Recent Advancements in Brain Tumor Segmentation and Classification using Deep Learning: A Review

Recent Advancements in Brain Tumor Segmentation and Classification using Deep Learning: A Review

Muhmmad Irfan Sharif

School of Computer Science and Engineering University of Electronic Science and Technology of China

Chengdu, China

Kaushik K

Srinivas Institute of Technology India.

AbstractBrain tumor is a deadlier disease with more fatality rate than the survival rate. Nowadays, with the advent of technology, it is becoming desirable to perform automated computer-based brain tumor image analysis. The two most important tasks covered in literature for brain tumor analysis are tumor segmentation and tumor classification. Deep learning approaches are gaining popularity because of their self-feature learning capability. The review aims to provide an introduction to recent works that use deep learning methodologies for brain tumor analysis. The manuscript first describes the problems in computer vision for brain tumor segmentation. Then, the recent cutting-edge works are discussed for segmenting brain tumor along with its prediction.

Keywords Brain tumor; classification; segmentation; deep learning; recent trends.

1. INTRODUCTION

   Machine learning (ML) [1-7] is a center of manmade consciousness [8]. It is the principal approach towards planning wise computers. AI for the most part utilizes acceptance and blend hence enabling PCs to get new information by reproducing human learning conduct. It redesigns current learning to ceaselessly improve computer execution. AI has additionally been broadly applied in numerous fields, for example, medical diagnosis [9-21], agriculture [22, 23], scene analysis [15, 24, 25] and computer vision [5, 26-33]. With the advancements of neural networks [34, 35], the idea of deep learning (DL) has been proposed. DL isn't just an improvement in neural systems, yet additionally another field in ML research. The effective utilization of DL brings ML closer to manmade reasoning. Since profound learning comprises of progressively concealed layers in contrast with neural systems, an increasingly conceptual significant level element portrayal for various classes is framed by utilizing different layers to join low-level features. Like computerized reasoning, deep adapting endeavors to fabricate and reproduce human mind to examine the learning procedure of neural system which mimics the learning component of human cerebrum when it happens to comprehend obscure ideas. DL can be utilized to consequently discover features from a given dataset for every application. By and large, traditional techniques depend on some earlier knowledge to concentrate in a specific application. In this manner, these techniques are hand crafted feature extraction strategies. DL can discover new features that are reasonable to

   explicit applications yet have never been recently found by specialists. These methodologies are also known as automatic features acquisition strategies [36-38]. The medical business is one of the most noticeable enterprises where DL can assume an immense job particularly with regards to medicinal imaging. There are various research works that demonstrated productivity and adequacy of the utilization of DL in the field. Brain is a vital part of human body [39] and comprised of many specific regions.

   Brain tumor is incited because of an ordinary boom of cells [40]. It comes under two categories that are malignant or benign. The benign tumors do not contain cancerous cells while the malignant tumors are comprised of cancerous cells [39, 41, 42]. Brain tumors associations [43] classify the brain tumor in four grades where grade I and II are referred to as benign and the remaining III and IV are labeled as malignant. The development rate of benign is relatively low when contrasted with malignant. When the benign isn't dealt with opportune then it is changed into malignant tumor.

   Along these lines, early discovery of tumor is desirable[44]. Grade II patients require ordinary treatment and checking through magnetic resonance imaging (MRI) [9, 42, 45, 46]. MRI depicts one of radio imaging types. Some more types include computed tomography (CT) and positron emission tomography (PET) [47] with the help of which brain tumor can be diagnosed in medical imaging[48]. MRI is mostly useful to analyze tumors in general clinical routines. It provides minute information about human cerebrum. One advantage of MRI is its non-intrusive nature and no ionization radiation.

   DL is most broadly utilized for brain tumor investigation in numerous scenarios for example, ordinary or irregular cerebrum tumor categorization and segmentation. A convolution neural network (CNN) represents a most famous DL structure utilized generally for characterization and segmentation of brain tumors. This is accomplished by convolving the images utilizing learned channels to assemble a chain of features activations. This deep learning work is done in a few layers to such an extent that the features acquired are interpretation and twisting invariant bringing about the high level of exactness.

   General computer vision steps for image analysis include: image acquiring (input), pre-processing [49-53], segmentation (to highlight required regions)[54], feature extraction (to

   acquire best characteristic points in an image) [55-57], features reduction (to acquire optimized features) [35, 41, 58- 61] and classification (to predict the image types e.g. HGG or LGG).

   Programmed division and prediction of brain tumors with the help of PC vision is significant for medical specialists as a second opinion for better treatment. Be that as it may, it stays hard for automated techniques to precisely segment and predict brain tumors. This is because of the reason that the tumor regions have a variety of variations in terms of shape, size and textures. Also change in the location of tumor regions is a real problem. Nowadays, deep learning strategies have become the cutting-edge strategies for brain tumor analysis as contrasted to old traditional methods. This manuscript mainly encompasses on a survey about recent trends for brain image segmentation and classification by employing DLstrategies. The major focus of this work is to give the readers an insight to these areas so that they can contribute further in an effective manner for brain tumor research.

2. LITERATURE REVIEW

1. Tumor segmentation

   To understand brain tumors and tumor structure, a preliminary study is presented below before discussing the literature for advanced deep methods in tumor segmentation.

   1. Preliminary study before segmenting tumor

      Gliomas are the most well-known brain tumors [62]. Some tumors because of low grade can be less destructive. Patients having such tumor can survive for many years [63]. Whereas some tumors of high grade are more destructive and can have survival duration of not more than two years [64]. Even though medical surgeries, which becomes a widely recognized treatment for brain cancers, along with other methods [65] are also the alternate solutions for tumor removal. MRI gives detailed pictures of cerebrum. Segmentation of some tumors, for instance, meningiomas can effectively be performed while it becomes hard to partition many other tumors, for example gliomas and glioblastomas[66]. These tumors are mostly combined with edema (swelling around the tumor region) and have poor contrast and structures [67]. This results in difficulty to fragment the tumor portion accurately. Additionally, images acquisition from different machines and different image sequence types may wind up having radically extraordinary grayscaleesteem when envisioned in various clinics. The theme of segmenting tumor is to localize the tumor region. Types of tumor tissues include dynamic tumor, necrotic and edema. Glioblastomas are tough to detect because of their outskirts being frequently fluffy and difficult to differentiate from normal or healthy tissues. More than one MRI sequences are considered which include T1, T1C, T2 and FLAIR.

