Brain tumor detection using statistical and machine learning method

doi:10.1016/j.cmpb.2019.05.015

Computer Methods and Programs in Biomedicine

Volume 177, August 2019, Pages 69-79

https://doi.org/10.1016/j.cmpb.2019.05.015 Get rights and content

Highlights

•
The method is proposed for lesion enhancement using Weiner filter with different wavelet bands and different statistical methods are utilized for brain tumor segmentation.
•
The results of segmentation are analyzed in term of pixel and features based. In pixel-based to classify the foreground, background, error rate, quality (Q) and are compared with ground truth annotation.
•
In features based, local binary pattern (LBP) and Gabor wavelet transform (GWF) are extracted from each segmented image. Moreover, both texture features are fused for accurate classification.

Abstract

Background and Objective

Brain tumor occurs because of anomalous development of cells. It is one of the major reasons of death in adults around the globe. Millions of deaths can be prevented through early detection of brain tumor. Earlier brain tumor detection using Magnetic Resonance Imaging (MRI) may increase patient's survival rate. In MRI, tumor is shown more clearly that helps in the process of further treatment. This work aims to detect tumor at an early phase.

Methods

In this manuscript, Weiner filter with different wavelet bands is used to de-noise and enhance the input slices. Subsets of tumor pixels are found with Potential Field (PF) clustering. Furthermore, global threshold and different mathematical morphology operations are used to isolate the tumor region in Fluid Attenuated Inversion Recovery (Flair) and T2 MRI. For accurate classification, Local Binary Pattern (LBP) and Gabor Wavelet Transform (GWT) features are fused.

Results

The proposed approach is evaluated in terms of peak signal to noise ratio (PSNR), mean squared error (MSE) and structured similarity index (SSIM) yielding results as 76.38, 0.037 and 0.98 on T2 and 76.2, 0.039 and 0.98 on Flair respectively. The segmentation results have been evaluated based on pixels, individual features and fused features. At pixels level, the comparison of proposed approach is done with ground truth slices and also validated in terms of foreground (FG) pixels, background (BG) pixels, error region (ER) and pixel quality (Q). The approach achieved 0.93 FG and 0.98 BG precision and 0.010 ER on a local dataset. On multimodal brain tumor segmentation challenge dataset BRATS 2013, 0.93 FG and 0.99 BG precision and 0.005 ER are acquired. Similarly on BRATS 2015, 0.97 FG and 0.98 BG precision and 0.015 ER are obtained. In terms of quality, the average Q value and deviation are 0.88 and 0.017. At the fused feature based level, specificity, sensitivity, accuracy, area under the curve (AUC) and dice similarity coefficient (DSC) are 1.00, 0.92, 0.93, 0.96 and 0.96 on BRATS 2013, 0.90, 1.00, 0.97, 0.98 and 0.98 on BRATS 2015 and 0.90, 0.91, 0.90, 0.77 and 0.95 on local dataset respectively.

Conclusion

The presented approach outperformed as compared to existing approaches.

Introduction

The brain is a complex human body organ and works through billions of cells. Brain tumors are caused due to the uncontrolled growth of cells. These cells might affect normal brain activities and also destroy normal cells [1], [2]. Gliomas are a primary type of brain tumors. It contains IV grades (I, II: High Grade (HG) and III, IV: Low Grade (LG)) [3]. Despite major developments in the medical field like surgical procedure, chemotherapy and radiotherapy, still malignant brain tumor cases are untreatable. According to reports, brain tumor is the 5th primary reason for the death of women from twenty to thirty nine years [4].

MRI is particularly helpful to validate gliomas because it gives detailed structure about the human brain [5]. For the brain analysis, T1c, Tl, Flair and T2 are the most commonly used MRI sequences. These sequences provide different information related to the brain tumor. MRI test is based on radio frequency (RF) in which time of repetition (TR) and time of echo (TE) is calculated. T1c and T1 MRI are produced through short TR and TE time and cerebrospinal fluid (CSF) is dark in these sequences. CSF is colorless found in brain/ spinal cord. T2 MRI is produced by larger TR and TE time; therefore, CSF is bright in this modality. Flair is produced through very large TR and TE time; hence abnormal region remains bright and CSF is dark [6]. T2 is more sensitive to water content and more suitable for the abnormal region where water accumulates inside tissues of the brain. Flair provides more clear differentiation between the abnormal region and CSF as compared to other modalities (Fig. 1).

