Intelligent Ultra-Light Deep Learning Model for Multi-Class Brain Tumor Detection
Abstract
:1. Introduction
2. Material and Methods
2.1. Proposed Method
2.2. Dataset
2.3. Preprocessing
2.3.1. Intensity Normalization
2.3.2. Discrete Wavelets Based Decomposition
2.3.3. Augmentation
2.4. Ultra-Light Deep Learning Architecture-Based Feature Extraction
2.5. Textural Features
2.6. Ultra-Light Brain Tumor Detection System
2.7. Performance Measures
3. Experimental Results and Discussion
3.1. UL-BTD Framework for Fastest Detection Time/Image Analysis
3.2. UL-DLA and GLCM Exclusive Analysis
3.3. Effect of Image Size on Tumor Prediction
3.4. Sensitivity Analysis of Coding Schemes: OvA and OvO
3.5. Reliability Performance and Complexity Analysis of Proposed Model
3.6. Comparison with State-of-the-Art
3.7. Contribution and Implications
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Siegel, R.L.; Miller, K.D.; Fedewa, S.A.; Ahnen, D.J.; Meester, R.G.; Barzi, A.; Jemal, A. Colorectal cancer statistics, 2017. CA A Cancer J. Clin. 2017, 67, 177–193. [Google Scholar] [CrossRef]
- Goodenberger, M.L.; Jenkins, R.B. Genetics of adult glioma. Cancer Genet. 2012, 205, 613–621. [Google Scholar] [CrossRef] [PubMed]
- Louis, D.N.; Perry, A.; Reifenberger, G.; Von Deimling, A.; Figarella-Branger, D.; Cavenee, W.K.; Ohgaki, H.; Wiestler, O.D.; Kleihues, P.; Ellison, D.W. The 2016 World Health Organization classification of tumors of the central nervous system: A summary. Acta Neuropathol. 2016, 131, 803–820. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Hollon, T.C.; Pandian, B.; Adapa, A.R.; Urias, E.; Save, A.V.; Khalsa, S.S.S.; Eichberg, D.G.; D’Amico, R.S.; Farooq, Z.U.; Lewis, S. Near real-time intraoperative brain tumor diagnosis using stimulated Raman histology and deep neural networks. Nat. Med. 2020, 26, 52–58. [Google Scholar] [CrossRef] [PubMed]
- DePaoli, D.; Lemoine, É.; Ember, K.; Parent, M.; Prud’homme, M.; Cantin, L.; Petrecca, K.; Leblond, F.; Côté, D.C. Rise of Raman spectroscopy in neurosurgery: A review. J. Biomed. Opt. 2020, 25, 050901. [Google Scholar] [CrossRef] [PubMed]
- Sultan, H.H.; Salem, N.M.; Al-Atabany, W. Multi-classification of brain tumor images using deep neural network. IEEE Access 2019, 7, 69215–69225. [Google Scholar] [CrossRef]
- Hsieh, K.L.-C.; Lo, C.-M.; Hsiao, C.-J. Computer-aided grading of gliomas based on local and global MRI features. Comput. Methods Programs Biomed. 2017, 139, 31–38. [Google Scholar] [CrossRef] [PubMed]
- Sachdeva, J.; Kumar, V.; Gupta, I.; Khandelwal, N.; Ahuja, C.K. A package-SFERCB-“Segmentation, feature extraction, reduction and classification analysis by both SVM and ANN for brain tumors”. Appl. Soft Comput. 2016, 47, 151–167. [Google Scholar] [CrossRef]
- Cheng, J.; Huang, W.; Cao, S.; Yang, R.; Yang, W.; Yun, Z.; Wang, Z.; Feng, Q. Enhanced performance of brain tumor classification via tumor region augmentation and partition. PLoS ONE 2015, 10, e0140381. [Google Scholar] [CrossRef] [PubMed]
- Jun, C. Brain Tumor Dataset. 2017. Available online: https://figshare.com/articles/brain_tumor_dataset/1512427 (accessed on 10 November 2021).
