Open Access Open Access  Restricted Access Subscription or Fee Access

A Comprehensive Comparative Analysis of Machine Learning Models for Brain Tumor Segmentation

(*) Corresponding author

Authors' affiliations



This research proposes a new approach to finding the best segmentation models for BraTS2020. Finding an efficient way of segmenting brain tumors from MRIs is essential, as minor improvements in the field may lead to lives being saved. The proposed approach compares 32 different segmentation models based on the U-net architecture on each available label. The models are trained in pre-processed T2-weighted, enhanced T1-weighted, and Fluid-Attenuated Inversion Recovery (FLAIR) Magnetic Resonance Images (MRI) from the BraTS2020 dataset. Rows and columns consisting of only black pixels are removed, and the images are then resized and stacked to fit the model. The models achieving the highest IoU score are SeResNet152 for enhancing tumors, SeResNet152 for peritumoral edemas, and EfficientNetB3 for necrotic tissue/tumor cores, ending at IoU scores of 0.85, 0.93, and 0.82, respectively. The IoU score was boosted further by combining the predicted outputs into one mask, which predicts the whole tumor. Combining the models’ output resulted in an IoU score of 0.94 for the whole tumor on the testing data. The results of this research can ease the workload of radiologists and potentially be used in developing a fully automated process for analyzing and diagnosing brain MRIs.
Copyright © 2023 Praise Worthy Prize - All rights reserved.


Brain Tumor; Machine Learning; Segmentation

Full Text:



Abbood, A. A., Shallal, Q. M., and Fadhel, M. A. (2021). Automated brain tumor classification using various deep learning models: A comparative study, In: Indonesian Journal of Electrical Engineering and Computer Science 22.1, p. 8.

Amin, J., Sharif, M., Raza, M., et al. (2019). Brain tumor detection using statistical and machine learning method. In: Computer Methods and Programs in Biomedicine 177, pp. 69-79.

Amin, J., Anjum, M. A., et al. (2022). A New Model for Brain Tumor Detection Using Ensemble Transfer Learning and Quantum Variational Classifier. In: Computational Intelligence and Neuroscience 2022.

Amin, J., Sharif, M., Yasmin, M., et al. (2018). Big data analysis for brain tumor detection: Deep convolutional neural networks. In: Future Generation Computer Systems 87, p. 8.

Bakas, S., Akbari, H., Sotiras, A., Bilello, M., Rozycki, M., Kirby, J., et al. (Nov. 2022a). BraTS-TCGA- GBM. (Accessed on 12/11/2022).

BraTS-TCGA-LGG. (Accessed on 12/11/2022). action?pageId=24282668#2428266828b503a73057431ba7736a00d8a2bc99

Bakas, S., Akbari, H., Sotiras, A., Bilello, M., Rozycki, M., Kirby, J. S., et al. (2017). Advancing The Cancer Genome Atlas glioma MRI collections with expert segmentation labels and radiomic features. In: Scientific Data 4 March, p. 13.

Bakas, S., Reyes, M., et al. (2018). Identifying the Best Machine Learning Algorithms for Brain Tumor Segmentation, Progression Assessment, and Overall Survival Prediction in the BRATS Challenge.

Brain Tumor: Statistics - Cancer.Net (n.d.). (Accessed on 08/27/2022).
https:// www. cancer. net/ cancer- types/ brain- tumor/statistics

BraTS2020 Dataset (Training + Validation) - Kaggle (2020). (Accessed on 01/11/2022).

Cinar, N., Ozcan, A., and Kaya, M. (2022). A hybrid DenseNet121-UNet model for brain tumor seg- mentation from MR Images. In: Biomedical Signal Processing and Control 76.March, p. 27.

Erickson, B. J. et al. (2017). Machine learning for medical imaging. In: Radiographics 37.2, pp. 505-515.

Hale, A. T. et al. (2018). Machine learning analyses can differentiate meningioma grade by features on magnetic resonance imaging. In: Neurosurgical Focus 45.5, pp. 1-6.

Iakubovskii, P. et al. (July 2022). GitHub - qubvel/segmentation models: Segmentation models with pretrained backbones. Keras and TensorFlow Keras. (Accessed on 12/03/2023) models

Iqbal, M. J. et al. (2021). Clinical applications of artificial intelligence and machine learning in cancer diagnosis: looking into the future. In: Cancer Cell International 21.1, pp. 1-11.

Kader, I. A. E. et al. (2021). Differential deep convolutional neural network model for brain tumor classification. In: Brain Sciences 11.3, p. 16.