For brain tumor detection, accurate segmentation is a crucial process. The manual segmentation is time consuming for the radiologist [7] and hence, automated/ semi-automated techniques are required for precise tumor detection [8], [9]. Nowadays fully automated methods [10] for classification between the tumor and non-tumor MRI [11] are common for clinical and research studies. These methods can provide more help to analyze the tumor region [12] and are developed rapidly in the last ten years [13]. Therefore, radiologists assume that computerized methods can improve their diagnosis abilities on the basis of automated machine learning methods [14], [15], [16], [17]. Despite more efforts and promising outcomes in medical images analysis, still reproducible and correct segmentation along with abnormalities characterization is a challenging task of brain tumor detection because of its variability in location, shape and size [18], [19]. Presenting an automated method to increase tumor detection performance is a motivation of this work. The contributions are as follows:

1.
Lesion enhancement: Weiner Filter in wavelet domain is employed for noise removal and to enhance lesion region.
2.
Lesion segmentation: PF clustering method is applied to find out the subset of tumor pixels. Moreover, the global threshold method is used for lesion segmentation. Finally, mathematical morphology is applied for refining the tumor segmentation results.
3.
Feature extraction: In this phase, GWT and LBP features are obtained through every segmented image and then both texture features are fused to improve discrimination results.

The paper is organized into different sections. Existing work for brain lesions is discussed in Section 2. The presented approach for tumor detection is presented in Section 3 and 4 shows the achieved outcomes. The conclusion of the presented work is defined in Section 5.

Section snippets

Related work

In the past, many approaches are proposed for the detection of tumor using MRI. Abbasi and Tajeripour [20] suggested a method in which preprocessing is performed using histogram matching and bias field correction. Various machine learning techniques are also used for tumor detection using MRI [21], [22] including k-means, Fuzzy C-Means (FCM), thresholding, kernel extreme learning machine (KELM) and level set methods. Yamamoto et al. [23] proposed rule based and level set methods for tumor

Proposed methodology

The presented method for tumor detection is shown in Fig. 2. Weiner filter with different wavelet bands is used for noise reduction and to enhance the region of interest (ROI), PF clustering is applied for segmentation purpose. In addition, an optimal gray level thresholding method is also used. Different texture features are obtained from each segmented slice. Furthermore, multiple classifiers are utilized for separation of MR slices.

Experiments and results

The presented technique is checked on two publicly accessible datasets and one locally collected dataset. The local dataset contains 86 images, where 49 tumor and 37 non-tumor images were collected from Nishtar Hospital Multan, Pakistan. BRATS 2013 challenge dataset consists of thirty cases with ground truth annotations in which 20 belong to HG and 10 to LG tumors. BRATS 2015 has 273 cases in which 54 LG and 220 HG gliomas are included. Two modalities (Flair and T2) of each case are utilized

Conclusion

Detection at an early stage of brain tumor is presented in this article. The major contribution of this research work is an in-depth focus on enhancement, segmentation and features fusion process. The proposed hybrid texture features from each segmented slice are supplied to multiple classifiers for the classification of tumor /non-tumor MR slices. On the basis of detailed performance evaluation, it is confirmed that features fusion and KNN outperform as compared to other classifiers. The

Conflict of interest

None.