   2. Deep segmentation strategies

      There is a variety of DL building blocks used by the scientists for segmenting the brain tumor in recent times. Some of such blocks contain deep convolutional neural networks (DCNNs) [68], recurrent neural networks (RNN), long short-term memory (LSTM), deep neural networks (DNN), auto encoders (AE) and generative adversarial networks (GAN). The accompanying sections discuss the existing literature in terms of these building blocks.

      Deep learning methodologies produce automatic features. The general strategy is to pass an image through the trained pipeline of deep learning building blocks and input image segmentation is performed depending on the deep features.

      1. DNN

         DNNs are the type of neural networks having a number of layers. DNN concentrates more on how the data from information is spoken to by means of a lot of nonlinear capacities before arriving at the yield layer. Havaei et al. [62] use a novel DNN model which encompasses both high and low level features. The authors claim that with support of GPUs, the approach is speedier as compared to cutting edge.

      2. RNN/ LSTM

         RNN is based on sequential data. S. Grivalskyet al. [69] select BraTS-17 dataset to perform HGG segmentation based on the proposed RNN architecture.

         LSTM is an upgraded form of RNNs intended to design for sequence data [70]. Each LSTM unit views a pixel and gets contribution from previous LSTM units. In this way, it recursively assembles data with all different pixels in the figure. A small number of research works are presented on brain tumor segmentation using LSTMs. Stollenga et al. [71] propose epic Pyramidal Multi-Dimensional LSTM (PyraMiD- LSTM) systems to utilize a somewhat extraordinary topology for the segmentation of tumor. The strategy is simpler to parallelize, requires less calculations in general, and performs best on GPU designs and 3D images. Good results of segmentation are achieved on MRBrainS13 dataset. LSTM- MA [72] utilizes multimodality based segmentation. Pixel- wise and super-pixel features are considered with LSTM classifier to perform semantic segmentation. The method is assessed on BrainWeb and MR BrainS datasets.

      3. AE

         AEs are another DL building blocks. Various variants of AE are used by researchers in order to segment brain tumor [73, 74]. Three layers stacked denoising AE is deployed in a research work to reconstruct the input image for segmentation [75]. In another work, deep spatial auto encoding strategy is used for segmenting the brain tumor. Some more works that emphasize auto encoders are [76-79].

      4. CNN

         CNN models are progressively perplexing with certain frameworks having excess of 100 layers which depicts a large number of loads and billions of associations between neurons [80, 81]. A normal CNN design comprises of ensuing layers of convolution, pooling, initiation and prediction. A large variety of works is found in literature that utilizes CNNs for segmenting brain.

         A novel method, Hough-CNN [82] is presented to deal with selected division by utilizing the deliberation qualities of CNNs. The technique depends on Hough casting, a methodology that takes into consideration completely programmed segmentation. This methodology doesn't just utilize CNN prediction results; however, it likewise performs casting by using the highlights delivered by the deepest segment of system.

         Javaria Amin et al. [32] introduce the concept of big data for DL in brain tumor segmentation. They propose seven-layer CNN architecture and assess their model on seven big MRI

         based tumor datasets. S. Hussain et al. [83] present five CNN methods to segment brain tumor. S Pereira et al. [84] use CNN with mini kernels, each having dimensions of 3×3. K Kamnitsas et al. [85] name their 3D architecture of CNN as DeepMedic. The authors include lingering connections on an early presented work. Cascaded fully CNN is applied for multi-class brain tumor extraction [86]. The anisotropic filters with multi fusion of layers make the approach more robust. U- net, also a variation of CNN, is used for tumor segmentation [87]. It is useful for training few numbers of images. A novel brain tumor division technique is proposed by incorporating a Fully-CNN with Conditional Random Fields (CRF) [88], as opposed to receiving CRF as a post-handling venture of FCNN. The model is prepared in three phases dependent on picture fixes and cuts individually. In another strategy [89], CNN is prepared legitimately on MRI modalities and in this manner gains a feature portrayal straightforwardly from the information.

         A cascaded design is proposed with two pathways: one which spotlights on little subtleties in gliomas and one on the bigger setting. A two-stage fix savvy learning methodology is likewise recommended that permits to learn models in a couple of hours. Convolutional idea of model completely permits to section a total mind picture in 25 sec to 3 min.

         Three distinctive 3D CNN models are investigated to take care of brain tumor division issue [90]. Two completely 3D CNN designs are suggested that propelled in two understood 2D models utilized for non exclusive image segmentation. A third model is additionally prepared which is a variation of two-pathway DeepMedic.

         S Pereira et al. propose Leaky Rectifier Linear Units (LRLU) [91] in their CNN model for tumor segmentation. The tumor is divided utilizing a completely programmed DL strategy called Input Cascade CNN [92]. An interesting CNN design contrasts from other customary CNNs because of its two-route preparation of image.

         U-net is also applied in [93] with dice loss function to handle unbalanced data. U-net is also considered in [94]. In this work, patch and semantic cascaded CNN models are proposed with the help of U-net architecture.

      5. GAN

         GAN is a variation of CNN and it produces great quality information by utilizing small data. GAN includes two turns, Generation turn and Discrimination turn. The former turn attempts to catch the model from which the information is taken hence creating pictures from irregular commotion inputs. The later stage utilizes traditional CNN that attempts to recognize genuine information and information produced by the generation stage.

      6. VoxResNet

         VoxResNet obtains the soul of deep residual learning. Chen et al. in their work [95] first propose a deep voxel wise lingering system alluded as VoxResNet which is then stretched out into a 3D variation for taking care of volumetric information, the learning on undertaking of volumetric cerebrum division.

