A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Deep Learning-Based Convolutional Neural Network with Cuckoo Search Optimization for MRI Brain Tumour Segmentation
In recent scenario there is a huge requirement of image processing in various applications, namely, pattern recognition, Image compression, multimedia computing, remote sensing, secured image data communication, biomedical imaging and content-based image restoration. The medical image processing is the process and technique where the human body images are created for the purpose of medical field to examine, reveal or diagnose the diseases. The internal anatomy of the human body is visualized by the medical imaging technique without opening the body. Proposed research consist of various steps preprocessing used to remove noise, lung MRI images that are diagnosed by a radiologist are segmented using basic thresholding and morphological operations to extract the lung parenchyma. Next the ROIs of pleural effusion are extracted followed by the extraction of the ROIs of pneumothorax. Ten shape and texture features area, convex area, equivalent diameter, mean, eccentricity, solidity, perimeter, entropy, smoothness and standard deviation are extracted from the ROIs. The CNN is trained to identify the feature vectors belonging to the four class’s pleural effusion, pneumothorax, normal lung and chest CT slices affected by other diseases. When the query MRI slice is applied, based on the training received, the classifies the query slice into the two classes for pneumonia or not. The classified result parameter optimized using Cuckoo search optimization algorithm (CSO). CSO algorithm a non-greedy local heuristic approach is used to solve optimization issues. The optimization results exhibit an accuracy of 94.18%.
Deep Learning-Based Convolutional Neural Network with Cuckoo Search Optimization for MRI Brain Tumour Segmentation
In recent scenario there is a huge requirement of image processing in various applications, namely, pattern recognition, Image compression, multimedia computing, remote sensing, secured image data communication, biomedical imaging and content-based image restoration. The medical image processing is the process and technique where the human body images are created for the purpose of medical field to examine, reveal or diagnose the diseases. The internal anatomy of the human body is visualized by the medical imaging technique without opening the body. Proposed research consist of various steps preprocessing used to remove noise, lung MRI images that are diagnosed by a radiologist are segmented using basic thresholding and morphological operations to extract the lung parenchyma. Next the ROIs of pleural effusion are extracted followed by the extraction of the ROIs of pneumothorax. Ten shape and texture features area, convex area, equivalent diameter, mean, eccentricity, solidity, perimeter, entropy, smoothness and standard deviation are extracted from the ROIs. The CNN is trained to identify the feature vectors belonging to the four class’s pleural effusion, pneumothorax, normal lung and chest CT slices affected by other diseases. When the query MRI slice is applied, based on the training received, the classifies the query slice into the two classes for pneumonia or not. The classified result parameter optimized using Cuckoo search optimization algorithm (CSO). CSO algorithm a non-greedy local heuristic approach is used to solve optimization issues. The optimization results exhibit an accuracy of 94.18%.
Deep Learning-Based Convolutional Neural Network with Cuckoo Search Optimization for MRI Brain Tumour Segmentation
Green Energy,Technology
Lahby, Mohamed (editor) / Al-Fuqaha, Ala (editor) / Maleh, Yassine (editor) / Sivanantham, Kalimuthu (author)
2022-04-22
20 pages
Article/Chapter (Book)
Electronic Resource
English
Computed tomography , Convolutional neural network , Cuckoo search optimization , Segmentation algorithm , Feature extraction , MRI image , Application of medical imaging Environment , Sustainable Development , Power Electronics, Electrical Machines and Networks , Cities, Countries, Regions , Renewable and Green Energy , Communications Engineering, Networks , Energy
| DOAJ | 2022