Brain tumour detection in magnetic resonance imaging using Levenberg–Marquardt backpropagation neural network

Abstract

Magnetic resonance imaging (MRI) is a high-quality medical image that is used to detect brain tumours in a complex and time-consuming manner. In this study, a back propagation neural network (BPNN) along with the Levenberg-Marquardt algorithm (LMA) is proposed to classify MRIs and diagnose brain tumours in a simple and fast process. The BPNN has 10 neurons in the hidden layer, and the default function of the feedforward feeds is mean squared error (MSE). The LMA is optimized as a multivariable adaptive approach and considerably decreases the MSE of the BPNN, so the errors of the tumour classification are diminished. The proposed method follows four steps including preprocessing, skull removal, feature extraction, and classification. The input MRIs are converted to greyscale, resized, and thresholding is performed in the preprocessing step and followed by skull removal. Morphological operations of closing, opening, and dilation are used to segment abnormal areas in the MRIs, and the opening operator recognizes the tumour more accurately. Using statistical analysis and a grey-level co-occurrence matrix (GLCM) 12 features are extracted from the MRIs and used as the inputs of the BPNN. To evaluate the proposed method, 670 normal and 670 abnormal brain MRIs are used as input data, and the classification is performed in 0.494 s. The accuracy, sensitivity, specificity, precision, dice, recall, and MSE are 98.7%, 97.61%, 99.7%, 97.61%, 98.6%, 97.61%, and 0.005, respectively. The approach is accurate and fast for medical images classification.

An adaptive neuro‐fuzzy inference system optimized by genetic algorithm for brain tumour detection in magnetic resonance images

“…A longer box means more scattered data and a smaller box means less scattered data. In Figure 5, the GA‐ANFIS shows a smaller boxplot and less scattered data [26], which denotes a better performance.…”

“…After the morphological operation, the feature vectors are extracted and feature selection is performed for categorizing and recognition. The features of mean, standard deviation, entropy, energy, contrast, homogeneity, correlation, root mean square (RMS), variance, covariance, skewness, and kurtosis are extracted by the GLCM [26]. By feature selection, only appropriate and informative features are chosen.…”

“…Two scenarios of classical image processing (CIP) and binary image with variable fuzzy level (BIVFL) were applied to diagnose brain tumours [25], that are not fully automated. The Levenberg-Marquardt algorithm (LMA) assisted a backpropagation neural network (BPNN) to classify MRIs and diagnose brain tumours [26], this technique is fast but the tumour types are not classified. The tumours were detected and classified into benign (low grade) or malignant (high grade) by a type-II fuzzy inference system (FIS) and ANFIS [27].…”

“…[1] Adaptive neuro-fuzzy inference system and multilevel linear clustering algorithm ANFIS + MLCA [2] Multilayer perceptron MLP [3] K-nearest neighbour and fuzzy c-mean KNN + FCM [8] Neural gas network NGN [11] Non-subsampled contourlet transform NSCT [14] Non-dominated sorting genetic algorithm-II NSGA-II [16][17][18][19][20] Grey-level difference method GLDM [19] Gliomas using radiomics, GA features and extremely randomized trees GRGE [21] Artificial bee colony algorithm ABC [24] Principal component analysis PCA [24] Template-based k-means TK [25] Classical image processing CIP [25] Binary image with variable fuzzy level BIVFL [26] Levenberg-Marquardt algorithm LMA [26] Backpropagation neural network BPNN [27] Fuzzy inference system FIS […”

Brain tumour segmentation of MR images based on custom attention mechanism with transfer‐learning

“…Therefore, the development of an automatic or semi-automatic computeraided diagnostic (CAD) system in real medical therapies is needed to reduce the workload of physicians and improve accuracy. CAD system for brain tumours consists of tumour detection [4][5][6], segmentation [7][8][9], and classification [10][11][12][13] processes from MR images.…”