- LeCun, Y.; Bengio, Y.; Hinton, G. Deep learning. Nature 2015, 521, 436–444. [Google Scholar] [CrossRef] [PubMed]
- Litjens, G.; Kooi, T.; Bejnordi, B.E.; Setio, A.A.A.; Ciompi, F.; Ghafoorian, M.; Van Der Laak, J.A.; Van Ginneken, B.; Sánchez, C.I. A survey on deep learning in medical image analysis. Med. Image Anal. 2017, 42, 60–88. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Swati, Z.N.K.; Zhao, Q.; Kabir, M.; Ali, F.; Ali, Z.; Ahmed, S.; Lu, J. Content-based brain tumor retrieval for MR images using transfer learning. IEEE Access 2019, 7, 17809–17822. [Google Scholar] [CrossRef]
- Soltaninejad, M.; Yang, G.; Lambrou, T.; Allinson, N.; Jones, T.L.; Barrick, T.R.; Howe, F.A.; Ye, X. Automated brain tumour detection and segmentation using superpixel-based extremely randomized trees in FLAIR MRI. Int. J. Comput. Assist. Radiol. Surg. 2017, 12, 183–203. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Soltaninejad, M.; Yang, G.; Lambrou, T.; Allinson, N.; Jones, T.L.; Barrick, T.R.; Howe, F.A.; Ye, X. Supervised learning based multimodal MRI brain tumour segmentation using texture features from supervoxels. Comput. Methods Programs Biomed. 2018, 157, 69–84. [Google Scholar] [CrossRef] [PubMed]
- Soltaninejad, M.; Zhang, L.; Lambrou, T.; Yang, G.; Allinson, N.; Ye, X. MRI brain tumor segmentation and patient survival prediction using random forests and fully convolutional networks. In Proceedings of the International MICCAI Brainlesion Workshop, Quebec City, QC, Canada, 14 September 2017; pp. 204–215. [Google Scholar]
- Zhang, W.; Yang, G.; Huang, H.; Yang, W.; Xu, X.; Liu, Y.; Lai, X. ME-Net: Multi-encoder net framework for brain tumor segmentation. Int. J. Imaging Syst. Technol. 2021, 31, 1834–1848. [Google Scholar] [CrossRef]
- Huang, H.; Yang, G.; Zhang, W.; Xu, X.; Yang, W.; Jiang, W.; Lai, X. A deep multi-task learning framework for brain tumor segmentation. Front. Oncol. 2021, 11, 690244. [Google Scholar] [CrossRef] [PubMed]
- Jin, Y.; Yang, G.; Fang, Y.; Li, R.; Xu, X.; Liu, Y.; Lai, X. 3D PBV-Net: An automated prostate MRI data segmentation method. Comput. Biol. Med. 2021, 128, 104160. [Google Scholar] [CrossRef]
- Liu, Y.; Yang, G.; Mirak, S.A.; Hosseiny, M.; Azadikhah, A.; Zhong, X.; Reiter, R.E.; Lee, Y.; Raman, S.S.; Sung, K. Automatic prostate zonal segmentation using fully convolutional network with feature pyramid attention. IEEE Access 2019, 7, 163626–163632. [Google Scholar] [CrossRef]
- Liu, Y.; Yang, G.; Hosseiny, M.; Azadikhah, A.; Mirak, S.A.; Miao, Q.; Raman, S.S.; Sung, K. Exploring uncertainty measures in Bayesian deep attentive neural networks for prostate zonal segmentation. IEEE Access 2020, 8, 151817–151828. [Google Scholar] [CrossRef]
- Guan, X.; Yang, G.; Ye, J.; Yang, W.; Xu, X.; Jiang, W.; Lai, X. 3D AGSE-VNet: An automatic brain tumor MRI data segmentation framework. BMC Med. Imaging 2022, 22, 6. [Google Scholar] [CrossRef]
- Rajput, Y.; Rajput, V.S.; Thakur, A.; Vyas, G. Advanced image enhancement based on wavelet & histogram equalization for medical images. IOSR J. Electron. Commun. Eng. 2012, 2, 12–16. [Google Scholar]
- Sajjad, M.; Khan, S.; Muhammad, K.; Wu, W.; Ullah, A.; Baik, S.W. Multi-grade brain tumor classification using deep CNN with extensive data augmentation. J. Comput. Sci. 2019, 30, 174–182. [Google Scholar] [CrossRef]
