An efficient brain tumor detection and classification using pre-trained convolutional neural network models.

Autor:	Rao KN; Department of ECE, MLR Institute of Technology, Hyderabad, India., Khalaf OI; Department of Solar, Al-Nahrain Research Center for Renewable Energy, Al-Nahrain University, Jadriya, Baghdad, Iraq., Krishnasree V; Department of ECE, VNR Vignana Jyothi Institute of Engineering and Technology, Telangana, India., Kumar AS; Department of ECE, VNR Vignana Jyothi Institute of Engineering and Technology, Telangana, India., Alsekait DM; Department of Computer Science and Information Technology, Applied College, Princess Nourah Bint Abdulrahman University, P.O. Box 84428, Riyadh 11671, Saudi Arabia., Priyanka SS; Department of ECE, Chaitanya Bharathi Institute of Technology, Telangana, India., Alattas AS; Information Science Department, Faculty of Arts & Humanities, King Abdul-Aziz University, Jeddah, Saudi Arabia., AbdElminaam DS; MEU Research Unit, Middle East University, Amman 11831, Jordan.; Jadara Research Center, Jadara University, Irbid, 21110, Jordan.
Jazyk:	angličtina
Zdroj:	Heliyon [Heliyon] 2024 Aug 26; Vol. 10 (17), pp. e36773. Date of Electronic Publication: 2024 Aug 26 (Print Publication: 2024).
DOI:	10.1016/j.heliyon.2024.e36773
Abstrakt:	In cases of brain tumors, some brain cells experience abnormal and rapid growth, leading to the development of tumors. Brain tumors represent a significant source of illness affecting the brain. Magnetic Resonance Imaging (MRI) stands as a well-established and coherent diagnostic method for brain cancer detection. However, the resulting MRI scans produce a vast number of images, which require thorough examination by radiologists. Manual assessment of these images consumes considerable time and may result in inaccuracies in cancer detection. Recently, deep learning has emerged as a reliable tool for decision-making tasks across various domains, including finance, medicine, cybersecurity, agriculture, and forensics. In the context of brain cancer diagnosis, Deep Learning and Machine Learning algorithms applied to MRI data enable rapid prognosis. However, achieving higher accuracy is crucial for providing appropriate treatment to patients and facilitating prompt decision-making by radiologists. To address this, we propose the use of Convolutional Neural Networks (CNN) for brain tumor detection. Our approach utilizes a dataset consisting of two classes: three representing different tumor types and one representing non-tumor samples. We present a model that leverages pre-trained CNNs to categorize brain cancer cases. Additionally, data augmentation techniques are employed to augment the dataset size. The effectiveness of our proposed CNN model is evaluated through various metrics, including validation loss, confusion matrix, and overall loss. The proposed approach employing ResNet50 and EfficientNet demonstrated higher levels of accuracy, precision, and recall in detecting brain tumors. Competing Interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. (© 2024 The Author(s).)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu