Salp Swarm Algorithm with Multilevel Thresholding Based Brain Tumor Segmentation Model  

作　　者：Hanan T.Halawani 

机构地区：[1]Computer Science Department,College of Computer Science and Information Systems,Najran University,Najran,55461,Saudi Arabia

出　　处：《Computers, Materials & Continua》2023年第3期6775-6788,共14页计算机、材料和连续体（英文）

基　　金：The author would like to express their gratitude to the Ministry of Education and the Deanship of Scientific Research-Najran University-Kingdom of Saudi Arabia for their financial and technical support under code number:NU/NRP/SERC/11/3.

摘　　要：Biomedical image processing acts as an essential part of severalmedical applications in supporting computer aided disease diagnosis. MagneticResonance Image (MRI) is a commonly utilized imaging tool used tosave glioma for clinical examination. Biomedical image segmentation plays avital role in healthcare decision making process which also helps to identifythe affected regions in the MRI. Though numerous segmentation models areavailable in the literature, it is still needed to develop effective segmentationmodels for BT. This study develops a salp swarm algorithm with multi-levelthresholding based brain tumor segmentation (SSAMLT-BTS) model. Thepresented SSAMLT-BTS model initially employs bilateral filtering based onnoise removal and skull stripping as a pre-processing phase. In addition,Otsu thresholding approach is applied to segment the biomedical imagesand the optimum threshold values are chosen by the use of SSA. Finally,active contour (AC) technique is used to identify the suspicious regions in themedical image. A comprehensive experimental analysis of the SSAMLT-BTSmodel is performed using benchmark dataset and the outcomes are inspectedin many aspects. The simulation outcomes reported the improved outcomesof the SSAMLT-BTS model over recent approaches with maximum accuracyof 95.95%.

关 键 词：Brain tumor segmentation noise removal multilevel thresholding healthcare PRE-PROCESSING 

分 类 号：R739.41[医药卫生—肿瘤] TP18[医药卫生—临床医学]