- Badža, M.M.; Barjaktarović, M.Č. Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network. Appl. Sci. 2020, 10, 1999. [Google Scholar] [CrossRef] [Green Version]
- Akbar, S.; Peikari, M.; Salama, S.; Nofech-Mozes, S.; Martel, A.L. The transition module: A method for preventing overfitting in convolutional neural networks. Comput. Methods Biomech. Biomed. Eng. Imaging Vis. 2019, 7, 260–265. [Google Scholar] [CrossRef] [PubMed]
- Pawara, P.; Okafor, E.; Groefsema, M.; He, S.; Schomaker, L.R.; Wiering, M.A. One-vs-One Classification for Deep Neural Networks. Pattern Recognit. 2020, 108, 107528. [Google Scholar] [CrossRef]
- Niu, X.-X.; Suen, C.Y. A novel hybrid CNN–SVM classifier for recognizing handwritten digits. Pattern Recognit. 2012, 45, 1318–1325. [Google Scholar] [CrossRef]
- Abraham, P.; Sarkar, R.; Brandel, M.G.; Wali, A.R.; Rennert, R.C.; Lopez Ramos, C.; Padwal, J.; Steinberg, J.A.; Santiago-Dieppa, D.R.; Cheung, V. Cost-effectiveness of intraoperative MRI for treatment of high-grade gliomas. Radiology 2019, 291, 689–697. [Google Scholar] [CrossRef] [PubMed]
- Lakomkin, N.; Hadjipanayis, C.G. The use of spectroscopy handheld tools in brain tumor surgery: Current evidence and techniques. Front. Surg. 2019, 6, 30. [Google Scholar] [CrossRef] [PubMed]
- Hu, S.; Kang, H.; Baek, Y.; El Fakhri, G.; Kuang, A.; Choi, H.S. Real-time imaging of brain tumor for image-guided surgery. Adv. Healthc. Mater. 2018, 7, 1800066. [Google Scholar] [CrossRef]
- Díaz-Pernas, F.J.; Martínez-Zarzuela, M.; Antón-Rodríguez, M.; González-Ortega, D. A Deep Learning Approach for Brain Tumor Classification and Segmentation Using a Multiscale Convolutional Neural Network. Healthcare 2021, 9, 153. [Google Scholar] [CrossRef]
- Schupper, A.J.; Rao, M.; Mohammadi, N.; Baron, R.; Lee, J.Y.; Acerbi, F.; Hadjipanayis, C.G. Fluorescence-guided surgery: A review on timing and use in brain tumor surgery. Front. Neurol. 2021, 12, 914. [Google Scholar] [CrossRef] [PubMed]
- Nguyen, Q.T.; Tsien, R.Y. Fluorescence-guided surgery with live molecular navigation—A new cutting edge. Nat. Rev. Cancer 2013, 13, 653–662. [Google Scholar] [CrossRef] [PubMed]
- Lindseth, F.; Langø, T.; Bang, J.; Nagelhus Hemes, T.A. Accuracy evaluation of a 3D ultrasound-based neuronavigation system. Comput. Aided Surg. 2002, 7, 197–222. [Google Scholar] [CrossRef] [PubMed]
- Sastry, R.; Bi, W.L.; Pieper, S.; Frisken, S.; Kapur, T.; Wells, W., III; Golby, A.J. Applications of ultrasound in the resection of brain tumors. J. Neuroimaging 2017, 27, 5–15. [Google Scholar] [CrossRef]
- Ganau, M.; Ligarotti, G.K.; Apostolopoulos, V. Real-time intraoperative ultrasound in brain surgery: Neuronavigation and use of contrast-enhanced image fusion. Quant. Imaging Med. Surg. 2019, 9, 350. [Google Scholar] [CrossRef] [PubMed]
- ul Rehman, A.; Qureshi, S.A. A review of the medical hyperspectral imaging systems and unmixing algorithms’ in biological tissues. Photodiagnosis Photodyn. Ther. 2020, 33, 102165. [Google Scholar] [CrossRef] [PubMed]