      7. Ensemble methods

         Numerous research works are found that use ensemble of more than one DL building blocks to highlight brain tumor from MR images [96, 97]. Joint Sequence Learning (JSL) [98]

         highlights a hybrid method for multi modalities tumor segmentation. The proposed method depicts the blend of auto encoder, LSTM and a CNN. The two-sided learning mechanism is deployed to handle data imbalance. The model claims better segmentation results on BRATS 2015 dataset. Zhao et al. propose the merger of CNN with RNNs for effective tumor segmentation [99]. S. Iqbal et al. [100] use fusion of LSTM and CNN features to extract brain tumor region and perform evaluation on BRATS 2015 dataset. Y. Gao [101] introduce 4D MRI segmentation using hybrid of CNN and LSTM. The results are acquired on BRIC clinical and IBIS datasets. S. P. Ang et al. [102] also use LSTM and CNN based architecture for different brain tissues types segmentation. Better results are claimed on fMRI dataset. Table 1 depicts some more recent works regarding brain tumor segmentation.

         TABLE I. RECENT METHODOLOGIES FOR BRAIN TUMOR SEGMENTATION

         Ref	Year	Methods	Dataset BRATS	Result (Dice)
         				Complete	Core	Enhance
         [99]	2018	Integrating FCNNs and CRFs	2015	0.81	0.65	0.60
         [83]		Pach based DCNN	2015	0.86	0.87	0.90
         [103]	2019	Mixed supervision with U-Net Extension	2018	80.92	63.23	66.61
         [104]		PPNet	2018	0.94	–	–
         [105]		Dual-force CNN	2017	0.89	0.73	0.73
         [106]		Patch based hybrid CNN	2013	0.86	0.86	0.88
         [107]	2020	Rescue-Net	2017	0.9463	0.856	0.9354

2. Deep learning strategies for brain tumor classification

The classification phase is generally dependent on a classifier method. As a general methodology, image features are extracted and passed to classifier for predictions [108]. Most of the research is focused on CNNs. In one research work, DNN is utilized as a classifier [109]. The preprocessing, feature extraction and reduction are performed using fuzzy c- mean and discrete wavelet transform (DWT) approaches. The reduced features are then transferred to DNN for predictions.

Saba et al. introduce fusion between handcrafted and CNN features to perform tumor classification [110]. P. Afshar et al. use capsule network in their work for brain tumor prediction [111]. Capsule networks (alluded to as CapsNets) are fresh AI structures proposed to defeat the weaknesses of CNNs. More work on CapsNet is done in [112]. A new architecture of 3D CNN is proposed by D Nie et al. [113]. The deep features are obtained by the proposed deep model which are then trained by support vector machine (SVM) classifier for tumor predictions. YPan [114] perform grading of brain tumors using CNN. In a work by Yan Xu [115], a previous pre trained CNN model named AlexNet is used to collect deep features for tumor classification. CNN is also used in another study

for tumor classification. Pre-trained GoogLeNet [118] based features are used as transfer learning to classify brain tumor.

I. Shahzadi et al. use cascade of LSTM and CNN for tumor classification [119]. Pre-trained VGG-16 is used as a CNN building block and output of this network is supplied to LSTM network.Table 2 illustrates some more recent deep learning based methodologies regarding brain tumor segmentation

TABLE II. RECENT METHODOLOGIES FOR BRAIN TUMOR CLASSIFICATION

1. DATASETS FOR BRAIN TUMOR ANALYSIS

   It is observed as a practice that a huge dataset is used for performing deep learning tasks. For brain tumor analysis, various datasets are publicly offered. The images in most of them are available in cases form. One case represents images belonging to a particular patient. Normally one case contains 154-155 sliced human brain images. So, for example, 10 HGG cases of BRATS 2012 challenge subset means that there are total 155×10=1550 images in this subset.. The various datasets used for brain image analysis include RIDER [124], Harvard [125], IBSR [126], BRATS (2012-18) [127-129] and SISS-

   ISLES [130],

2. DISCUSSION

   As a general practice, in clinics, brain tumor analysis is performed manually by medical experts. This becomes a challenging job because of varying appearances and confusing brain structure. As a result, manual brain image analysis becomes tedious. On the other hand, automated segmentation and classification ease the task of neurologists because these methodologies assist them in making the final decision.

   The robustness in terms of speed makes such computer assisted systems a positive contributor in medical diagnostics. DL approaches are beneficial in brain tumor research by aiding in automatic feature acquisitions. This overcomes the consumption of time as compared to manual engineering of features. The advent of GPUs makes the computation processes very fast. In addition, performance increases with increase of training data. Apart from such advantages, some limitations are also observed by using DL approaches in brain tumor domain. The high cost of GPUs makes the DL process very costly. Also, no detailed literature is found to help in the selection of specific deep network architecture for particular brain analysis problem. This study will provide hint to the researchers in observing that what recent DL models are employed in brain analysis so that further research can be performed by keeping in view the existing DL methodologies.

3. CONCLUSION

A lot of work in performed in recent days on brain tumor MRI image segmentation and prediction with deep approaches. Still MRI is a challenging area where room for further research is available. The segmentation and classification both provide the medical experts a major advantage of second opinion based on automated results and a quick time analysis response. This saves a lot of time in manual brain image analysis. At the same time, this domain suffers because of robustness issues in terms of accuracy. This manuscript mainly focuses on existing DL techniques of segmenting and classifying brain tumors. In addition, publicly available datasets are also discussed.

REFERENCES

1. J. Amin, M. Sharif, M. Raza, and M. Yasmin, "Detection of Brain Tumor based on Features Fusion and Machine Learning," Journal of Ambient Intelligence and Humanized Computing, pp. 1-17, 2018.

2. J. Amin, M. Sharif, M. Raza, T. Saba, and M. A. Anjum, "Brain tumor detection using statistical and machine learning method," Computer methods and programs in biomedicine, vol. 177, pp. 69-79, 2019.

3. J. Amin, M. Sharif, M. Yasmin, T. Saba, and M. Raza, "Use of machine intelligence to conduct analysis of human brain data for detection of abnormalities in its cognitive functions," Multimedia Tools and Applications, pp. 1-19, 2019.

4. V. K. Jain, S. Kumar, and S. L. Fernandes, "Extraction of emotions from multilingual text using intelligent text processing and computational linguistics," Journal of Computational Science, vol. 21, pp. 316-326, 2017.

5. J. H. Shah, Z. Chen, M. Sharif, M. Yasmin, and S. L. Fernandes, "A novel biomechanics-based approach for person re-identification by generating dense color sift salience features," Journal of Mechanics in Medicine and Biology, vol. 17, p. 1740011, 2017.