References (44)

M. Talo et al.
Application of deep transfer learning for automated brain abnormality classification using MR images
Cognit. Syst. Res.
(2019)
H. Fujita
Computer-aided diagnosis: the emerging of three CAD systems induced by Japanese health care needs
Comput. Methods Programs Biomed.
(2008)
J. Amin et al.
Big data analysis for brain tumor detection: deep convolutional neural networks
Future Gener. Comput. Syst.
(2018)
V. Simi et al.
Segmentation of glioblastoma multiforme from MR images–A comprehensive review
Egypt. J. Radiol. Nuclear Med.
(2015)
S. Abbasi et al.
Detection of brain tumor in 3D MRI images using local binary patterns and histogram orientation gradient
Neurocomputing
(2017)
E. Abdel-Maksoud et al.
Brain tumor segmentation based on a hybrid clustering technique
Egypt. Inform. J.
(2015)
D. Yamamoto
Computer-aided detection of multiple sclerosis lesions in brain magnetic resonance images: false positive reduction scheme consisted of rule-based, level set method, and support vector machine
Comput. Med. ImagingGraphics
(2010)
A. Gudigar et al.
Application of multiresolution analysis for automated detection of brain abnormality using MR images: a comparative study
Future Gener. Comput. Syst.
(2019)
E.-S.A. El-Dahshan et al.
Hybrid intelligent techniques for MRI brain images classification
Digital Signal Process.
(2010)
M. Graña
Computer aided diagnosis system for Alzheimer disease using brain diffusion tensor imaging features selected by Pearson's correlation
Neurosci. Lett.
(2011)

I. Cabria et al.

MRI segmentation fusion for brain tumor detection

Inf. Fusion

(2017)

A. Vishnuvarthanan et al.

An automated hybrid approach using clustering and nature inspired optimization technique for improved tumor and tissue segmentation in magnetic resonance brain images

Appl. Soft Comput.

(2017)

L. Liu et al.

Local binary features for texture classification: taxonomy and experimental study

Pattern Recognit.

(2017)

N. Nabizadeh et al.

Brain tumors detection and segmentation in MR images: gabor wavelet vs. statistical features

Comput. Electr. Eng.

(2015)

M. Sharif et al.

Detection and classification of citrus diseases in agriculture based on optimized weighted segmentation and feature selection

Comput. Electron. Agric.

(2018)

N. Gupta et al.

Glioma detection on brain MRIs using texture and morphological features with ensemble learning

Biomed. Signal Process. Control

(2019)

A. Pinto et al.

Hierarchical brain tumour segmentation using extremely randomized trees

Pattern Recognit.

(2018)

M. Havaei

Brain tumor segmentation with deep neural networks

Medical Image Anal.

(2017)

K. Kamnitsas

Efficient multi-scale 3D CNN with fully connected CRF for accurate brain lesion segmentation

Med. image Anal.

(2017)

X. Zhao et al.

A deep learning model integrating FCNNs and CRFs for brain tumor segmentation

Medical Image Anal.

(2018)

S. Chen et al.

Dual-force convolutional neural networks for accurate brain tumor segmentation

Pattern Recognit.

(2019)

A. Kavitha et al.

Brain tumor segmentation using genetic algorithm with SVM classifier

Int. J. Adv. Res. Electr. Electron. Instrum. Eng.

(2016)

Cited by (179)