- Fabelo, H.; Ortega, S.; Lazcano, R.; Madroñal, D.; Callicó, G.M.; Juárez, E.; Salvador, R.; Bulters, D.; Bulstrode, H.; Szolna, A. An intraoperative visualization system using hyperspectral imaging to aid in brain tumor delineation. Sensors 2018, 18, 430. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Schulder, M.; Carmel, P.W. Intraoperative magnetic resonance imaging: Impact on brain tumor surgery. Cancer Control 2003, 10, 115–124. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Fan, Y.; Xia, Y.; Zhang, X.; Sun, Y.; Tang, J.; Zhang, L.; Liao, H. Optical coherence tomography for precision brain imaging, neurosurgical guidance and minimally invasive theranostics. Biosci. Trends 2018, 12, 12–23. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Kut, C.; Chaichana, K.L.; Xi, J.; Raza, S.M.; Ye, X.; McVeigh, E.R.; Rodriguez, F.J.; Quiñones-Hinojosa, A.; Li, X. Detection of human brain cancer infiltration ex vivo and in vivo using quantitative optical coherence tomography. Sci. Transl. Med. 2015, 7, ra100–ra292. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Anaraki, A.K.; Ayati, M.; Kazemi, F. Magnetic resonance imaging-based brain tumor grades classification and grading via convolutional neural networks and genetic algorithms. Biocybern. Biomed. Eng. 2019, 39, 63–74. [Google Scholar] [CrossRef]
- Paul, J.S.; Plassard, A.J.; Landman, B.A.; Fabbri, D. Deep learning for brain tumor classification. In Proceedings of the Medical Imaging 2017: Biomedical Applications in Molecular, Structural, and Functional Imaging, Orlando, FL, USA, 12–14 February 2017; p. 1013710. [Google Scholar]
- Afshar, P.; Mohammadi, A.; Plataniotis, K.N. Brain tumor type classification via capsule networks. In Proceedings of the 2018 25th IEEE International Conference on Image Processing (ICIP), Athens, Greece, 7–10 October 2018; pp. 3129–3133. [Google Scholar]
- Kurup, R.V.; Sowmya, V.; Soman, K. Effect of data pre-processing on brain tumor classification using capsulenet. In Proceedings of the International Conference on Intelligent Computing and Communication Technologies (ICICCT 2019), Istanbul, Turkey, 29–30 April 2019; pp. 110–119. [Google Scholar]
- Gumaei, A.; Hassan, M.M.; Hassan, M.R.; Alelaiwi, A.; Fortino, G. A hybrid feature extraction method with regularized extreme learning machine for brain tumor classification. IEEE Access 2019, 7, 36266–36273. [Google Scholar] [CrossRef]
- Masood, M.; Nazir, T.; Nawaz, M.; Mehmood, A.; Rashid, J.; Kwon, H.-Y.; Mahmood, T.; Hussain, A. A Novel Deep Learning Method for Recognition and Classification of Brain Tumors from MRI Images. Diagnostics 2021, 11, 744. [Google Scholar] [CrossRef] [PubMed]
- Swati, Z.N.K.; Zhao, Q.; Kabir, M.; Ali, F.; Ali, Z.; Ahmed, S.; Lu, J. Brain tumor classification for MR images using transfer learning and fine-tuning. Comput. Med. Imaging Graph. 2019, 75, 34–46. [Google Scholar] [CrossRef] [PubMed]
- Rehman, A.; Naz, S.; Razzak, M.I.; Akram, F.; Imran, M. A deep learning-based framework for automatic brain tumors classification using transfer learning. Circuits Syst. Signal Process. 2020, 39, 757–775. [Google Scholar] [CrossRef]
- Kutlu, H.; Avcı, E. A novel method for classifying liver and brain tumors using convolutional neural networks, discrete wavelet transform and long short-term memory networks. Sensors 2019, 19, 1992. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Kaplan, K.; Kaya, Y.; Kuncan, M.; Ertunç, H.M. Brain tumor classification using modified local binary patterns (LBP) feature extraction methods. Med. Hypotheses 2020, 139, 109696. [Google Scholar] [CrossRef] [PubMed]
