{"id":{"repo_id":"csusb","oai_identifier":"oai:scholarworks.lib.csusb.edu:etd-2760"},"canonical_url":"https://search.ndltd.org/etd/csusb/oai:scholarworks.lib.csusb.edu:etd-2760","repository":{"repo_id":"csusb","name":"CSUniversity San Bernardino","base_url":"https://scholarworks.lib.csusb.edu/do/oai/"},"display":{"title":"Brain Tumor Detection Using MRI Images","abstract":"When abnormal cells develop within the brain, a tumor is formed. Early tumor detection improves the likelihood of a patient's recovery. Compared to CT scan pictures, magnetic resonance imaging (MRI) is a trustworthy method for finding malignancies. In this project, we will use deep learning methods to detect tumors faster with higher accuracy using MRI images. Specifically, we will investigate the performance of transfer learning models based on convolutional neural networks (CNN) structures on the tumor detection problem. A machine learning approach called transfer learning uses a model already trained for the present task. The advantage of this technique is that we do not need to train the model from scratch, which will save time and increase accuracy. With the help of the Visual Geometry Group (VGG 16), Inception V3, and Resnet 50, this study attempts to identify brain tumors. It also uses a methodical approach for hyperparameter tuning to improve the trained models' accuracy. The main objective is to develop a practical approach for detecting brain tumors using MRIs to make quick, efficient, and precise decisions regarding the patients' conditions. Our suggested methodology is evaluated on the Kaggle dataset, taken from BRATS 2015 for brain tumor diagnosis using MRI images, including 3700 MRI brain images, with 3300 showing tumors. The simulation results show that training the deep learning models could achieve an accuracy of 96.0% for VGG-16, 94% for Resnet50, and 90.7% for the InceptionV3 model. In order to improve the accuracy even further, Bayesian Optimization is leveraged as a hyperparameter tuning technique to obtain the best set of parameters. We could achieve the accuracy of 97.5% for VGG-16, 95% for Resnet50, and 91.5% for InceptionV3.","abstract_html":"When abnormal cells develop within the brain, a tumor is formed. Early tumor detection improves the likelihood of a patient&#x27;s recovery. Compared to CT scan pictures, magnetic resonance imaging (MRI) is a trustworthy method for finding malignancies. In this project, we will use deep learning methods to detect tumors faster with higher accuracy using MRI images. Specifically, we will investigate the performance of transfer learning models based on convolutional neural networks (CNN) structures on the tumor detection problem. A machine learning approach called transfer learning uses a model already trained for the present task. The advantage of this technique is that we do not need to train the model from scratch, which will save time and increase accuracy. With the help of the Visual Geometry Group (VGG 16), Inception V3, and Resnet 50, this study attempts to identify brain tumors. It also uses a methodical approach for hyperparameter tuning to improve the trained models&#x27; accuracy. The main objective is to develop a practical approach for detecting brain tumors using MRIs to make quick, efficient, and precise decisions regarding the patients&#x27; conditions. Our suggested methodology is evaluated on the Kaggle dataset, taken from BRATS 2015 for brain tumor diagnosis using MRI images, including 3700 MRI brain images, with 3300 showing tumors. The simulation results show that training the deep learning models could achieve an accuracy of 96.0% for VGG-16, 94% for Resnet50, and 90.7% for the InceptionV3 model. In order to improve the accuracy even further, Bayesian Optimization is leveraged as a hyperparameter tuning technique to obtain the best set of parameters. We could achieve the accuracy of 97.5% for VGG-16, 95% for Resnet50, and 91.5% for InceptionV3.","abstract_has_math":false,"creators":["Patel, Mayur"],"institution":null,"degree_name":"Master of Science in Computer Science","degree_level":"Restricted Project: Campus only access","degree_discipline":"School of Computer Science and Engineering","degree_department":null,"school":null,"contributors":["Ghasemkhani Amir"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:08:51Z","subjects":["Brain Tumor","Artificial Intelligence","Hyperparameter Tuning","Bayesian Optimization","Other Computer Engineering"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[],"person_identifier_entries":[]},"links":{"outbound_url":"https://scholarworks.lib.csusb.edu/etd/1602","outbound_label":"Repository record","outbound_source":"dc:identifier","record_url":"https://scholarworks.lib.csusb.edu/etd/1602","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-2760","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Ghasemkhani Amir"]},{"key":"dc:creator","label":"Author","values":["Patel, Mayur"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.available","label":"Dc