6. S. L. Fernandes, V. P. Gurupur, N. R. Sunder, N. Arunkumar, and S. Kadry, "A novel nonintrusive decision support approach for heart rate measurement," Pattern Recognition Letters, 2017.

7. R. J. Martis and S. L. Fernandes, Frontiers in Development of Intelligent Systems for Human Biomechanics Using Soft Computing: World Scientific, 2017.

8. J. Liu, Y. Pan, M. Li, Z. Chen, L. Tang, C. Lu, and J. Wang, "Applications of deep learning to MRI images: A survey," Big Data Mining and Analytics, vol. 1, pp. 1-18, 2018.

9. T. Saba, S. T. F. Bokhari, M. Sharif, M. Yasmin, and M. Raza, "Fundus image classification methods for the detection of glaucoma: A review," Microscopy research and technique, vol. 81, pp. 1105-1121, 2018.

10. J. Amin, M. Sharif, A. Rehman, M. Raza, and M. R. Mufti, "Diabetic retinopathy detection and classification using hybrid feature set," Microscopy research and technique, vol. 81, pp. 990-996, 2018.

11. S. Masood, M. Sharif, M. Raza, M. Yasmin, M. Iqbal, and M. Younus Javed, "Glaucoma disease: A survey," Current Medical Imaging Reviews, vol. 11, pp. 272-283, 2015.

12. S. Masood, M. Sharif, A. Masood, M. Yasmin, and M. Raza, "A survey on medical image segmentation," Current Medical Imaging Reviews, vol. 11, pp. 3-14, 2015.

13. S. Masood, M. Sharif, M. Yasmin, M. Raza, and S. Mohsin, "Brain image Compression: A brief survey," Research Journal of Applied Sciences, Engineering and Technology, vol. 5, pp. 49-59, 2013.

14. S. Akbar, M. Sharif, M. U. Akram, T. Saba, T. Mahmood, and M. Kolivand, "Automated techniques for blood vessels segmentation through fundus retinal images: A review," Microscopy research and technique, vol. 82, pp. 153-170, 2019.

15. F. Bokhari, T. Syedia, M. Sharif, M. Yasmin, and S. L. Fernandes, "Fundus image segmentation and feature extraction for the detection of glaucoma: A new approach," Current Medical Imaging Reviews, vol. 14, pp. 77-87, 2018.

16. B. Mughal, N. Muhammad, M. Sharif, A. Rehman, and T. Saba, "Removal of pectoral muscle based on topographic map and shape- shifting silhouette," BMC cancer, vol. 18, p. 778, 2018.

17. M. A. Khan, T. Akram, M. Sharif, A. Shahzad, K. Aurangzeb, M. Alhussein, S. I. Haider, and A. Altamrah, "An implementation of normal distribution based segmentation and entropy controlled features

    selection for skin lesion detection and classification," BMC cancer, vol. 18, p. 638, 2018.

18. S. Naqi, M. Sharif, M. Yasmin, and S. L. Fernandes, "Lung nodule detection using polygon approximation and hybrid features from CT images," Current Medical Imaging Reviews, vol. 14, pp. 108-117, 2018.

19. S. L. Fernandes, V. P. Gurupur, H. Lin, and R. J. Martis, "A Novel fusion approach for early lung cancer detection using computer aided diagnosis techniques," Journal of Medical Imaging and Health Informatics, vol. 7, pp. 1841-1850, 2017.

20. N. Raja, V. Rajinikanth, S. L. Fernandes, and S. C. Satapathy, "Segmentation of breast thermal images using Kapur's entropy and hidden Markov random field," Journal of Medical Imaging and Health Informatics, vol. 7, pp. 1825-1829, 2017.

21. H. Ali, M. Sharif, M. Yasmin, and M. H. Rehmani, "Color-based template selection for detection of gastric abnormalities in video endoscopy," Biomedical Signal Processing and Control, vol. 56, p. 101668, 2020.

22. M. Sharif, M. A. Khan, Z. Iqbal, M. F. Azam, M. I. U. Lali, and M. Y. Javed, "Detection and classification of citrus diseases in agriculture based on optimized weighted segmentation and feature selection," Computers and electronics in agriculture, vol. 150, pp. 220-234, 2018.

23. H. Ali, M. Lali, M. Z. Nawaz, M. Sharif, and B. Saleem, "Symptom based automated detection of citrus diseases using color histogram and textural descriptors," Computers and electronics in agriculture, vol. 138, pp. 92-104, 2017.

24. G. J. Ansari, J. H. Shah, M. Yasmin, M. Sharif, and S. L. Fernandes, "A novel machine learning approach for scene text extraction," Future Generation Computer Systems, vol. 87, pp. 328-340, 2018.

25. R. J. Martis, V. P. Gurupur, H. Lin, A. Islam, and S. L. Fernandes, "Recent advances in big data analytics, internet of things and machine learning," ed: Elsevier, 2018.

26. S. M. Naqi, M. Sharif, and M. Yasmin, "Multistage segmentation model and SVM-ensemble for precise lung nodule detection," International journal of computer assisted radiology and surgery, vol. 13, pp. 1083-1095, 2018.

27. A. Liaqat, M. A. Khan, J. H. Shah, M. Sharif, M. Yasmin, and S. L. Fernandes, "Automated ulcer and bleeding classification from WCE images using multiple features fusion and selection," Journal of Mechanics in Medicine and Biology, vol. 18, p. 1850038, 2018.

28. M. Sharif, M. A. Khan, M. Faisal, M. Yasmin, and S. L. Fernandes, "A framework for offline signature verification system: Best features selection approach," Pattern Recognition Letters, 2018.

29. J. H. Shah, M. Sharif, M. Yasmin, and S. L. Fernandes, "Facial expressions classification and false label reduction using LDA and threefold SVM," Pattern Recognition Letters, 2017.

30. J. Amin, M. Sharif, M. Yasmin, H. Ali, and S. L. Fernandes, "A method for the detection and classification of diabetic retinopathy using structural predictors of bright lesions," Journal of Computational Science, vol. 19, pp. 153-164, 2017.