- Russakovsky, O.; Deng, J.; Su, H.; Krause, J.; Satheesh, S.; Ma, S.; Huang, Z.; Karpathy, A.; Khosla, A.; Bernstein, M. Imagenet large scale visual recognition challenge. Int. J. Comput. Vis. 2015, 115, 211–252. [Google Scholar] [CrossRef] [Green Version]
- Moccia, S.; Foti, S.; Routray, A.; Prudente, F.; Perin, A.; Sekula, R.F.; Mattos, L.S.; Balzer, J.R.; Fellows-Mayle, W.; De Momi, E. Toward improving safety in neurosurgery with an active handheld instrument. Ann. Biomed. Eng. 2018, 46, 1450–1464. [Google Scholar] [CrossRef] [PubMed]
- Ye, Q.; Xia, J.; Yang, G. Explainable AI for COVID-19 CT classifiers: An initial comparison study. In Proceedings of the 2021 IEEE 34th International Symposium on Computer-Based Medical Systems (CBMS), Online, 7–9 June 2021; pp. 521–526. [Google Scholar]
- Yang, G.; Ye, Q.; Xia, J. Unbox the black-box for the medical explainable ai via multi-modal and multi-centre data fusion: A mini-review, two showcases and beyond. Inf. Fusion 2022, 77, 29–52. [Google Scholar] [CrossRef]
Tumor Class | Patients | MRI Scans | Planer Geometry Details | |
---|---|---|---|---|
Anatomical Plane | Scans Division | |||
Glioma | 89 | 1426 | Transverse | 494 |
Coronal | 437 | |||
Sagittal | 495 | |||
Meningioma | 82 | 708 | Transverse | 209 |
Coronal | 268 | |||
Sagittal | 231 | |||
Pituitary | 62 | 930 | Transverse | 291 |
Coronal | 319 | |||
Sagittal | 320 |
Type | Variation | Extreme Value |
---|---|---|
Rotation | Clockwise | −3 |
Anti-clockwise | 3 | |
Reflection | X-Reflection | 1 |
Y-Reflection | 1 | |
Shear | X-Shear | [−0.05, 0.05] |
Y-Shear | [−0.05, 0.05] |
Parameter | Methods/Values Under Trail | Selection |
---|---|---|
Initial Learning Rate | Single (0.01, 0.001, 0.002, 0.005, 0.008, 0.009, 0.0002, 0.0001, 0.0003, 0.0004, 0.0005, 0.0007, 0.0008, 0.0009), Piecewise (Learn rate drop factor: 0.1, 0.2, 0.3, 0.4, 0.5; Learn rate schedule: 10, 20, 25 epochs) | Single: 0.0001 |
Epochs (Maximum) | 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 60, 75, 100 | 20 |
Mini Batch (Size) | 2, 4, 8, 16, 32 | 16 |
Pairs of Conv. Layersand Filter Stacks | 2, 3, 4, 5 | 4 |
Kernel Depth (Size of Each Filter Stack) | 4, 8, 16, 32, 64, 128, 256 | 64, 128, 256 |
Kernel Size | 3 × 3, 5 × 5, and 7 × 7 | 3 × 3 |
ReLU | 2, 3, 4, 5 | 4 |
Pooling Type | MaxPool, AvgPool | MaxPool (2 × 2) |
Fully Connected Layers | 1, 2, 3, 4 | 2 |
Number of Filter Stacks | 2, 3, 4, 5 | 4 |
L2-Regularization | 0.01, 0.05, 0.001, 0.005, 0.008, 0.0009, 0.0001, 0.0002, 0.0005, 0.0006, 0.0007, 0.0008, 0.0009 | 0.0005 |
Solver Name | Stochastic Gradient Descent with Momentum (SGDM), ADAptive Moment estimation (ADAM) | SGDM |
Momentum (for Sgdm only) | 0.80, 0.82, 0.82, 0.85,0.87, 0.90, 0.91, 0.92, 0.93, 0.94, 0.95, 0.96, 0.97, 0.98, 0.99, 1.00 | 0.97 |
Input Image Size | 64 × 64, 128 × 128, 224 × 224, 256 × 256 | 224 × 224 |
Dropout Layers | 1, 2, 3 | 1 |
Dropout Rate (%) | 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 35, 45, 50, 60 | 5 |
Classifier | Param | Tumor Type | TP | FN | FP | TPR (Ind.) | PPV (Ind.) | F-Measure (Ind.) | Acc. (Ind.) | Acc. (Avg.) |
---|---|---|---|---|---|---|---|---|---|---|
SVM kernel | Linear | G | 276 | 9 | 6 | 96.84 | 97.87 | 0.97 | 97.55 | |
M | 135 | 7 | 8 | 95.07 | 94.41 | 0.95 | 97.55 | 98.26 | ||
P | 186 | 0 | 2 | 100.00 | 98.94 | 1.00 | 99.67 | |||
RBF | G | 278 | 7 | 5 | 97.54 | 98.23 | 0.98 | 98.04 | ||
M | 136 | 6 | 7 | 95.78 | 95.11 | 0.95 | 97.88 | 98.58 | ||