Date Available","values":["2022-12-14T08:00:00Z"]},{"key":"dc:date.created","label":"Dc Date Created","values":["2023-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["School of Computer Science and Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Restricted Project: Campus only access"]},{"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":["Brain Tumor","Artificial Intelligence","Hyperparameter Tuning","Bayesian Optimization","Other Computer Engineering"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://scholarworks.lib.csusb.edu/etd/1602"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["When abnormal cells develop within the brain, a tumor is formed. Early tumor detection improves the likelihood of a patient's recovery. Compared to CT scan pictures, magnetic resonance imaging (MRI) is a trustworthy method for finding malignancies. In this project, we will use deep learning methods to detect tumors faster with higher accuracy using MRI images. Specifically, we will investigate the performance of transfer learning models based on convolutional neural networks (CNN) structures on the tumor detection problem. A machine learning approach called transfer learning uses a model already trained for the present task. The advantage of this technique is that we do not need to train the model from scratch, which will save time and increase accuracy. With the help of the Visual Geometry Group (VGG 16), Inception V3, and Resnet 50, this study attempts to identify brain tumors. It also uses a methodical approach for hyperparameter tuning to improve the trained models' accuracy. The main objective is to develop a practical approach for detecting brain tumors using MRIs to make quick, efficient, and precise decisions regarding the patients' conditions. Our suggested methodology is evaluated on the Kaggle dataset, taken from BRATS 2015 for brain tumor diagnosis using MRI images, including 3700 MRI brain images, with 3300 showing tumors. The simulation results show that training the deep learning models could achieve an accuracy of 96.0% for VGG-16, 94% for Resnet50, and 90.7% for the InceptionV3 model. In order to improve the accuracy even further, Bayesian Optimization is leveraged as a hyperparameter tuning technique to obtain the best set of parameters. We could achieve the accuracy of 97.5% for VGG-16, 95% for Resnet50, and 91.5% for InceptionV3."]},{"key":"dc:title","label":"Title","values":["Brain Tumor Detection Using MRI Images"]}]}],"canonical_facts":{"dc:contributor":["Ghasemkhani Amir"],"dc:creator":["Patel, Mayur"],"dc:date.available":["2022-12-14T08:00:00Z"],"dc:date.created":["2023-01-01T08:00:00Z"],"dc:description.abstract":["When abnormal cells develop within the brain, a tumor is formed. Early tumor detection improves the likelihood of a patient's recovery. Compared to CT scan pictures, magnetic resonance imaging (MRI) is a trustworthy method for finding malignancies. In this project, we will use deep learning methods to detect tumors faster with higher accuracy using MRI images. Specifically, we will investigate the performance of transfer learning models based on convolutional neural networks (CNN) structures on the tumor detection problem. A machine learning approach called transfer learning uses a model already trained for the present task. The advantage of this technique is that we do not need to train the model from scratch, which will save time and increase accuracy. With the help of the Visual Geometry Group (VGG 16), Inception V3, and Resnet 50, this study attempts to identify brain tumors. It also uses a methodical approach for hyperparameter tuning to improve the trained models' accuracy. The main objective is to develop a practical approach for detecting brain tumors using MRIs to make quick, efficient, and precise decisions regarding the patients' conditions. Our suggested methodology is evaluated on the Kaggle dataset, taken from BRATS 2015 for brain tumor diagnosis using MRI images, including 3700 MRI brain images, with 3300 showing tumors. The simulation results show that training the deep learning models could achieve an accuracy of 96.0% for VGG-16, 94% for Resnet50, and 90.7% for the InceptionV3 model. In order to improve the accuracy even further, Bayesian Optimization is leveraged as a hyperparameter tuning technique to obtain the best set of parameters. We could achieve the accuracy of 97.5% for VGG-16, 95% for Resnet50, and 91.5% for InceptionV3."],"dc:identifier":["https://scholarworks.lib.csusb.edu/etd/1602"],"dc:subject":["Brain Tumor","Artificial Intelligence","Hyperparameter Tuning","Bayesian Optimization","Other Computer Engineering"],"dc:title":["Brain Tumor Detection Using MRI Images"],"thesis:degree_discipline":["School of Computer Science and Engineering"],"thesis:degree_level":["Restricted Project: Campus only access"],"thesis:degree_name":["Master of Science in Computer Science"]},"updated_at":"2026-11-19T08:08:51Z"}