31. M. Raza, M. Sharif, M. Yasmin, M. A. Khan, T. Saba, and S. L. Fernandes, "Appearance based pedestrians gender recognition by employing stacked auto encoders in deep learning," Future Generation Computer Systems, vol. 88, pp. 28-39, 2018.

32. J. Amin, M. Sharif, M. Yasmin, and S. L. Fernandes, "Big data analysis for brain tumor detection: Deep convolutional neural networks," Future Generation Computer Systems, vol. 87, pp. 290-297, 2018.

33. V. Rajinikanth, N. Madhavaraja, S. C. Satapathy, and S. L. Fernandes, "Otsu's multi-thresholding and active contour snake model to segment dermoscopy images," Journal of Medical Imaging and Health Informatics, vol. 7, pp. 1837-1840, 2017.

34. M. Yasmin, M. Sharif, and S. Mohsin, "Neural networks in medical imaging applications: A survey," World Applied Sciences Journal, vol. 22, pp. 85-96, 2013.

35. M. A. Khan, T. Akram, M. Sharif, M. Y. Javed, N. Muhammad, and

    M. Yasmin, "An implementation of optimized framework for action classification using multilayers neural network on selected fused features," Pattern Analysis and Applications, vol. 22, pp. 1377-1397, 2019.

36. M. Fayyaz, M. Yasmin, M. Sharif, J. H. Shah, M. Raza, and T. Iqbal, "Person re-identification with features-based clustering and deep features," Neural Computing and Applications, pp. 1-22, 2019.

37. S. M. Naqi, M. Sharif, and A. Jaffar, "Lung nodule detection and classification based on geometric fit in parametric form and deep learning," Neural Computing and Applications, pp. 1-19, 2018.

38. M. Sharif, U. Tanvir, E. U. Munir, M. A. Khan, and M. Yasmin, "Brain tumor segmentation and classification by improved binomial thresholding and multi-features selection," Journal of Ambient Intelligence and Humanized Computing, pp. 1-20, 2018.

39. J. Amin, M. Sharif, M. Yasmin, and S. L. Fernandes, "A distinctive pproach in brain tumor detection and classification using MRI," Pattern Recognition Letters, 2017.

40. S. Sajjad, A. Hanan Abdullah, M. Sharif, and S. Mohsin, "Psychotherapy through video game to target illness related problematic behaviors of children with brain tumor," Current Medical Imaging Reviews, vol. 10, pp. 62-72, 2014.

41. V. Rajinikanth, S. C. Satapathy, S. L. Fernandes, and S. Nachiappan, "Entropy based segmentation of tumor from brain MR imagesa study with teaching learning based optimization," Pattern Recognition Letters, vol. 94, pp. 87-95, 2017.

42. V. Rajinikanth, S. L. Fernandes, B. Bhushan, and N. R. Sunder, "Segmentation and analysis of brain tumor using Tsallis entropy and regularised level set," in Proceedings of 2nd International Conference on Micro-Electronics, Electromagnetics and Telecommunications, 2018, pp. 313-321.

43. L. M. DeAngelis, "Brain tumors," New England Journal of Medicine,

    vol. 344, pp. 114-123, 2001.

44. J. Naik and S. Patel, "Tumor detection and classification using decision tree in brain MRI," International Journal of Computer Science and Network Security (IJCSNS), vol. 14, p. 87, 2014.

45. U. R. Acharya, S. L. Fernandes, J. E. WeiKoh, E. J. Ciaccio, M. K. M. Fabell, U. J. Tanik, V. Rajinikanth, and C. H. Yeong, "Automated Detection of Alzheimers Disease Using Brain MRI ImagesA Study with Various Feature Extraction Techniques," Journal of Medical Systems, vol. 43, p. 302, 2019.‌

46. S. L. Fernandes, U. J. Tanik, V. Rajinikanth, and K. A. Karthik, "A reliable framework for accurate brain image examination and treatment planning based on early diagnosis support for clinicians," Neural Computing and Applications, pp. 1-12, 2019.

47. R. A. Gatenby, O. Grove, and R. J. Gillies, "Quantitative imaging in cancer evolution and ecology," Radiology, vol. 269, pp. 8-14, 2013.

48. J. Amin, M. Sharif, M. A. Anjum, M. Raza, and S. A. C. Bukhari, "Convolutional neural network with batch normalization for glioma and stroke lesion detection using MRI," Cognitive Systems Research, vol. 59, pp. 304-311, 2020.

49. M. Yasmin, M. Sharif, S. Masood, M. Raza, and S. Mohsin, "Brain image enhancement-A survey," World Applied Sciences Journal, vol. 17, pp. 1192-1204, 2012.

50. M. Yasmin, S. Mohsin, M. Sharif, M. Raza, and S. Masood, "Brain image analysis: a survey," World Applied Sciences Journal, vol. 19, pp. 1484-1494, 2012.

51. M. Raza, M. Sharif, M. Yasmin, S. Masood, and S. Mohsin, "Brain image representation and rendering: A survey," Research Journal of Applied Sciences, Engineering and Technology, vol. 4, pp. 3274-3282, 2012.

52. M. Yasmin, M. Sharif, S. Masood, M. Raza, and S. Mohsin, "Brain image reconstruction: A short survey," World Applied Sciences Journal, vol. 19, pp. 52-62, 2012.

53. M. Yasmin, M. Sharif, S. Mohsin, and F. Azam, "Pathological brain image segmentation and classification: a survey," Current Medical Imaging Reviews, vol. 10, pp. 163-177, 2014.

54. M. Nasir, M. Attique Khan, M. Sharif, I. U. Lali, T. Saba, and T. Iqbal, "An improved strategy for skin lesion detection and classification using uniform segmentation and feature selection based approach," Microscopy research and technique, vol. 81, pp. 528-543, 2018.‌

55. R. Ranjan, R. Arya, S. L. Fernandes, E. Sravya, and V. Jain, "A fuzzy neural network approach for automatic K-complex detection in sleep EEG signal," Pattern Recognition Letters, vol. 115, pp. 74-83, 2018.

56. S. L Fernandes and J. G Bala, "A novel decision support for composite sketch matching using fusion of probabilistic neural network and dictionary matching," Current Medical Imaging Reviews, vol. 13, pp. 176-184, 2017.