P | 186 | 0 | 1 | 100.00 | 99.47 | 1.00 | 99.83 | |||
Ploy-2 | G | 281 | 4 | 9 | 98.60 | 96.90 | 0.98 | 97.88 | ||
M | 133 | 9 | 4 | 93.66 | 97.08 | 0.95 | 97.88 | 98.47 | ||
P | 185 | 1 | 1 | 99.46 | 99.46 | 1.00 | 99.67 | |||
Ploy-3 | G | 283 | 2 | 3 | 99.30 | 98.95 | 0.99 | 99.18 | ||
M | 137 | 5 | 2 | 96.48 | 98.56 | 0.98 | 98.86 | 99.24 | ||
P | 186 | 0 | 2 | 100.00 | 98.94 | 0.99 | 99.67 | |||
Ploy-4 | G | 183 | 102 | 156 | 64.21 | 53.98 | 0.59 | 50.39 | 52.89 | |
M | 61 | 81 | 148 | 42.96 | 29.19 | 0.35 | 53.36 | |||
P | 18 | 168 | 47 | 9.68 | 27.69 | 0.14 | 54.93 | |||
k-NN k | 1 | G | 281 | 4 | 13 | 98.60 | 95.58 | 0.97 | 97.23 | |
M | 129 | 13 | 3 | 90.85 | 97.73 | 0.94 | 97.39 | 98.15 | ||
P | 186 | 0 | 1 | 100.00 | 99.47 | 1.00 | 99.83 | |||
3 | G | 280 | 5 | 16 | 98.25 | 94.60 | 0.96 | 96.56 | ||
M | 125 | 17 | 6 | 88.03 | 95.42 | 0.92 | 96.25 | 97.42 | ||
P | 185 | 1 | 1 | 99.46 | 99.46 | 1.00 | 99.66 | |||
5 | G | 278 | 7 | 26 | 97.54 | 91.45 | 0.94 | 94.55 | ||
M | 112 | 30 | 9 | 78.87 | 92.56 | 0.85 | 93.62 | 95.48 | ||
P | 182 | 4 | 6 | 97.85 | 96.81 | 0.97 | 98.28 | |||
7 | G | 276 | 9 | 28 | 96.84 | 90.79 | 0.94 | 93.85 | ||
M | 108 | 34 | 11 | 76.06 | 90.76 | 0.83 | 92.62 | 94.69 | ||
P | 181 | 5 | 9 | 97.31 | 95.26 | 0.96 | 97.58 | |||
9 | G | 274 | 11 | 39 | 96.14 | 87.54 | 0.92 | 91.68 | ||
M | 95 | 47 | 10 | 66.90 | 90.48 | 0.77 | 90.63 | 93.10 | ||
P | 182 | 4 | 13 | 97.85 | 93.33 | 0.96 | 97.01 | |||
RF Nt | 500 | G | 274 | 11 | 6 | 96.14 | 97.86 | 0.97 | 97.21 | |
M | 136 | 6 | 12 | 95.78 | 91.89 | 0.94 | 97.05 | 97.81 | ||
P | 183 | 3 | 2 | 98.39 | 98.92 | 0.99 | 99.16 | |||
550 | G | 275 | 10 | 7 | 96.49 | 97.52 | 0.97 | 97.22 | ||
M | 136 | 6 | 10 | 95.78 | 93.15 | 0.94 | 97.38 | 97.92 | ||
P | 183 | 3 | 2 | 98.39 | 98.92 | 0.99 | 99.17 | |||
600 | G | 276 | 9 | 6 | 96.84 | 97.87 | 0.97 | 97.55 | ||
M | 136 | 6 | 9 | 95.78 | 93.79 | 0.95 | 97.55 | 98.14 | ||
P | 184 | 2 | 2 | 98.93 | 98.93 | 0.99 | 99.33 | |||
650 | G | 274 | 11 | 7 | 96.14 | 97.51 | 0.97 | 97.05 | ||
M | 136 | 6 | 9 | 95.78 | 93.79 | 0.95 | 97.53 | 97.81 | ||
P | 183 | 3 | 4 | 98.39 | 97.86 | 0.98 | 98.83 | |||
1000 | G | 275 | 10 | 7 | 96.49 | 97.52 | 0.97 | 97.21 | ||
M | 134 | 8 | 9 | 94.37 | 93.71 | 0.94 | 97.21 | 97.81 | ||
P | 184 | 2 | 4 | 98.93 | 97.87 | 0.98 | 99.00 |
Dataset Type | Tumor Type | TP | FN | FP | TPR | PPV | F-Measure | Acc. (Ind.) | Acc. (Avg.) |
---|---|---|---|---|---|---|---|---|---|
A-CE-MRI | Glioma | 551 | 19 | 32 | 96.67 | 94.51 | 0.96 | 95.82 | |
Meningioma | 246 | 37 | 19 | 86.93 | 92.83 | 0.90 | 95.43 | 96.88 | |
Pituitary | 371 | 1 | 6 | 99.73 | 98.41 | 0.99 | 99.40 | ||
WT-CE-MRI | Glioma | 276 | 9 | 17 | 96.84 | 94.20 | 0.96 | 95.70 | |
Meningioma | 118 | 24 | 10 | 83.10 | 92.19 | 0.87 | 94.44 | 96.15 | |
Pituitary | 184 | 2 | 8 | 98.93 | 95.83 | 0.97 | 98.30 | ||
CE-MRI | Glioma | 272 | 13 | 19 | 95.44 | 93.47 | 0.94 | 94.71 | |
Meningioma | 117 | 25 | 12 | 82.39 | 90.70 | 0.86 | 93.93 | 95.59 | |
Pituitary | 184 | 2 | 9 | 98.93 | 95.34 | 0.97 | 98.12 |
Dataset Type | Tumor Type | TP | FN | FP | TPR | PPV | F-Measure | Acc. (Ind.) | Acc. (Avg.) |
---|---|---|---|---|---|---|---|---|---|
A-CE-MRI | Glioma | 233 | 52 | 258 | 81.75 | 80.90 | 0.81 | 82.11 | 86.26 |
Meningioma | 85 | 57 | 406 | 59.86 | 59.86 | 0.60 | 81.16 | ||
Pituitary | 173 | 13 | 318 | 93.01 | 94.54 | 0.94 | 95.53 | ||
