TY - JOUR
T1 - Deep Learning and Improved Particle Swarm Optimization Based Multimodal Brain Tumor Classication
AU - Tahir, Ayesha Bin T.
AU - Khan, Muhamamd Attique
AU - Alhaisoni, Majed
AU - Khan, Junaid Ali
AU - Nam, Yunyoung
AU - Wang, Shui Hua
AU - Javed, Kashif
N1 - Publisher Copyright:
© 2021 Tech Science Press. All rights reserved.
PY - 2021/3/22
Y1 - 2021/3/22
N2 - A brain tumor reects abnormal cell growth. Challenges: Surgery, radiation therapy, and chemotherapy are used to treat brain tumors, but these procedures are painful and costly. Magnetic resonance imaging (MRI) is a non-invasive modality for diagnosing tumors, but scans must be interpretated by an expert radiologist. Methodology:We used deep learning and improved particle swarm optimization (IPSO) to automate brain tumor classication. MRI scan contrast is enhanced by ant colony optimization (ACO); the scans are then used to further train a pretrained deep learning model, via transfer learning (TL), and to extract features from two dense layers. We fused the features of both layers into a single, more informative vector. An IPSO algorithm selected the optimal features, which were classied using a support vector machine. Results: We analyzed high- and low-grade glioma images from the BRATS 2018 dataset; the identication accuracies were 99.9% and 99.3%, respectively. Impact: The accuracy of our method is signicantly higher than existing techniques; thus, it will help radiologists to make diagnoses, by providing a second opinion..
AB - A brain tumor reects abnormal cell growth. Challenges: Surgery, radiation therapy, and chemotherapy are used to treat brain tumors, but these procedures are painful and costly. Magnetic resonance imaging (MRI) is a non-invasive modality for diagnosing tumors, but scans must be interpretated by an expert radiologist. Methodology:We used deep learning and improved particle swarm optimization (IPSO) to automate brain tumor classication. MRI scan contrast is enhanced by ant colony optimization (ACO); the scans are then used to further train a pretrained deep learning model, via transfer learning (TL), and to extract features from two dense layers. We fused the features of both layers into a single, more informative vector. An IPSO algorithm selected the optimal features, which were classied using a support vector machine. Results: We analyzed high- and low-grade glioma images from the BRATS 2018 dataset; the identication accuracies were 99.9% and 99.3%, respectively. Impact: The accuracy of our method is signicantly higher than existing techniques; thus, it will help radiologists to make diagnoses, by providing a second opinion..
KW - Brain tumor
KW - classication
KW - contrast enhancement
KW - deep learning
KW - feature selection
UR - http://www.scopus.com/inward/record.url?scp=85103625328&partnerID=8YFLogxK
U2 - 10.32604/cmc.2021.015154
DO - 10.32604/cmc.2021.015154
M3 - Article
AN - SCOPUS:85103625328
SN - 1546-2218
VL - 68
SP - 1099
EP - 1116
JO - Computers, Materials and Continua
JF - Computers, Materials and Continua
IS - 1
ER -