Brain tumor detection: a long short-term memory (LSTM)-based learning model

Abstract

To overcome the problems of automated brain tumor classification, a novel approach is proposed based on long short-term memory (LSTM) model using magnetic resonance images (MRI). First, N4ITK and Gaussian filters having size 5 × 5 are used to boost the of multi-sequence MRI quality. The presented deep LSTM model having four layers is utilized for classification. In each layer, optimal hidden units (HU) are selected such as 200 HU, 225 HU, 200 HU and 225 HU, respectively. These hidden or concealed units are chosen after performing extensive experiments to acquire better results. The results are validated on different versions of BRATS datasets (BRATS 2012–15, 2018) and SISS-ISLES 2015 dataset. The presented method attained dice similarity coefficient (DSC) 1.00 on 2012 synthetic, 0.95 on 2013, 0.99 on 2013 Leader board, 0.99 on 2014, 0.98 on 2015, 0.99 on 2018 and 0.95 on SISS-ISLES 2015. The methodology is also checked on real patient’s cases of brain tumor collected from Pakistan ordinance factory and achieved 0.97 DSC. The results confirm that the presented method provides more help for radiologists to classify brain tumor precisely.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5

Abbreviations

\( x_{\text{t}} \) :

Input image

\( h_{t} \) :

Hidden layer

\( O_{t} \) :

Model output

B :

Bias

c t :

Cell state time step

R :

Recurrent weights

\( \varSigma \) :

Sigmoid activation function

\( \odot \) :

Hadamard product

RNN:

Recurrent neural network

NNs:

Feedforward neural networks

SE:

Sensitivity

SP:

Specificity

FN:

False negative

FP:

False positive

TN:

True negative

TP:

True positive

DSC:

Dice similarity coefficient

DWI:

Diffusion-weighted imaging

FNR:

False negative rate

FLAIR:

Fluid-attenuated inversion recovery

T1c:

T1-weighted contrast-enhanced

T1:

T1-weighted

RF:

Random forests

SVMs:

Support vector machines

CNNs:

Convolutional neural networks

MRFs:

Markov random fields

CEN:

Convolutional encoder networks

\( {\text{HGG}} \) :

High-grade glioma

CRFs:

Conditional random fields

\( {\text{LGG}} \) :

Low-grade glioma

KNN:

K-nearest neighbor

DT:

Decision trees

MRI:

Magnetic resonance images

JSI:

Jaccard similarity index

\( {\text{FPR}} \) :

False positive rate

PPV:

Positive predictive value

ACC:

Accuracy

References

  1. 1.

    Moise D, Madhusoodanan S (2006) Psychiatric symptoms associated with brain tumors: a clinical enigma. CNS Spectr 11:28–31

    Article  Google Scholar 

  2. 2.

    Morrison J (2015) When psychological problems mask medical disorders: a guide for psychotherapists. Guilford Publications, New York

    Google Scholar 

  3. 3.

    Sharif M et al (2018) Brain tumor segmentation and classification by improved binomial thresholding and multi-features selection. J Ambient Intell Humaniz Comput 1–20

  4. 4.

    Amin J et al (2017) A distinctive approach in brain tumor detection and classification using MRI. Pattern Recognit Lett. https://doi.org/10.1016/j.patrec.2017.10.036

    Article  Google Scholar 

  5. 5.

    Amin J et al (2018) Detection of brain tumor based on features fusion and machine learning. J Ambient Intell Humaniz Comput 1–17

  6. 6.

    Gooya A et al (2012) GLISTR: glioma image segmentation and registration. IEEE Trans Med Imaging 31:1941–1954

    Article  Google Scholar 

  7. 7.

    Singhal AB et al (2011) Ischemic stroke: basic pathophysiology and neuroprotective strategies. In: Acute ischemic stroke. Springer, pp 1–24

  8. 8.

    Maier O et al (2015) Extra tree forests for sub-acute ischemic stroke lesion segmentation in MR sequences. J Neurosci Methods 240:89–100

    Article  Google Scholar 

  9. 9.

    Menze BH et al (2015) The multimodal brain tumor image segmentation benchmark (BRATS). IEEE Trans Med Imaging 34:1993

    Article  Google Scholar 

  10. 10.

    Prastawa M et al (2004) A brain tumor segmentation framework based on outlier detection. Med Image Anal 8:275–283

    Article  Google Scholar 

  11. 11.

    Reza S, Iftekharuddin K (2014) Improved brain tumor tissue segmentation using texture features. In: Proceedings MICCAI BraTS (brain tumor segmentation challenge), pp 27–30

  12. 12.

    Kleesiek J et al (2014) Ilastik for multi-modal brain tumor segmentation. In: Proceedings MICCAI BraTS (brain tumor segmentation challenge), pp 12–17

  13. 13.

    Havaei M et al (2017) Brain tumor segmentation with deep neural networks. Med Image Anal 35:18–31

    Article  Google Scholar 

  14. 14.