WT-CE-MRI | Glioma | 245 | 40 | 223 | 85.96 | 73.35 | 0.79 | 78.39 | 83.36 |
Meningioma | 59 | 83 | 409 | 41.55 | 59.60 | 0.49 | 79.19 | ||
Pituitary | 164 | 22 | 304 | 88.17 | 91.11 | 0.90 | 92.49 | ||
CE-MRI | Glioma | 197 | 88 | 264 | 69.12 | 84.55 | 0.76 | 78.80 | 82.40 |
Meningioma | 107 | 35 | 354 | 75.35 | 51.94 | 0.61 | 77.48 | ||
Pituitary | 157 | 29 | 304 | 84.41 | 90.23 | 0.87 | 90.93 |
Input Layer | Tumor Type | TP | FN | FP | TPR (Ind.) | PPV (Ind.) | F-Measure (Ind.) | Acc. (Ind.) | Acc. (Avg.) |
---|---|---|---|---|---|---|---|---|---|
64 × 64 | Glioma | 277 | 8 | 14 | 97.19 | 95.19 | 0.96 | 96.38 | |
Meningioma | 125 | 17 | 9 | 88.03 | 93.28 | 0.91 | 95.75 | 97.04 | |
Pituitary | 184 | 2 | 4 | 98.93 | 97.87 | 0.98 | 98.99 | ||
128 × 128 | Glioma | 282 | 3 | 10 | 98.95 | 96.58 | 0.98 | 97.86 | |
Meningioma | 127 | 15 | 4 | 89.44 | 96.95 | 0.93 | 96.90 | 97.92 | |
Pituitary | 185 | 1 | 5 | 99.46 | 97.37 | 0.98 | 99.00 | ||
224 × 224 | Glioma | 283 | 2 | 3 | 99.30 | 98.95 | 0.99 | 99.18 | |
Meningioma | 137 | 5 | 2 | 96.48 | 98.56 | 0.98 | 98.86 | 99.24 | |
Pituitary | 186 | 0 | 2 | 100.00 | 98.94 | 0.99 | 99.67 | ||
256 × 256 | Glioma | 278 | 7 | 13 | 97.54 | 95.53 | 0.97 | 96.73 | |
Meningioma | 127 | 15 | 7 | 89.44 | 94.78 | 0.92 | 96.41 | 97.60 | |
Pituitary | 186 | 0 | 2 | 100.00 | 98.94 | 1.00 | 99.66 |
Multi-Class Scheme | Tumor Type | TP | FN | FP | TPR (Ind.) | PPV (Ind.) | F-Measure (Ind.) | Acc. (Ind.) | Acc. (Avg.) |
---|---|---|---|---|---|---|---|---|---|
OvA | Glioma | 277 | 8 | 7 | 97.19 | 97.54 | 0.97 | 97.55 | |
Meningioma | 134 | 8 | 8 | 94.37 | 94.37 | 0.94 | 97.39 | 98.26 | |
Pituitary | 186 | 0 | 1 | 100.00 | 99.47 | 1.00 | 99.83 | ||
OvO | G-M | 279 | 6 | 13 | 97.89 | 95.55 | 96.71 | 95.55 | |
M-P | 142 | 0 | 0 | 100.00 | 100.00 | 100.00 | 100.00 | 98.45 | |
G-P | 285 | 0 | 1 | 100.00 | 99.65 | 99.82 | 99.79 |
Reference | Accuracy |
---|---|
Diaz-Pernas et al. [32] | 97.30 |
Badža and Barjaktarović [25] | 96.56 |
Proposed UL-BTD | 98.46 (av.) |
Comparison Basis | UL-DLA | SVM (HFS) |
---|---|---|
Tumor detection time/image (ms) | 7.96 | 11.69 |
Memory usage (GPU, GB) | 2.1 | 2.8 |
Technique | Translation Source | Utility | Drawbacks | Compatible with Neuro Navigation System |
---|---|---|---|---|
Intraoperative Fluorescence Imaging (iFI) | Optical fluorescence imaging is based on the accumulation of fluorescence optical contrast media at the desired ROI | In vivo assistance during surgery [33] | The limited resolution, quantification, depth of penetration 5 mm, and availability of targeted contrast agents with a high signal-to-back-ground ratio (SBR) [34] | √ |
Intraoperative Ultrasound Imaging (IoUS) | IoUS uses high-frequency sound waves as the viewing source to form the image | Integrated to the neuro-navigation system with accuracy 1.40 ± 0.45 mm (arithmetic mean) [35] | Image quality subject to interoperation variability motion artifacts, and low image resolution restricts it for deep tumors [36] | √ |
Intraoperative Raman Spectroscopy (iRS) | RS is based on light interaction within a material between its chemical bonds | Intraoperative label-free molecular information [5] with an accuracy of 1 mm [37] | WHO says that its accuracy, sensitivity, and specificity are lower than 60% for grade invasive cancer cells in the normal brain or between grade 3 and 4 gliomas | √ |
