Multi-task model for classification and segmentation of brain tumors
DOI:
https://doi.org/10.59681/2175-4411.v16.iEspecial.2024.1296Keywords:
Artificial Intelligence, Supervised Machine Learning, Brain Neoplasms, Deep LearningAbstract
Objective: To validate whether a multi-task model (MTL) for brain tumor classification and segmentation outperforms a single-task (ST) approach. Method: The model architecture consists of an encoder, branching into a fully connected (classification) and a decoder (segmentation). For the ST, only the classification branch was considered. Both were trained using the 5-fold cross-validation approach with the SARTAJ and Figshare datasets. Results: The MTL achieved an accuracy of 95.99% ± 0.70% compared to 95.87% ± 1.01% for the ST. Conclusion: Both models presented similar performances, however the MTL revealed some advantages, such as greater stability of metrics, resulting from the lower standard deviation in all metrics. Compared to the literature, the MTL achieved an accuracy only 3% below the best model analyzed and also had a significantly lower number of parameters, up to 187 times fewer.
References
Thierheimer M, Cioffi G, Waite KA, Kruchko C, Ostrom QT, Barnholtz-Sloan JS. Mortality trends in primary malignant brain and central nervous system tumors vary by histopathology, age, race, and sex. J Neurooncol. 2023;162:167-177.
Deorah S, Lynch CF, Sibenaller ZA, Ryken TC. Trends in Brain Cancer Incidence and Survival in the United States: Surveillance, Epidemiology, and End Results Program, 1973 to 2001. Neurosurg Focus. 2006;20(4):E1.
Yang S, Zhu F, Ling X, Liu Q, Zhao P. Intelligent Health Care: Applications of Deep Learning in Computational Medicine. Front Genet. 2021;12:607471.
Trombetta GBW, Fröhlich W da R, Rigo SJ, Rodrigues CA. Application of Deep Learning for Diagnosis of COVID-19-Induced Pneumonia from X-ray Images. J Health Inform [Internet]. March 15, 2021 [cited March 9, 2024];12. Available from: https://jhi.sbis.org.br/index.php/jhi-sbis/article/view/828.
LeCun Y, Bengio Y, Hinton G. Deep Learning. Nature. 2015;521:436.
Zhang Y, Yang Q. An Overview of Multi-Task Learning. Natl Sci Rev. 2018;5(1):30-43.
Crawshaw M. Multi-Task Learning with Deep Neural Networks: A Survey. arXiv. 2020.
Ruder S. An Overview of Multi-Task Learning in Deep Neural Networks. arXiv. 2017.
Tardy M, Mateus D. Leveraging Multi-Task Learning to Cope With Poor and Missing Labels of Mammograms. Front Radiol. 2021;1.
Oliveira B, et al. A multi-task convolutional neural network for classification and segmentation of chronic venous disorders. Sci Rep. 2023;13:761.
Ngo DK, Tran MT, Kim SH, Yang HJ, Lee GS. Multi-Task Learning for Small Brain Tumor Segmentation from MRI. Appl Sci. 2020;10(21):7790.
Gómez-Guzmán MA, et al. Classifying Brain Tumors on Magnetic Resonance Imaging by Using Convolutional Neural Networks. Electronics. 2023;12:955.
Ullah N, et al. An Effective Approach to Detect and Identify Brain Tumors Using Transfer Learning. Appl Sci. 2022;12:5645.
Rasheed Z, et al. Brain Tumor Classification from MRI Using Image Enhancement and Convolutional Neural Network Techniques. Brain Sci. 2023;13(9):1320.
Bhuvaji S, Kadam A, Bhumkar P, Dedge S, Kanchan S. Brain Tumor Classification (MRI). Kaggle. 2020.
Cheng J. Brain Tumor Dataset. Figshare. 2017.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Submission of a paper to Journal of Health Informatics is understood to imply that it is not being considered for publication elsewhere and that the author(s) permission to publish his/her (their) article(s) in this Journal implies the exclusive authorization of the publishers to deal with all issues concerning the copyright therein. Upon the submission of an article, authors will be asked to sign a Copyright Notice. Acceptance of the agreement will ensure the widest possible dissemination of information. An e-mail will be sent to the corresponding author confirming receipt of the manuscript and acceptance of the agreement.