57. S. Fernandes and J. Bala, "A comparative study on various state of the art face recognition techniques under varying facial expressions," Int. Arab J. Inf. Technol., vol. 14, pp. 254-259, 2017.

58. N. Arunkumar, K. Ramkumar, V. Venkatraman, E. Abdulhay, S. L. Fernandes, S. Kadry, and S. Segal, "Classification of focal and non

    focal EEG using entropies," Pattern Recognition Letters, vol. 94, pp. 112-117, 2017.

59. J. Amin, M. Sharif, M. Raza, T. Saba, and A. Rehman, "Brain Tumor Classification: Feature Fusion," in 2019 International Conference on Computer and Information Sciences (ICCIS), 2019, pp. 1-6.

60. M. I. Sharif, J. P. Li, M. A. Khan, and M. A. Saleem, "Active Deep neural Network Features Selection for Segmentation and Recognition of Brain Tumors using MRI Images," Pattern Recognition Letters, 2019.

61. M. I. Sharif, J. P. Li, J. Naz, and I. Rashid, "A Comprehensive Review on Multi-Organs Tumor Detection based on Machine Learning," Pattern Recognition Letters, 2019.

62. M. Havaei, A. Davy, D. Warde-Farley, A. Biard, A. Courville, Y. Bengio, C. Pal, P.-M. Jodoin, and H. Larochelle, "Brain tumor segmentation with deep neural networks," Medical image analysis, vol. 35, pp. 18-31, 2017.

63. V. Rajinikanth, S. C. Satapathy, N. Dey, and R. Vijayarajan, "DWT- PCA image fusion technique to improve segmentation accuracy in brain tumor analysis," in Microelectronics, Electromagnetics and Telecommunications, ed: Springer, 2018, pp. 453-462.

64. S. Bakas, M. Reyes, A. Jakab, S. Bauer, M. Rempfler, A. Crimi, R. T. Shinohara, C. Berger, S. M. Ha, and M. Rozycki, "Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge," arXiv preprint arXiv:1811.02629, 2018.

65. D. Wang, C. Wang, L. Wang, and Y. Chen, "A comprehensive review in improving delivery of small-molecule chemotherapeutic agents overcoming the blood-brain/brain tumor barriers for glioblastoma treatment," ed: Taylor & Francis, 2019.

66. M. L. White, Y. Zhang, F. Yu, N. Shonka, M. R. Aizenberg, P. Adapa, and S. A. J. Kazmi, "Post-operative perfusion and diffusion MR imaging and tumor progression in high-grade gliomas," PloS one, vol. 14, p. e0213905, 2019.‌

67. A. Chaudhari and J. Kulkarni, "Cerebral edema segmentation using textural feature," Biocybernetics and Biomedical Engineering, 2019.

68. A. In, C. Direkolu, and M. ah, "Review of MRI-based brain tumor image segmentation using deep learning methods," Procedia Computer Science, vol. 102, pp. 317-324, 2016.‌

69. S. Grivalsky, M. Tamajka, and W. Benesova, "Segmentation of gliomas in magnetic resonance images using recurrent neural networks," in 2019 42nd International Conference on Telecommunications and Signal Processing (TSP), 2019, pp. 539-542.

70. J. Amin, M. Sharif, M. Raza, T. Saba, R. Sial, and S. A. Shad, "Brain tumor detection: a long short-term memory (LSTM)-based learning model," Neural Computing and Applications, pp. 1-9, 2019.

71. M. F. Stollenga, W. Byeon, M. Liwicki, and J. Schmidhuber, "Parallel multi-dimensional LSTM, with application to fast biomedical volumetric image segmentation," in Advances in neural information processing systems, 2015, pp. 2998-3006.

72. K. Xie and Y. Wen, "LSTM-MA: A LSTM Method with Multi- Modality and Adjacency Constraint for Brain Image Segmentation," in 2019 IEEE International Conference on Image Processing (ICIP), 2019, pp. 240-244.‌

73. A. Myronenko, "3D MRI brain tumor segmentation using autoencoder regularization," in International MICCAI Brainlesion Workshop, 2018, pp. 311-320.

74. J. Amin, M. Sharif, N. Gul, M. Raza, M. A. Anjum, M. W. Nisar, and

    S. A. C. Bukhari, "Brain Tumor Detection by Using Stcked Autoencoders in Deep Learning," Journal of Medical Systems, vol. 44, p. 32, 2020.

75. K. Vaidhya, S. Thirunavukkarasu, V. Alex, and G. Krishnamurthi, "Multi-modal brain tumor segmentation using stacked denoising autoencoders," in BrainLes 2015, 2015, pp. 181-194.‌

76. V. Alex, K. Vaidhya, S. Thirunavukkarasu, C. Kesavadas, and G. Krishnamurthi, "Semisupervised learning using denoising autoencoders for brain lesion detection and segmentation," Journal of Medical Imaging, vol. 4, p. 041311, 2017.

77. Z. Xiao, R. Huang, Y. Ding, T. Lan, R. Dong, Z. Qin, X. Zhang, and

    W. Wang, "A deep learning-based segmentation method for brain tumor in MR images," in 2016 IEEE 6th International Conference on Computational Advances in Bio and Medical Sciences (ICCABS), 2016, pp. 1-6.

78. P. Korfiatis, T. L. Kline, and B. J. Erickson, "Automated segmentation of hyperintense regions in FLAIR MRI using deep learning," Tomography, vol. 2, p. 334, 2016.

79. G. Zhao, X. Wang, Y. Niu, L. Tan, and S.-X. Zhang, "Segmenting brain tissues from Chinese visible human dataset by deep-learned features with stacked autoencoder," BioMed research international, vol. 2016, 2016.

80. Z. Akkus, A. Galimzianova, A. Hoogi, D. L. Rubin, and B. J. Erickson, "Deep learning for brain MRI segmentation: state of the art and future directions," Journal of digital imaging, vol. 30, pp. 449-459, 2017.

81. M. Sharif, T. Akram, M. Raza, T. Saba, and A. Rehman, "Hand-crafted and deep convolutional neural network features fusion and selection strategy: an application to intelligent human action recognition," Applied Soft Computing, p. 105986, 2019.

