{"id":{"repo_id":"uab","oai_identifier":"oai:digitalcommons.library.uab.edu:etd-collection-3602"},"canonical_url":"https://search.ndltd.org/etd/uab/oai:digitalcommons.library.uab.edu:etd-collection-3602","repository":{"repo_id":"uab","name":"University of Alabama Birmingham","base_url":"https://digitalcommons.library.uab.edu/do/oai/"},"display":{"title":"A Deep Learning Approach For Identifying Key Biomarkers In Medical Imaging Applications","abstract":"Although Deep Leaning has achieved great success in many domains in recent years, research into its applicability and eﬀectiveness in medical applications has been limited for various reasons. Some of these barriers are related to the historic impression that neural networks are black boxes, especially when applied to medical diagnosis and oﬀer no in terpretation into prediction. In this dissertation, I demonstrated that deep learning can be an eﬀective utility in medical diagnosis and biomarker detection. In particular, I have developed deep learning models that can identify biomarkers in brain hemorrhage, brain tumor, pneumonia, diabetic retinopathy and Parkinsons disease, with high sensitivity and speciﬁcity. I demonstrate models and methodologies in this thesis to address a number of current challenges in utilizing neural nets for medical applications, including (1) demonstra tion of methods and techniques for eﬃciently managing 3D and higher dimensional data, (2) demonstration of the utility of class-activation mapping for gleaning intuition into the bases of decision making within a given trained neural net, and (3) demonstration the utility of combining class activation mapping with augmented weak supervision techniques to improve the eﬃciency of training of neural nets and allow for more rapid development of clinically relevant algorithms. This thesis is organized through an exploration of appropriate method ologies and techniques applied to four independent, and disparate types of problems in the medical domain, involving two 2D datasets, one 3D dataset, and one higher dimensional (Diﬀusion-Weighted Imaging Magnetic Resonance Imaging) dataset. I close my discussion by proposing a general framework for developing and deploying deep learning systems into existing hospital infrastructure with considerations for privacy and security","abstract_html":"Although Deep Leaning has achieved great success in many domains in recent years, research into its applicability and eﬀectiveness in medical applications has been limited for various reasons. Some of these barriers are related to the historic impression that neural networks are black boxes, especially when applied to medical diagnosis and oﬀer no in terpretation into prediction. In this dissertation, I demonstrated that deep learning can be an eﬀective utility in medical diagnosis and biomarker detection. In particular, I have developed deep learning models that can identify biomarkers in brain hemorrhage, brain tumor, pneumonia, diabetic retinopathy and Parkinsons disease, with high sensitivity and speciﬁcity. I demonstrate models and methodologies in this thesis to address a number of current challenges in utilizing neural nets for medical applications, including (1) demonstra tion of methods and techniques for eﬃciently managing 3D and higher dimensional data, (2) demonstration of the utility of class-activation mapping for gleaning intuition into the bases of decision making within a given trained neural net, and (3) demonstration the utility of combining class activation mapping with augmented weak supervision techniques to improve the eﬃciency of training of neural nets and allow for more rapid development of clinically relevant algorithms. This thesis is organized through an exploration of appropriate method ologies and techniques applied to four independent, and disparate types of problems in the medical domain, involving two 2D datasets, one 3D dataset, and one higher dimensional (Diﬀusion-Weighted Imaging Magnetic Resonance Imaging) dataset. I close my discussion by proposing a general framework for developing and deploying deep learning systems into existing hospital infrastructure with considerations for privacy and security","abstract_has_math":false,"creators":["Odaibo, David"],"institution":null,"degree_name":null,"degree_level":"Dissertation","degree_discipline":null,"degree_department":null,"school":null,"contributors":["Murat Tanik","Leon Jololian","Nakhmani Arie","Earl Wells","Karthikeyan Lingasubramanian"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T09:27:05Z","subjects":["CT","Deep Learning","Medical Imaging","MRI","Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.library.uab.edu/etd-collection/2610","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://digitalcommons.library.uab.edu/etd-collection/2610","record_label":"Repository record","record_kind":"record","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.library.uab.