    Pereira S et al (2016) Brain tumor segmentation using convolutional neural networks in MRI images. IEEE Trans Med Imaging 35:1240–1251

    Article  Google Scholar 

  15. 15.

    Havaei M et al (2015) A convolutional neural network approach to brain tumor segmentation. In: International workshop on brainlesion: glioma, multiple sclerosis, stroke and traumatic brain injuries, pp 195–208

  16. 16.

    Dvořák P, Menze B (2015) Local structure prediction with convolutional neural networks for multimodal brain tumor segmentation. In: International MICCAI workshop on medical computer vision, pp 59–71

  17. 17.

    Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556

  18. 18.

    Fan J et al (2019) BIRNet: brain image registration using dual-supervised fully convolutional networks. Med Image Anal 54:193–206

    Article  Google Scholar 

  19. 19.

    Zhang X et al (2019) A survey on deep learning based brain computer interface: recent advances and new frontiers. arXiv:1905.04149

  20. 20.

    Zhou T et al (2019) Effective feature learning and fusion of multimodality data using stage-wise deep neural network for dementia diagnosis. Hum Brain Mapp 40:1001–1016

    Article  Google Scholar 

  21. 21.

    Farahani K et al (2014) Brats 2014 challenge manuscripts. http://www.braintumorsegmentation.org. Accessed 23 July 2019

  22. 22.

    Amin J et al (2018) Big data analysis for brain tumor detection: deep convolutional neural networks. Future Gener Comput Syst 87:290–297

    Article  Google Scholar 

  23. 23.

    https://medium.com/datathings/the-magic-of-lstm-neural-networks-6775e8b540cd. Accessed by 10 Apr 2019

  24. 24.

    Rajinikanth V et al (2017) Entropy based segmentation of tumor from brain MR images–a study with teaching learning based optimization. Pattern Recognit Lett 94:87–95

    Article  Google Scholar 

  25. 25.

    Raja NSM et al (2018) Contrast enhanced medical MRI evaluation using Tsallis entropy and region growing segmentation. J Ambient Intell Humaniz Comput 1–12

  26. 26.

    Fernandes SL et al (2019) A reliable framework for accurate brain image examination and treatment planning based on early diagnosis support for clinicians. Neural Comput Appl 1–12

  27. 27.

    Rajinikanth V et al (2019) Shannon’s entropy and watershed algorithm based technique to inspect ischemic Stroke wound. In: Smart intelligent computing and applications. Springer, pp 23–31

  28. 28.

    Hsieh TM et al (2011) Automatic segmentation of meningioma from non-contrasted brain MRI integrating fuzzy clustering and region growing. BMC Med Inform Decis Mak 11:54

    Article  Google Scholar 

  29. 29.

    Tustison NJ et al (2015) Optimal symmetric multimodal templates and concatenated random forests for supervised brain tumor segmentation (simplified) with ANTsR. Neuroinformatics 13:209–225

    Article  Google Scholar 

  30. 30.

    Juan-Albarracín J et al (2015) Automated glioblastoma segmentation based on a multiparametric structured unsupervised classification. PLoS ONE 10:e0125143

    Article  Google Scholar 

  31. 31.

    Pinto A et al (2015) Brain tumour segmentation based on extremely randomized forest with high-level features. In: Engineering in medicine and biology society (EMBC), 2015 37th annual international conference of the IEEE, pp 3037–3040

  32. 32.

    Soltaninejad M et al (2017) Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI. Int J Comput Assist Radiol Surg 12:183–203

    Article  Google Scholar 

  33. 33.

    Amin J et al (2017) A method for the detection and classification of diabetic retinopathy using structural predictors of bright lesions. J Comput Sci 19:153–164

    Article  Google Scholar 

  34. 34.

    Reza S, Iftekharuddin K (2013) Multi-class abnormal brain tissue segmentation using texture. Multimodal Brain Tumor Segmentation 38

  35. 35.

    Pereira S et al. (2013) Automatic brain tissue segmentation of multi-sequence MR images using random decision forests. In: Proceedings of the MICCAI grand challenge on MR brain image segmentation (MRBrainS’13)

  36. 36.

    Meier R et al (2013) A hybrid model for multimodal brain tumor segmentation. Multimodal Brain Tumor Segmentation 31:31–37

    Google Scholar 

  37. 37.

    Roth HR et al. (2015) Deeporgan: multi-level deep convolutional networks for automated pancreas segmentation. In: International conference on medical image computing and computer-assisted intervention, pp 556–564

  38. 38.

    Erhan D et al (2010) Why does unsupervised pre-training help deep learning? J Mach Learn Res 11:625–660

    MathSciNet  MATH  Google Scholar 

  39. 39.

    Amin J et al (2018) Diabetic retinopathy detection and classification using hybrid feature set. Microsc Res Tech 81:990–996

    Article  Google Scholar 

  40. 40.