Hyperspectral Imaging (HSI) | HSI is finding the spectrum for each pixel in the image of a scene [38] | Image-guided surgery using an intraoperative visualization system for delineation of the brain tumor [39] | Penetration of depth, as well as real-time detection of tumors, is a challenge | √ |
Intraoperative Magnetic Resonance Imaging (iMRI) | MRI is based on protons’ density distribution in the brain | It controls brain shift during resection of non-fluorescing gliomas (~1 h) [4,40] | Low magnetic field strength, contrast agents have their own problems and need more research for FDA approved and optimized contrast agents | √ |
Optical Coherence Tomography (OCT) | OCT is based on the detection of light backscattered by a tissue [41] | It provides micrometer-scale resolution with quick volumetric imaging [42] | Using ionizing radiation much sensitive to the brain | √ |
UL-BTD Framework | Deep learning-based solution using artificial intelligence (AI) and iMRI scans | Real-time brain tumor surgical resection support, can be used with desktop | Only need artificial intelligence (AI) and iMRI scans, no drawback | √ |
Model/Study Reference | k-Fold Cross-Validation/Dataset Partitioning | Augmentation/Smote | Classification Technique | Precision (%Avg.) | Recall (%Avg.) | F-Score (%Avg.) | Accuracy (%Avg.) |
---|---|---|---|---|---|---|---|
[9] | 5-fold | √ | SVM + BoW | × | × | × | 91.14 |
[44] | 5-fold; 10-fold | √ | CNN | × | × | × | 91.43 |
[45] | × | × | Capsule Networks (CapsNet) | × | × | × | 89.56 |
[46] | 5-fold | × | CapsNet | 92.67 | 94.67 | 93.33 | 92.60 |
[6] | 68% training and 32% in test | √ | CNN | 96.06 | 94.43 | × | 96.13 |
[43] | 5-fold | √ | GA based CNN | × | × | × | 94.20 |
[47] | 5-fold; 70% training and 30% in test | × | Hybrid PCA-NGIST + RELM | × | × | × | 94.23 |
[52] | 10-fold | × | nLBP + k-NN | × | × | × | 95.56 |
[25] | 5-fold; 10-fold | √ | CNN | 95.79 | 96.51 | 96.11 | 96.56 |
[32] | 5-fold; 10-fold | √ | Multi-scale CNN | × | 96.67 | × | 97.30 |
Proposed framework | 10-fold; 5 fold for limited experiments | √ | UL-DLA + GLCM + SVM (UL-BTD system) | 97.07 | 97.30 | 97.20 | 99.24 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Qureshi, S.A.; Raza, S.E.A.; Hussain, L.; Malibari, A.A.; Nour, M.K.; Rehman, A.u.; Al-Wesabi, F.N.; Hilal, A.M. Intelligent Ultra-Light Deep Learning Model for Multi-Class Brain Tumor Detection. Appl. Sci. 2022, 12, 3715. https://doi.org/10.3390/app12083715
Qureshi SA, Raza SEA, Hussain L, Malibari AA, Nour MK, Rehman Au, Al-Wesabi FN, Hilal AM. Intelligent Ultra-Light Deep Learning Model for Multi-Class Brain Tumor Detection. Applied Sciences. 2022; 12(8):3715. https://doi.org/10.3390/app12083715
Chicago/Turabian StyleQureshi, Shahzad Ahmad, Shan E. Ahmed Raza, Lal Hussain, Areej A. Malibari, Mohamed K. Nour, Aziz ul Rehman, Fahd N. Al-Wesabi, and Anwer Mustafa Hilal. 2022. "Intelligent Ultra-Light Deep Learning Model for Multi-Class Brain Tumor Detection" Applied Sciences 12, no. 8: 3715. https://doi.org/10.3390/app12083715
APA StyleQureshi, S. A., Raza, S. E. A., Hussain, L., Malibari, A. A., Nour, M. K., Rehman, A. u., Al-Wesabi, F. N., & Hilal, A. M. (2022). Intelligent Ultra-Light Deep Learning Model for Multi-Class Brain Tumor Detection. Applied Sciences, 12(8), 3715. https://doi.org/10.3390/app12083715