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.library.uab.edu%3Aetd-collection-3602","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Murat Tanik","Leon Jololian","Nakhmani Arie","Earl Wells","Karthikeyan Lingasubramanian"]},{"key":"dc:creator","label":"Author","values":["Odaibo, David"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.created","label":"Dc Date Created","values":["2019-01-01T08:00:00Z"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Dissertation"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["CT","Deep Learning","Medical Imaging","MRI","Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.library.uab.edu/etd-collection/2610"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Although Deep Leaning has achieved great success in many domains in recent years, research into its applicability and eﬀectiveness in medical applications has been limited for various reasons. Some of these barriers are related to the historic impression that neural networks are black boxes, especially when applied to medical diagnosis and oﬀer no in terpretation into prediction. In this dissertation, I demonstrated that deep learning can be an eﬀective utility in medical diagnosis and biomarker detection. In particular, I have developed deep learning models that can identify biomarkers in brain hemorrhage, brain tumor, pneumonia, diabetic retinopathy and Parkinsons disease, with high sensitivity and speciﬁcity. I demonstrate models and methodologies in this thesis to address a number of current challenges in utilizing neural nets for medical applications, including (1) demonstra tion of methods and techniques for eﬃciently managing 3D and higher dimensional data, (2) demonstration of the utility of class-activation mapping for gleaning intuition into the bases of decision making within a given trained neural net, and (3) demonstration the utility of combining class activation mapping with augmented weak supervision techniques to improve the eﬃciency of training of neural nets and allow for more rapid development of clinically relevant algorithms. This thesis is organized through an exploration of appropriate method ologies and techniques applied to four independent, and disparate types of problems in the medical domain, involving two 2D datasets, one 3D dataset, and one higher dimensional (Diﬀusion-Weighted Imaging Magnetic Resonance Imaging) dataset. I close my discussion by proposing a general framework for developing and deploying deep learning systems into existing hospital infrastructure with considerations for privacy and security"]},{"key":"dc:description.note","label":"Dc Description Note","values":["etdadmin_upload_702821"]},{"key":"dc:title","label":"Title","values":["A Deep Learning Approach For Identifying Key Biomarkers In Medical Imaging Applications"]}]}],"canonical_facts":{"dc:contributor":["Murat Tanik","Leon Jololian","Nakhmani Arie","Earl Wells","Karthikeyan Lingasubramanian"],"dc:creator":["Odaibo, David"],"dc:date.created":["2019-01-01T08:00:00Z"],"dc:description.abstract":["Although Deep Leaning has achieved great success in many domains in recent years, research into its applicability and eﬀectiveness in medical applications has been limited for various reasons. Some of these barriers are related to the historic impression that neural networks are black boxes, especially when applied to medical diagnosis and oﬀer no in terpretation into prediction. In this dissertation, I demonstrated that deep learning can be an eﬀective utility in medical diagnosis and biomarker detection. In particular, I have developed deep learning models that can identify biomarkers in brain hemorrhage, brain tumor, pneumonia, diabetic retinopathy and Parkinsons disease, with high sensitivity and speciﬁcity. I demonstrate models and methodologies in this thesis to address a number of current challenges in utilizing neural nets for medical applications, including (1) demonstra tion of methods and techniques for eﬃciently managing 3D and higher dimensional data, (2) demonstration of the utility of class-activation mapping for gleaning intuition into the bases of decision making within a given trained neural net, and (3) demonstration the utility of combining class activation mapping with augmented weak supervision techniques to improve the eﬃciency of training of neural nets and allow for more rapid development of clinically relevant algorithms. This thesis is organized through an exploration of appropriate method ologies and techniques applied to four independent, and disparate types of problems in the medical domain, involving two 2D datasets, one 3D dataset, and one higher dimensional (Diﬀusion-Weighted Imaging Magnetic Resonance Imaging) dataset. I close my discussion by proposing a general framework for developing and deploying deep learning systems into existing hospital infrastructure with considerations for privacy and security"],"dc:description.note":["etdadmin_upload_702821"],"dc:identifier":["https://digitalcommons.library.uab.edu/etd-collection/2610"],"dc:subject":["CT","Deep Learning","Medical Imaging","MRI","Engineering"],"dc:title":["A Deep Learning Approach For Identifying Key Biomarkers In Medical Imaging Applications"],"thesis:degree_level":["Dissertation"]},"updated_at":"2026-11-19T09:27:05Z"}