Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
, 11 (1)

A Review on a Deep Learning Perspective in Brain Cancer Classification

Affiliations
Review

A Review on a Deep Learning Perspective in Brain Cancer Classification

Gopal S Tandel et al. Cancers (Basel).

Abstract

A World Health Organization (WHO) Feb 2018 report has recently shown that mortality rate due to brain or central nervous system (CNS) cancer is the highest in the Asian continent. It is of critical importance that cancer be detected earlier so that many of these lives can be saved. Cancer grading is an important aspect for targeted therapy. As cancer diagnosis is highly invasive, time consuming and expensive, there is an immediate requirement to develop a non-invasive, cost-effective and efficient tools for brain cancer characterization and grade estimation. Brain scans using magnetic resonance imaging (MRI), computed tomography (CT), as well as other imaging modalities, are fast and safer methods for tumor detection. In this paper, we tried to summarize the pathophysiology of brain cancer, imaging modalities of brain cancer and automatic computer assisted methods for brain cancer characterization in a machine and deep learning paradigm. Another objective of this paper is to find the current issues in existing engineering methods and also project a future paradigm. Further, we have highlighted the relationship between brain cancer and other brain disorders like stroke, Alzheimer's, Parkinson's, and Wilson's disease, leukoriaosis, and other neurological disorders in the context of machine learning and the deep learning paradigm.

Keywords: brain; cancer; deep learning; extreme learning; imaging; machine learning; neurological disorders; pathophysiology.

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Cell cycle proliferation. (image courtesy: AtheroPointTM, Roseville, CA, USA).
Figure 2
Figure 2
(a) Axial view, (b) Sagittal view, (c) Coronal view and (d) T1-weighted, (e) T2-weighted and (f) FLAIR Images of MRI. (image courtesy: AtheroPointTM ).
Figure 3
Figure 3
Working of ML-based algorithms.
Figure 4
Figure 4
Brain MR images: (a) normal brain, (b) benign tumor (7 O’ clock arrow) and (c) malignant tumor (7 O’ clock arrow) (reproduced from [63] with permission).
Figure 5
Figure 5
Process model of ANN-based classification model [63].
Figure 6
Figure 6
Hybrid characterization system for brain cancer characterization [88].
Figure 7
Figure 7
Illustration of different types as per their grades: row 1 and row 2 consists of T1ce brain images and its corresponding texture images, respectively. The images are pointed to by arrow are as follows: a1 (T1ce) and a2 (Texture): meningioma; b1 (T1ce) and b2 (Texture): Grade-II; c1 (T1ce), c2 (Texture): Grade-III; d1 (T1ce) and d2 (Texture): Grade-IV; e1 (T1ce) and e2 (Texture): metastasis (reproduced from [92] with permission).
Figure 8
Figure 8
Process model using SVM-based grade estimation method [92].
Figure 9
Figure 9
Process model of SVM-based grade estimation method [92].
Figure 10
Figure 10
Extreme learning machine.
Figure 11
Figure 11
CNN architecture (image courtesy: AtheroPointTM).
Figure 12
Figure 12
Segmentation results from two different patients. Class1: ground truth; Class 2 (enhancing region): green; Class 3 (necrotic region): yellow, Class 4 (T1abnormality-hypointensity region on T1, excluding enhancing and necrotic regions): red, and Class 5 (FLAIR abnormality excluding classes 2-4): blue (reproduced from [102] with permission).
Figure 13
Figure 13
Process model for segmentation [102].
Figure 14
Figure 14
Segmentation results from two different patients. Green: edema, yellow: enhanced tumor, pink: necrosis, blue: non-enhanced tumor (reproduced from [103] with permission).
Figure 15
Figure 15
Model Architecture (reproduced from [103] with permission).
Figure 16
Figure 16
Plausible solution for brain tumor grading.
Figure 17
Figure 17
Comparison of brain tumor with other brain disorders (image permission requested from sources). (a) Normal Brain [AtheroPointTM]; (b) Multiple Sclerosis [113]; (c) Stroke [114]; (d) Leukoaraiosis [115]; (e) Alzheimer’s Disease [116]; (f) Parkinson’s Disease [117]; (g) Wilson’sDisease [118]; (h) Brain Tumor [119].

Similar articles

See all similar articles

Cited by 5 PubMed Central articles

References

    1. International Agency for Research on Cancer. [(accessed on 1 November 2018)]; Available online: https://gco.iarc.fr/
    1. Brain Tumor Basics. [(accessed on 1 November 2018)]; Available online: https://www.thebraintumourcharity.org/
    1. American Cancer Society website. [(accessed on 1 November 2018)]; Available online: www.cancer.org/cancer.html.
    1. Brain Tumor Diagnosis. [(accessed on 1 November 2018)]; Available online: https://www.cancer.net/cancer-types/brain-tumor/diagnosis.
    1. WHO Statistics on Brain Cancer. [(accessed on 1 November 2018)]; Available online: http://www.who.int/cancer/en/
Feedback