Lee, J.-g. et al. (2017). 2017-Lee-Deep Learning in Medical Imaging Gen. In: Korean Journal of Radiology 18.4, p. 15.

Linhares, P. et al. (2020). Glioblastoma: Is there any blood biomarker with true clinical relevance? In: International Journal of Molecular Sciences 21.16, pp. 1-16.

Lu, F. and Lysack, J. T. (2022). Lessons Learned From Commonly Missed Head and Neck Cancers on Cross-Sectional Imaging. In: Canadian Association of Radiologists Journal, pp. 2-4.

Mahbub, M. K. et al. (2022). Deep Neural Networks for Brain Tumor Detection from MRI Images. In: Lecture Notes in Networks and Systems 348.February, pp. 473-485.

Maji, D., Sigedar, P., and Singh, M. (2022). Attention Res-UNet with Guided Decoder for semantic segmentation of brain tumors. In: Biomedical Signal Processing and Control 71.PA, p. 9.

Menze, B. H. et al. (2015). The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS). In: IEEE Transactions on Medical Imaging 34.10, p. 32.

Naser, M. A. and Deen, M. J. (2020). Brain tumor segmentation and grading of lower-grade glioma using deep learning in MRI images. In: Computers in Biology and Medicine 121.April, p. 103758.

Nickparvar, M. (2021). Brain Tumor MRI Dataset.

Noreen, N. et al. (2020). A Deep Learning Model Based on Concatenation Approach for the Diagnosis of Brain Tumor. In: IEEE Access 8, pp. 55135-55144.

Rajeev, S. K. et al. (2022). Biomedical Signal Processing and Control A biologically-inspired hybrid deep learning approach for brain tumor classification from magnetic resonance imaging using improved gabor wavelet transform and Elmann-BiLSTM network. In: Biomedical Signal Processing and Control 78.April, p. 103949. ISSN: 1746-8094.

Raza, A. et al. (2022). A Hybrid Deep Learning-Based Approach for Brain Tumor Classification. In: Electronics (Switzerland) 11.7, p. 17.

Rehman, A. et al. (2020). A Deep Learning-Based Framework for Automatic Brain Tumors Classification Using Transfer Learning. In: Circuits, Systems, and Signal Processing 39.2, p. 19.

Shafi, A. S. et al. (2021). Classification of brain tumors and auto-immune disease using ensemble learning. In: Informatics in Medicine Unlocked 24.May, p. 100608.

Sharif, M. et al. (2018). Brain tumor segmentation and classification by improved binomial thresholding and multi-features selection. In: Journal of Ambient Intelligence and Humanized Computing 0.0, p. 11.

Sharif, M. I. et al. (2020). A comprehensive review on multi-organs tumor detection based on ma- chine learning. In: Pattern Recognition Letters 131, pp. 30-37.

Shinde, A. S. et al. (2022). Advances in Engineering Software Performance analysis of machine learning algorithm of detection and classification of brain tumor using computer vision. In: Ad- vances in Engineering Software 173, p. 103221.

Shuvo, M. B. et al. (2021). CNL-UNet: A novel lightweight deep learning architecture for multimodal biomedical image segmentation with false output suppression. In: Biomedical Signal Processing and Control 70.April, p. 14.

Swati, Z. N. K. et al. (2019). Brain tumor classification for MR images using transfer learning and fine-tuning. In: Computerized Medical Imaging and Graphics 75, p. 13.

Tandel, G. S., Tiwari, A., and Kakde, O. G. (2021). Performance optimisation of deep learning models using majority voting algorithm for brain tumor classification. In: Computers in Biology and Medicine 135.February, p. 27.

Uberg, L. L. (2022). Analysis of Brain Tumor Using MRI Images. In: Proceedings of the 17th Conference on Computer Science and Intelligence Systems, M. Ganzha, L. Maciaszek, M. Paprzycki, D. Ślęzak (eds). ACSIS, Vol. 30, pages 201-204 (2022), p. 4.

Xiang, Y. et al. (2020). Implementation of artificial intelligence in medicine: Status analysis and development suggestions. In: Artificial Intelligence in Medicine 102.December 2019, p. 7.

Yue, W. et al. (2022). Deep learning-based automatic segmentation for size and volumetric measurement of breast cancer on magnetic resonance imaging. In: Frontiers in Oncology 12.August, pp. 1-14.


  • There are currently no refbacks.

Please send any question about this web site to
Copyright © 2005-2024 Praise Worthy Prize