    Sharif M et al (2018) A framework for offline signature verification system: best features selection approach. Pattern Recognit Lett. https://doi.org/10.1016/j.patrec.2018.01.021

    Article  Google Scholar 

  41. 41.

    Bokhari F et al (2018) Fundus image segmentation and feature extraction for the detection of glaucoma: a new approach. Curr Med Imaging Rev 14:77–87

    Article  Google Scholar 

  42. 42.

    Naqi S et al (2018) Lung nodule detection using polygon approximation and hybrid features from CT images. Curr Med Imaging Rev 14:108–117

    Article  Google Scholar 

  43. 43.

    Liaqat A et al (2018) Automated ulcer and bleeding classification from WCE images using multiple features fusion and selection. J Mech Med Biol 18:1850038

    Article  Google Scholar 

  44. 44.

    Zikic D et al (2014) Segmentation of brain tumor tissues with convolutional neural networks. In: Proceedings MICCAI-BRATS, pp 36–39

  45. 45.

    Dollár P, Zitnick CL (2013) Structured forests for fast edge detection. In: 2013 IEEE international conference on computer vision (ICCV), pp 1841–1848

  46. 46.

    Rao V et al (2015) Brain tumor segmentation with deep learning. In: MICCAI multimodal brain tumor segmentation challenge (BraTS), pp 56–59

  47. 47.

    Vaidhya K et al (2015) Multi-modal brain tumor segmentation using stacked denoising autoencoders. In: International workshop on brainlesion: glioma, multiple sclerosis, stroke and traumatic brain injuries, 2015, pp 181–194

  48. 48.

    Brosch T et al (2015) Deep convolutional encoder networks for multiple sclerosis lesion segmentation. In: International conference on medical image computing and computer-assisted intervention, 2015, pp 3–11

  49. 49.

    Cong W et al (2016) A modified brain MR image segmentation and bias field estimation model based on local and global information. In: Computational and mathematical methods in medicine, vol 2016

  50. 50.

    Blazejewska AI et al (2019) Intracortical smoothing of small-voxel fMRI data can provide increased detection power without spatial resolution losses compared to conventional large-voxel fMRI data. NeuroImage 189:601–614

    Article  Google Scholar 

  51. 51.

    Tustison NJ et al (2010) N4ITK: improved N3 bias correction. IEEE Trans Med Imaging 29:1310–1320

    Article  Google Scholar 

  52. 52.

    Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9:1735–1780

    Article  Google Scholar 

  53. 53.

    Zhao X et al (2018) A deep learning model integrating FCNNs and CRFs for brain tumor segmentation. Med Image Anal 43:98–111

    Article  Google Scholar 

  54. 54.

    Bakas S et al (2018) Identifying the best machine learning algorithms for brain tumor segmentation, progression assessment, and overall survival prediction in the BRATS challenge. In: Computer vision and pattern recognition, pp 1–49

  55. 55.

    Raup DM, Crick RE (1979) Measurement of faunal similarity in paleontology. J Paleontol 1213–1227

  56. 56.

    Rehman ZU et al (2019) Fully automated multi-parametric brain tumour segmentation using superpixel based classification. Expert Syst Appl 118:598–613

    Article  Google Scholar 

  57. 57.

    Chang J et al (2019) A mix-pooling CNN architecture with FCRF for brain tumor segmentation. J Vis Commun Image Represent 58:316–322

    Article  Google Scholar 

  58. 58.

    Selvapandian A, Manivannan K (2018) Fusion based glioma brain tumor detection and segmentation using ANFIS classification. Comput Methods Programs Biomed 166:33–38

    Article  Google Scholar 

  59. 59.

    Li H et al (2019) A novel end-to-end brain tumor segmentation method using improved fully convolutional networks. Comput Biol Med 108:150–160

    Article  Google Scholar 

  60. 60.

    Wang Y et al (2019) Multimodal brain tumor image segmentation using WRN-PPNet. Comput Med Imaging Graph 75:56–65

    Article  Google Scholar 

  61. 61.

    Praveen G et al (2019) Combination of hand-crafted and unsupervised learned features for ischemic stroke lesion detection from Magnetic Resonance Images. Biocybern Biomed Eng 39:410–425

    Article  Google Scholar 

  62. 62.

    Mallick PK et al (2019) Brain MRI image classification for cancer detection using deep wavelet autoencoder-based deep neural network. IEEE Access 7:46278–46287

    Article  Google Scholar 

  63. 63.

    Sajid S et al (2019) Brain tumor detection and segmentation in MR images using deep learning. Arab J Sci Eng 44:1–13

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Muhammad Sharif.

Ethics declarations

Conflict of interest

All authors have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Amin, J., Sharif, M., Raza, M. et al. Brain tumor detection: a long short-term memory (LSTM)-based learning model. Neural Comput & Applic 32, 15965–15973 (2020). https://doi.org/10.1007/s00521-019-04650-7

Download citation

Keywords

  • MRI
  • LSTM
  • HU
  • Brain tumor
  • Detection
  • Prediction