Feature selection based on hybrid optimization for magnetic resonance imaging brain tumor classification and segmentation
Journal Title: Applied Medical Informatics - Year 2019, Vol 41, Issue 1
Abstract
With the health information technology being infused into clinical health, e-health is becoming a key factor in delivering improvements in the health sector. Brain tumor data feature selection is crucial for the development of a viable cancer detection system based on brain tumor data. Our study aimed to obtain an optimal feature subset through a hybrid algorithm of Simulated Annealing-Genetic Algorithms (SA-GA). Two real datasets of brain tumor Magnetic Resonance Images are used to assess the performances of the proposed approach. The first dataset was freely downloaded from the Harvard Medical School brain atlas. The second brain tumor dataset was created from Nanfang Hospital, Guangzhou, China, and General Hospital, Tianjing Medical University, China from 2005 to 2012. The proposed approach is compared to the methods of simulated annealing, genetic algorithm and with the state-of-the-art methods used separately. The obtained results show that SA-GA exceeds simulated annealing and genetic algorithms when they are applied in isolation, in terms of accuracy and computing time. The evaluation shows that our method overtakes the state-of-the-art methods with a segmentation accuracy rate of 97.82%±0.74 for glioma tumor and 95.12% ±3.21for pituitary tumor.
Authors and Affiliations
Ahmed KHARRAT, Mahmoud NEJI
Keywords
Related Articles
- EP ID EP655028
- DOI -
- Views 55
- Downloads 0
Cerebral Visual Impairment and Dysgenesis of Corpus Callosum in Multidisabled Children Aged 1 to 9 Years Old
Aims: To emphasize the functional vision characteristics in visually impaired multiple disabled children (MDVI) aged 2 to 9 years old related to brain damages on magnetic resonance imaging in different cortical and subco...
Using Electronic Health Record Systems in Medical Education – A Needs Assessment
The aim of this study was to determine the features needed in electronic health record systems by supervising physicians for teaching medical students and residents. We conducted in-depth semi-structured interviews with...
Correlation between vectorcardiographic and echocardiographic parameters in patients with cardiac resynchronization therapy
Introduction: Echocardiographic improvement of left ventricular (LV) contraction after cardiac resynchronization therapy (CRT) is an acute endpoint. However, small changes of CRT stimulation parameters may induce large v...
Analysis of the baccalaureate performances of nursing candidates at the Timişoara, Cluj-Napoca and Bucharest Universities
Introduction: The substantial technological developments in medicine and nursing may warrant research to improve the scientific curriculum during high school. The aim of this study was to analyse the high school baccalau...
Content-Based Image Retrieval: A Comprehensive User Interactive Simulation Tool for Endoscopic Image Databases
Until few years ago, radiological methods were widely used for the examination and investigation of the digestive tract. Today, wireless capsule endoscopy represents an innovative, noninvasive, effective solution that do...