U-Net Based Approach for Automated Brain Tumor Classification

Notice

This is an unedited manuscript accepted for publication and provided as an Article in Press for early access at the author’s request. The article will undergo copyediting, typesetting, and galley proof review before final publication. Please be aware that errors may be identified during production that could affect the content. All legal disclaimers of the journal apply.

Year : 2025 | Volume : 14 | Issue : 02 | Page : –
    By

    Anjali Yadav,

  • Hardik Tated,

  • Suyog Tathe,

  • Neha Vishwakarma,

  • Pratiksha Madre,

  1. Professor, Department of electronic and telecommunication, Sinhgad Institute of Technology and science, Narhe, Pune, India
  2. Student, Department of electronic and telecommunication, Sinhgad Institute of Technology and science, Narhe, Pune, India
  3. Student, Department of electronic and telecommunication, Sinhgad Institute of Technology and science, Narhe, Pune, India
  4. Student, Department of electronic and telecommunication, Sinhgad Institute of Technology and science, Narhe, Pune, India
  5. Student, Department of electronic and telecommunication, Sinhgad Institute of Technology and science, Narhe, Pune, India

Abstract

document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_abs_192332’);});Edit Abstract & Keyword

Brain tumor detection and identification play vital roles in diagnostic procedures in the field of medicine, with the conventional analysis of MRI images requiring a lot of time and also subject to variability. The proposed study involves the use of a CNN-U-Net based approach for brain tumor detection and identification automatically. The study uses a database of 3,064 contrast-enhanced T1-weighted MRI images from 233 patients with the tumors of meningioma, glioma, and pituitary tumors. The proposed methodology incorporates advanced preprocessing techniques, including image denoising, normalization, and tumor mask extraction, to improve image quality. The CNN-based U-Net model is trained on 80% of the dataset, employing batch normalization, dropout, and early stopping mechanisms to ensure generalization and prevent overfitting. Rotation, flipping, and zooming methods for data augmentation are utilized to improve model robustness. Accuracy, precision, recall, F1-score, and confusion matrix, are utilized to measure model performance with a test accuracy of 91.5%. To allow real-time usage within the clinic, the built model has been integrated within a web application system of automatic detection and categorization of tumors. Through the use of deep learning within neuro-oncology, this study attempts to advance diagnostic accuracy, reduce the necessity of human interpretation, and simplify the detection of brain tumors with the final patient benefit.

Keywords: Artificial intelligence in healthcare, CNN-based U-Net, MRI image processing, Brain Tumor segmentation.

[This article belongs to Research and Reviews: Journal of Oncology and Hematology ]

How to cite this article:
Anjali Yadav, Hardik Tated, Suyog Tathe, Neha Vishwakarma, Pratiksha Madre. U-Net Based Approach for Automated Brain Tumor Classification. Research and Reviews: Journal of Oncology and Hematology. 2025; 14(02):-.
How to cite this URL:
Anjali Yadav, Hardik Tated, Suyog Tathe, Neha Vishwakarma, Pratiksha Madre. U-Net Based Approach for Automated Brain Tumor Classification. Research and Reviews: Journal of Oncology and Hematology. 2025; 14(02):-. Available from: https://journals.stmjournals.com/rrjooh/article=2025/view=0


document.addEventListener(‘DOMContentLoaded’,function(){frmFrontForm.scrollToID(‘frm_container_ref_192332’);});Edit

References

  1. Agrawal M, Singh D, Chauhan H. Transfer learning-based brain tumor classification using MRI images. Neural Comput Appl. 2024;36(2):1235–1246.
  2. Chattopadhyay A, Maitra M. A convolutional neural network model for brain tumor classification using batch normalization and fine-tuning. Biomed Signal Process Control. 2022;71:103135.
  3. Saleh A, Albarqi HA, Mohammed MA, García-Zapirain B, Damaševičius R, Maskeliūnas R, et al. Brain tumor classification using CNN architectures: Xception, ResNet50, and Inception V3. J Supercomput. 2020;76(9):7333–7345.
  4. Hashemzehi R, Shoaran M, Daei A. A hybrid deep learning model based on CNN and NADE for brain tumor classification. Multimedia Tools Appl. 2019;78(22):31807–31824.
  5. Hua R, Yu J, Zhang Y, Chen J, Wu J. Brain tumor segmentation in multimodal MRI using cascaded V-Net. Neurocomputing. 2020;392:1–10.
  6. Bhimavarapu U, Kanakala S, Krishna G, Varma PK. Unsupervised clustering and machine learning-based classification of glioma using fuzzy c-means on MRI data. J King Saud Univ Comput Inf Sci. 2024;36(1):52–61.
  7. Amin J, Sharif M, Anwar SM. Machine learning techniques for brain tumor classification: A review. J Imaging. 2021;7(5):77.
  8. Kumar S, Choudhary A, Singh M. Brain tumor classification using convolutional neural network. In: 2021 4th International Conference on Innovations in Computing Technologies (ICICT). IEEE; 2021. p. 1–6. doi:10.1109/ICICT51256.2021.9358544
  9. Amin J, Sharif M, Yasmin M, Fernandes SL. A hybrid deep learning framework for brain tumor classification. J Med Imaging Health Inform. 2021;11(6):1601–1607. doi:10.1166/jmihi.2021.3587
  10. Chen D, Shi L, Li X. Deep CNN models for brain tumor detection using MRI images with enhanced contrast. Int J Imaging Syst Technol. 2021;31(2):755–764. doi:10.1002/ima.22543
  11. Das M, Jani HS, Mohan PM. Brain tumor detection and segmentation with deep learning techniques. Comput Biol Med. 2021;131:104248. doi:10.1016/j.compbiomed.2021.104248
  12. Tandel N, Patel H, Choksi JK. MRI-based tumor classification using CNN and image enhancement techniques. Med Biol Eng Comput. 2022;60(2):357–369. doi:10.1007/s11517-021-02434-5.
Regular Issue Subscription Review Article
Volume 14
Issue 02
Received 27/03/2025
Accepted 29/03/2025
Published 11/06/2025
Publication Time 76 Days
114

[first_name] [last_name]

My IP

PlumX Metrics