82. F. Milletari, S.-A. Ahmadi, C. Kroll, A. Plate, V. Rozanski, J. Maiostre, J. Levin, O. Dietrich, B. Ertl-Wagner, and K. Bötzel, "Hough-CNN: deep learning for segmentation of deep brain regions in MRI and ultrasound," Computer Vision and Image Understanding, vol. 164, pp. 92-102, 2017.

83. S. Hussain, S. M. Anwar, and M. Majid, "Segmentation of glioma tumors in brain using deep convolutional neural network," Neurocomputing, vol. 282, pp. 248-261, 2018.

84. S. Pereira, A. Pinto, V. Alves, and C. A. Silva, "Brain tumor segmentation using convolutional neural networks in MRI images," IEEE transactions on medical imaging, vol. 35, pp. 1240-1251, 2016.

85. K. Kamnitsas, E. Ferrante, S. Parisot, C. Ledig, A. V. Nori, A. Criminisi, D. Rueckert, and B. Glocker, "DeepMedic for brain tumor segmentation," in International workshop on Brainlesion: Glioma, multiple sclerosis, stroke and traumatic brain injuries, 2016, pp. 138- 149.

86. G. Wang, W. Li, S. Ourselin, and T. Vercauteren, "Automatic brain tumor segmentation using cascaded anisotropic convolutional neural networks," in International MICCAI Brainlesion Workshop, 2017, pp. 178-190.

87. O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," in International Conference on Medical image computing and computer-assisted intervention, 2015, pp. 234-241.

88. X. Zhao, Y. Wu, G. Song, Z. Li, Y. Fan, and Y. Zhang, "Brain tumor segmentation using a fully convolutional neural network with conditional random fields," in International Workshop on Brainlesion: Glioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries, 2016, pp. 75-87.

89. M. Havaei, F. Dutil, C. Pal, H. Larochelle, and P.-M. Jodoin, "A convolutional neural network approach to brain tumor segmentation," in BrainLes 2015, 2015, pp. 195-208.

90. A. Casamitjana, S. Puch, A. Aduriz, E. Sayrol, and V. Vilaplana, "3d convolutional networks for brain tumor segmentation," Proceedings of the MICCAI Challenge on Multimodal Brain Tumor Image Segmentation (BRATS), pp. 65-68, 2016.

91. S. Pereira, A. Pinto, V. Alves, and C. A. Silva, "Deep convolutional neural networks for the segmentation of gliomas in multi-sequence MRI," in BrainLes 2015, 2015, pp. 131-143.

92. M. Sajjad, S. Khan, K. Muhammad, W. Wu, A. Ullah, and S. W. Baik, "Multi-grade brain tumor classification using deep CNN with extensive data augmentation," Journal of Computational Science, vol. 30, pp. 174-182, 2019.

93. F. Isensee, P. Kickingereder, W. Wick, M. Bendszus, and K. H. Maier- Hein, "Brain tumor segmentation and radiomics survival prediction: Contribution to the brats 2017 challenge," in International MICCAI Brainlesion Workshop, 2017, pp. 287-297.

94. H. Dong, G. Yang, F. Liu, Y. Mo, and Y. Guo, "Automatic brain tumor detection and segmentation using U-Net based fully convolutional networks," in annual conference on medical image understanding and analysis, 2017, pp. 506-517.

95. H. Chen, Q. Dou, L. Yu, and P.-A. Heng, "Voxresnet: Deep voxelwise residual networks for volumetric brain segmentation," arXiv preprint arXiv:1608.05895, 2016.

96. M. Sharif, J. Amin, M. Raza, M. Yasmin, and S. C. Satapathy, "An Integrated Design of Particle Swarm Optimization (PSO) with Fusion of Features for Detection of Brain Tumor," Pattern Recognition Letters, 2019.

97. J. Amin, M. Sharif, M. Yasmin, T. Saba, M. A. Anjum, and S. L. Fernandes, "A New Approach for Brain Tumor Segmentation and Classification Based on Score Level Fusion Using Transfer Learning," Journal of Medical Systems, vol. 43, p. 326, 2019.

98. K.-L. Tseng, Y.-L. Lin, W. Hsu, and C.-Y. Huang, "Joint sequence learning and cross-modality convolution for 3d biomedical

    segmentation," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 6393-6400.

99. X. Zhao, Y. Wu, G. Song, Z. Li, Y. Zhang, and Y. Fan, "A deep learning model integrating FCNNs and CRFs for brain tumor segmentation," Medical image analysis, vol. 43, pp. 98-111, 2018.

100. S. Iqbal, M. U. Ghani Khan, T. Saba, Z. Mehmood, N. Javaid, A. Rehman, and R. Abbasi, "Deep learning model integrating features and novel classifiers fusion for brain tumor segmentation," Microscopy research and technique, 2019.

101. Y. Gao, J. M. Phillips, Y. Zheng, R. Min, P. T. Fletcher, and G. Gerig, "Fully convolutional structured LSTM networks for joint 4D medical image segmentation," in 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), 2018, pp. 1104-1108.‌

102. S. P. Ang, S. L. Phung, M. M. Schira, A. Bouzerdoum, and S. T. M. Duong, "Human Brain Tissue Segmentation in fMRI using Deep Long- Term Recurrent Convolutional Network," in 2018 Digital Image Computing: Techniques and Applications (DICTA), 2018, pp. 1-7.

103. P. Mlynarski, H. Delingette, A. Criminisi, and N. Ayache, "Deep learning with mixed supervision for brain tumor segmentation," Journal of Medical Imaging, vol. 6, p. 034002, 2019.

104. Y. Wang, C. Li, T. Zhu, and J. Zhang, "Multimodal brain tumor image segmentation using WRN-PPNet," Computerized Medical Imaging and Graphics, vol. 75, pp. 56-65, 2019.

105. S. Chen, C. Ding, and M. Liu, "Dual-force convolutional neural networks for accurate brain tumor segmentation," Pattern Recognition, vol. 88, pp. 90-100, 2019.

106. S. Sajid, S. Hussain, and A. Sarwar, "Brain Tumor Detection and Segmentation in MR Images Using Deep Learning," Arabian Journal for Science and Engineering, vol. 44, pp. 9249-9261, 2019.

