{"id":{"repo_id":"csusb","oai_identifier":"oai:scholarworks.lib.csusb.edu:etd-2988"},"canonical_url":"https://search.ndltd.org/etd/csusb/oai:scholarworks.lib.csusb.edu:etd-2988","repository":{"repo_id":"csusb","name":"CSUniversity San Bernardino","base_url":"https://scholarworks.lib.csusb.edu/do/oai/"},"display":{"title":"LUNG LESION SEGMENTATION USING DEEP LEARNING APPROACHES","abstract":"The amount of data generated in the medical imaging field, especially in a modern context, is growing significantly. As the amount of data grows, it's prudent to make use of automated techniques that can leverage datasets to solve problems that are error-prone or have inconsistent solutions. Deep learning approaches have gained traction in medical imaging tasks due to their superior performance with larger datasets and ability to discern the intricate features of 3D volumes, a task inefficient if done manually. Specifically for the task of lung nodule segmentation, several different methods have been tried before such as region growing etc. but this project focuses on using an Attention U-Net model to automatically segment the nodule boundaries. Specifically, this is done on the LUNA16 dataset as a benchmark which is a popular reference point for comparison. To achieve this, specifically, the Attention U-Net was trained with 5-fold cross-validation on the training dataset. In addition to the segmentation outputs, averaged training and validation curves over all folds were also shown as the model is trained for 70 epochs. To conclude, these results present a useful automated method to segment the lung nodules. In practical situations, this would be of significant help to radiologists as it is less error-prone and not as susceptible to inter-observer variability. These automated tools along with other radiologist interactions could potentially significantly improve patient outcomes.","abstract_html":"The amount of data generated in the medical imaging field, especially in a modern context, is growing significantly. As the amount of data grows, it&#x27;s prudent to make use of automated techniques that can leverage datasets to solve problems that are error-prone or have inconsistent solutions. Deep learning approaches have gained traction in medical imaging tasks due to their superior performance with larger datasets and ability to discern the intricate features of 3D volumes, a task inefficient if done manually. Specifically for the task of lung nodule segmentation, several different methods have been tried before such as region growing etc. but this project focuses on using an Attention U-Net model to automatically segment the nodule boundaries. Specifically, this is done on the LUNA16 dataset as a benchmark which is a popular reference point for comparison. To achieve this, specifically, the Attention U-Net was trained with 5-fold cross-validation on the training dataset. In addition to the segmentation outputs, averaged training and validation curves over all folds were also shown as the model is trained for 70 epochs. To conclude, these results present a useful automated method to segment the lung nodules. In practical situations, this would be of significant help to radiologists as it is less error-prone and not as susceptible to inter-observer variability. These automated tools along with other radiologist interactions could potentially significantly improve patient outcomes.","abstract_has_math":false,"creators":["Tummala, Sree Snigdha"],"institution":null,"degree_name":"Master of Science in Computer Science","degree_level":"Project","degree_discipline":"School of Computer Science and Engineering","degree_department":null,"school":null,"contributors":["Jennifer Jin"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:09:24Z","subjects":["Lung Lesion Segmentation","CNNs","LUNA16 dataset","Medical Imaging","Segmentation Maps","Attention Mechanism","Other Computer Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://scholarworks.lib.csusb.edu/etd/1824","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://scholarworks.lib.csusb.edu/etd/1824","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://scholarworks.lib.csusb.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Ascholarworks.lib.csusb.edu%3Aetd-2988","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Jennifer Jin"]},{"key":"dc:creator","label":"Author","values":["Tummala, Sree Snigdha"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2023-11-06T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2023-12-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["School of Computer Science and Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Project"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Computer Science"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Lung Lesion Segmentation","CNNs","LUNA16 dataset","Medical Imaging","Segmentation Maps","Attention Mechanism","Other Computer Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://scholarworks.lib.csusb.edu/etd/1824"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["The amount of data generated in the medical imaging field, especially in a modern context, is growing significantly. As the amount of data grows, it's prudent to make use of automated techniques that can leverage datasets to solve problems that are error-prone or have inconsistent solutions. Deep learning approaches have gained traction in medical imaging tasks due to their superior performance with larger datasets and ability to discern the intricate features of 3D volumes, a task inefficient if done manually. Specifically for the task of lung nodule segmentation, several different methods have been tried before such as region growing etc. but this project focuses on using an Attention U-Net model to automatically segment the nodule boundaries. Specifically, this is done on the LUNA16 dataset as a benchmark which is a popular reference point for comparison. To achieve this, specifically, the Attention U-Net was trained with 5-fold cross-validation on the training dataset. In addition to the segmentation outputs, averaged training and validation curves over all folds were also shown as the model is trained for 70 epochs. To conclude, these results present a useful automated method to segment the lung nodules. In practical situations, this would be of significant help to radiologists as it is less error-prone and not as susceptible to inter-observer variability. These automated tools along with other radiologist interactions could potentially significantly improve patient outcomes."]},{"key":"dc:title","label":"Title","values":["LUNG LESION SEGMENTATION USING DEEP LEARNING APPROACHES"]}]}],"canonical_facts":{"dc:contributor":["Jennifer Jin"],"dc:creator":["Tummala, Sree Snigdha"],"dc:date.available":["2023-11-06T08:00:00Z"],"dc:date.created":["2023-12-01T08:00:00Z"],"dc:description.abstract":["The amount of data generated in the medical imaging field, especially in a modern context, is growing significantly. As the amount of data grows, it's prudent to make use of automated techniques that can leverage datasets to solve problems that are error-prone or have inconsistent solutions. Deep learning approaches have gained traction in medical imaging tasks due to their superior performance with larger datasets and ability to discern the intricate features of 3D volumes, a task inefficient if done manually. Specifically for the task of lung nodule segmentation, several different methods have been tried before such as region growing etc. but this project focuses on using an Attention U-Net model to automatically segment the nodule boundaries. Specifically, this is done on the LUNA16 dataset as a benchmark which is a popular reference point for comparison. To achieve this, specifically, the Attention U-Net was trained with 5-fold cross-validation on the training dataset. In addition to the segmentation outputs, averaged training and validation curves over all folds were also shown as the model is trained for 70 epochs. To conclude, these results present a useful automated method to segment the lung nodules. In practical situations, this would be of significant help to radiologists as it is less error-prone and not as susceptible to inter-observer variability. These automated tools along with other radiologist interactions could potentially significantly improve patient outcomes."],"dc:identifier":["https://scholarworks.lib.csusb.edu/etd/1824"],"dc:subject":["Lung Lesion Segmentation","CNNs","LUNA16 dataset","Medical Imaging","Segmentation Maps","Attention Mechanism","Other Computer Engineering"],"dc:title":["LUNG LESION SEGMENTATION USING DEEP LEARNING APPROACHES"],"thesis:degree_discipline":["School of Computer Science and Engineering"],"thesis:degree_level":["Project"],"thesis:degree_name":["Master of Science in Computer Science"]},"updated_at":"2026-11-19T08:09:24Z"}