thesis | PDF",1785735921,131,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"predicting-kidney-post-transplantation-function-from-optical-coherence-tomography-images-using-machine-learning-approaches-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@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/predicting-kidney-post-transplantation-function-from-optical-coherence-tomography-images-using-machine-learning-approaches-thesis/121494/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the thesis use OCT images for predicting kidney post-transplantation function?","Question",{"text":75,"@type":76},"It extracts texture features from OCT images after image acquisition and segmentation, then uses these features as inputs to machine learning models for classification of post-transplant outcomes.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What role do KDPI scores play in the prediction framework?",{"text":80,"@type":76},"The thesis compares OCT-texture-only baselines with KDPI-score-only baselines and evaluates improvements when OCT texture data are combined with KDPI scores.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the imbalanced data problem handled in the study?",{"text":84,"@type":76},"The work evaluates multiple strategies, including data-level balancing with SMOTE-Tomek and algorithm-level cost-sensitive learning, and uses evaluation metrics suited to imbalanced learning.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,118,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":116,"slug":117},40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]