107. S. Nema, A. Dudhane, S. Murala, and S. Naidu, "RescueNet: An unpaired GAN for brain tumor segmentation," Biomedical Signal Processing and Control, vol. 55, p. 101641, 2020.

108. M. Sharif, J. Amin, M. W. Nisar, M. A. Anjum, and N. Ahmad, "A Unified Patch based Method for Brain Tumor Detection using Features Fusion," Cognitive Systems Research, 2019.

109. H. Mohsen, E.-S. A. El-Dahshan, E.-S. M. El-Horbaty, and A.-B. M. Salem, "Classification using deep learning neural networks for brain tumors," Future Computing and Informatics Journal, vol. 3, pp. 68-71, 2018.

110. T. Saba, A. S. Mohamed, M. El-Affendi, J. Amin, and M. Sharif, "Brain Tumor Detection using Fusion of Hand Crafted and Deep Learning Features," Cognitive Systems Research, 2019.

111. P. Afshar, A. Mohammadi, and K. N. Plataniotis, "Brain tumor type classification via capsule networks," in 2018 25th IEEE International Conference on Image Processing (ICIP), 2018, pp. 3129-3133.

112. P. Afshar, K. N. Plataniotis, and A. Mohammadi, "Capsule Networks for Brain Tumor Classification Based on Mri Images and Coarse Tumor Boundaries," in ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019, pp. 1368-1372.

113. D. Nie, H. Zhang, E. Adeli, L. Liu, and D. Shen, "3D deep learning for multi-modal imaging-guided survival time prediction of brain tumor patients," in International Conference on Medical Image Computing and Computer-Assisted Intervention, 2016, pp. 212-220.

114. Y. Pan, W. Huang, Z. Lin, W. Zhu, J. Zhou, J. Wong, and Z. Ding, "Brain tumor grading based on neural networks and convolutional neural networks," in 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2015, pp. 699-702.

115. Y. Xu, Z. Jia, Y. Ai, F. Zhang, M. Lai, I. Eric, and C. Chang, "Deep convolutional activation features for large scale brain tumor histopathology image classification and segmentation," in 2015 IEEE international conference on acoustics, speech and signal processing (ICASSP), 2015, pp. 947-951.

116. J. S. Paul, A. J. Plassard, B. A. Landman, and D. Fabbri, "Deep learning for brain tumor classification," in Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging, 2017, p. 1013710.‌

117. H. H. Sultan, N. M. Salem, and W. Al-Atabany, "Multi-classification of Brain Tumor Images using Deep Neural Network," IEEE Access, 2019.

118. S. Deepak and P. Ameer, "Brain tumor classification using deep CNN features via transfer learning," Computers in biology and medicine, vol. 111, p. 103345, 2019.

119. I. Shahzadi, T. B. Tang, F. Meriadeau, and A. Quyyum, "CNN-LSTM: Cascaded Framework For Brain Tumour Classification," in 2018 IEEE-EMBS Conference on Biomedical Engineering and Sciences (IECBES), 2018, pp. 633-637.

120. Z. N. K. Swati, Q. Zhao, M. Kabir, F. Ali, Z. Ali, S. Ahmed, and J. Lu, "Brain tumor classification for MR images using transfer learning and fine-tuning," Computerized Medical Imaging and Graphics, vol. 75, pp. 34-46, 2019.

121. F. Özyurt, E. Sert, and D. Avc, "An expert system for brain tumor detection: Fuzzy C-means with super resolution and convolutional neural network with extreme learning machine," Medical hypotheses, vol. 134, p. 109433, 2020.

122. T. Saba, A. S. Mohamed, M. El-Affendi, J. Amin, and M. Sharif, "Brain tumor detection using fusion of hand crafted and deep learning features," Cognitive Systems Research, vol. 59, pp. 221-230, 2020.

123. N. Ghassemi, A. Shoeibi, and M. Rouhani, "Deep neural network with generative adversarial networks pre-training for brain tumor classification based on MR images," Biomedical Signal Processing and Control, vol. 57, p. 101678, 2020.

124. S. Armato, Reinhard Beichel, Luc Bidaut, Larry Clarke, Bar- bara Croft, and M. G. e. a. Chuck Fenimore, "RIDER(Reference Database to Evaluate Response) Committee Combined Report, 9/25/2008 Sponsored by NIH, NCI, CIP, ITDB Causes of and Methods for Estimating/Ameliorating Variance in the Evaluation of Tumor Change in Response-to Therapy, https://wiki. cancerimagingarchive.net/display/Public/Collections.," 2008

125. D. Summers, "Harvard Whole Brain Atlas: www. med. harvard. edu/AANLIB/home. html," Journal of Neurology, Neurosurgery & Psychiatry, vol. 74, pp. 288-288, 2003.

126. M. Shakeri, S. Tsogkas, E. Ferrante, S. Lippe, S. Kadoury, N. Paragios, and I. Kokkinos, "Sub-cortical brain structure segmentation using F- CNN's," in Biomedical Imaging (ISBI), 2016 IEEE 13th International Symposium on, 2016, pp. 269-272.‌

127. S. Bakas, M. Reyes, A. Jakab, S. Bauer, M. Rempfler, A. Crimi, R. T. Shinohara, C. Berger, S. M. Ha, and M. Rozycki, "Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge," Computer Vision and Pattern Recognition, pp. 1-49, 2018.

128. B. H. Menze, A. Jakab, S. Bauer, J. Kalpathy-Cramer, K. Farahani, J. Kirby, Y. Burren, N. Porz, J. Slotboom, and R. Wiest, "The multimodal brain tumor image segmentation benchmark (BRATS)," IEEE transactions on medical imaging, vol. 34, pp. 1993-2024, 2014.

129. S. Bakas, H. Akbari, A. Sotiras, M. Bilello, M. Rozycki, J. S. Kirby, J.

     B. Freymann, K. Farahani, and C. Davatzikos, "Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features," Scientific data, vol. 4, p. 170117, 2017.

130. M. Kistler, S. Bonaretti, M. Pfahrer, R. Niklaus, and P. Büchler, "The virtual skeleton database: an open access repository for biomedical research and collaboration," Journal of medical Internet research, vol. 15, pp. 1-14, 2013.