A review on brain tumor segmentation based on deep learning methods with federated learning techniques
2023, Computerized Medical Imaging and Graphics
Brain tumors have become a severe medical complication in recent years due to their high fatality rate. Radiologists segment the tumor manually, which is time-consuming, error-prone, and expensive. In recent years, automated segmentation based on deep learning has demonstrated promising results in solving computer vision problems such as image classification and segmentation. Brain tumor segmentation has recently become a prevalent task in medical imaging to determine the tumor location, size, and shape using automated methods. Many researchers have worked on various machine and deep learning approaches to determine the most optimal solution using the convolutional methodology. In this review paper, we discuss the most effective segmentation techniques based on the datasets that are widely used and publicly available. We also proposed a survey of federated learning methodologies to enhance global segmentation performance and ensure privacy. A comprehensive literature review is suggested after studying more than 100 papers to generalize the most recent techniques in segmentation and multi-modality information. Finally, we concentrated on unsolved problems in brain tumor segmentation and a client-based federated model training strategy. Based on this review, future researchers will understand the optimal solution path to solve these issues.
FP-CNN: Fuzzy pooling-based convolutional neural network for lung ultrasound image classification with explainable AI
2023, Computers in Biology and Medicine
The COVID-19 pandemic wreaks havoc on healthcare systems all across the world. In pandemic scenarios like COVID-19, the applicability of diagnostic modalities is crucial in medical diagnosis, where non-invasive ultrasound imaging has the potential to be a useful biomarker. This research develops a computer-assisted intelligent methodology for ultrasound lung image classification by utilizing a fuzzy pooling-based convolutional neural network FP-CNN with underlying evidence of particular decisions. The fuzzy-pooling method finds better representative features for ultrasound image classification. The FPCNN model categorizes ultrasound images into one of three classes: covid, disease-free (normal), and pneumonia. Explanations of diagnostic decisions are crucial to ensure the fairness of an intelligent system. This research has used Shapley Additive Explanation (SHAP) to explain the prediction of the FP-CNN models. The prediction of the black-box model is illustrated using the SHAP explanation of the intermediate layers of the black-box model. To determine the most effective model, we have tested different state-of-the-art convolutional neural network architectures with various training strategies, including fine-tuned models, single-layer fuzzy pooling models, and fuzzy pooling at all pooling layers. Among different architectures, the Xception model with all pooling layers having fuzzy pooling achieves the best classification results of 97.2% accuracy. We hope our proposed method will be helpful for the clinical diagnosis of covid-19 from lung ultrasound (LUS) images.
HOG transformation based feature extraction framework in modified Resnet50 model for brain tumor detection
2023, Biomedical Signal Processing and Control
Brain tumor happens due to the instant and uncontrolled cell growth. It may lead to death if not cured at an early stage. In spite of several promising results and substantial efforts in this research area, the real challenge is to provide the accurate classification and segmentation. The key issue in brain tumor detection develops from the irregular changes in the tumor size, shape and location. In assessing the MRI images, computer-aided diagnoses are playing an extraordinary role and can help clinicians/radiologist. Nowadays, brain tumor has become the most incursive ailment that leads to a very short life expectancy when it reaches its highest grade. This research paper has created a new model using histogram of gradient (HOG) based neural features from MRI images for tumors detection. This research has conducted the feature optimization approach to achieve additional instinctive features from the complex feature vector. We developed a Modified ResNet50 model with HOG technique. The modified ResNet50 model can accurately extract the deep feature using deep learning approach. This model is applied along with the upgraded layered architecture in order to keep the optimal computational efficiency. We have also used the augmentation and feature extraction techniques using machine learning-based ensemble classifier that further provides the optimized fusion vector to identify the tumor. Such hybrid approach provides excellent performance with the detection accuracy of 88% with HOG and modified ResNet50. The results are also compared with the recent state of art methods.
Comparing of brain tumor diagnosis with developed local binary patterns methods
2024, Neural Computing and Applications
An Explainable AI Framework for Artificial Intelligence of Medical Things
2024, arXiv
Brain Tumor Detection and Categorization with Segmentation of Improved Unsupervised Clustering Approach and Machine Learning Classifier
2024, Bioengineering

View all citing articles on Scopus

View full text

Brain tumor detection using statistical and machine learning method

Highlights

Abstract

Background and Objective

Methods

Results

Conclusion

Introduction

Section snippets

Related work

Proposed methodology

Experiments and results

Conclusion

Conflict of interest

Cognit. Syst. Res.

Comput. Methods Programs Biomed.

Future Gener. Comput. Syst.

Egypt. J. Radiol. Nuclear Med.

Neurocomputing

Egypt. Inform. J.

Comput. Med. ImagingGraphics

Future Gener. Comput. Syst.

Digital Signal Process.

Neurosci. Lett.

Inf. Fusion

Appl. Soft Comput.

Pattern Recognit.

Comput. Electr. Eng.

Comput. Electron. Agric.

Biomed. Signal Process. Control

Pattern Recognit.

Medical Image Anal.

Med. image Anal.

Medical Image Anal.

Pattern Recognit.

Brain tumor segmentation using genetic algorithm with SVM classifier

Int. J. Adv. Res. Electr. Electron. Instrum. Eng.