Detection of brain abnormality by a novel Lu-Net deep neural CNN
model from MR images
Abstract The identification and classification of tumors in the human mind from MR images at an early stage play a pivotal role in diagnosis such diseases. This work presents the novel Deep Neural network with less number of layers and less complex in designed named U-Net (LU-Net) for the detection of tumors. The work is comprised of classifying the brain MR images into normal and abnormal class from the dataset of 253 images of high pixels. The MR images were 1st resized, cropped, preprocessed, and augmented for the accurate and fast training of deep neural models. The performance of the Lu-Net model is evaluated using five types of statistical assessment metrics Precision, Recall, Specificity, F-score, and Accuracy, and compared with the other two types of model Le-Net and VGG-16. The CNN models were trained and tested on augmented images and validation is performed on 50 untrained data. EXISTING SYSTEM
Input Image Gray Multihroushould Contour
Conversion Detection
Output CNN Edge
Classificatio Detection n
Fig 1. Existing System
DISADVANTAGES OF EXISTING SYSTEM:
Existing method achieved low performance
Prediction efficiency low They proposed an ANN model with combining optimizer named gray wolf optimizer (GWO). Pashaei et al. proposed a CNN model for the extraction of features from brain MRI in their work. In their CNN architecture model had five Conv layers with a 3 × 3 filter in each layer. They achieved 81% accuracy using their CNN model, ELM classifier model was used to enhance the performance where CNN features were used in it. Their Model performed well on pituitary tumor class but received low accuracy on meningioma which indicates the discrimination capability of the proposed classifier (Pashaei, Sajedi, & Jazayeri, 2018). They segmented the tumor image in the 1st step and then the augmentation process has been applied on it and in the final step they performed classification using CNN. They also segmented and augmented the images before classifying them. The training on images was performed with LSTM and SVM was used for the classification purpose. Muhammad Nazir et al. applied the BNN approach to classifying the two types of brain tumor benign and malignant. In the 1st step, they preprocessed the images using filtering, and then in the next step feature extraction was done by finding the average color moment from every MR image. PROPOSED SYSTEM
Image Data Pre-processing Fuzzy C-means
set Segmentation
GLCM
Output FF BNN Classification
Fig 2. Proposed System
ADVANTAGES OF PROPOSED SYSTEM:
Comparatively more accurate
Quality of the Image is increasing High Performance
SOFTWARE REQUIREMENTS:
Operating system : Windows XP/7
Coding Language : MATLAB Tool : MATLAB R2013A REFRENCES 1. Ahmed, Heba M., Youssef, Bayumy A. B., Elkorany, Ahmed S., Saleeb, Adel A., & Abd El-Samie, Fathi (2018). Hybrid gray wolf optimizer–artificial neural network classification approach for magnetic resonance brain images. Applied Optics, 57(7), B25. 2. Alfonse, Marco, & Salem, Abdel-Badeeh M. (2016). An automatic classification of brain tumors through MRI using support vector machine. Egyptian Computer Science Journal, 40(03), 1110–2586. 3.Anjali, R., & Priya, S. (2017). An efficient classifier for brain tumor classification. International Journal of Computer Science and Mobile Computing, 6(8), 40–48, www. ijcsmc.com. 3. Chakrabarty, Navoneel (2019). Brain MRI images for brain tumor detection. Kaggle. 4. Çinar, Ahmet, & Yildirim, Muhammed (2020). Detection of tumors on brain MRI images using the hybrid convolutional neural network architecture. Medical Hypotheses, 139(March),2020. 5. Varuna Shree, N., & Kumar, T. N. R. (2018). Identification and classification of brain tumor MRI images with feature extraction using DWT and probabilistic neural network.