Tehran University of Medical Sciences

Science Communicator Platform

Stay connected! Follow us on X network (Twitter):
Share this content! By
A Quantitative Comparison Between Focal Loss and Binary Cross-Entropy Loss in Brain Tumor Auto-Segmentation Using U-Net Publisher



Ms Neyestanak Mahdi SHAFIEI ; H Jahani HAMID ; M Khodarahmi MOHSEN ; J Zahiri JAVAD ; M Hosseini MOSTAFA ; Aa Fatoorchi Amir ALI ; Ms Yekaninejad Mir SAEID
Authors

Source: Journal of Biostatistics and Epidemiology Published:2025


Abstract

Introduction: Brain tumors are among the most fatal cancers and cause the deaths of many people annually. Early diagnosis of a brain tumor can help save the patient’s life. Methods: We have collected a dataset consisting of 314 brain MRI images in all planes taken after administering a contrast medium with the dimension of 800*512, which offers the highest resolution. First, skull stripping has been implemented to separate the brain from other parts in the images. Next, we have annotated the tumors in the images under the supervision of experienced radiologists to create ground truth. To determine the most effective model versions for all three loss functions, hyperparameter tuning was performed. Following the comparison, the study further evaluates the effectiveness of two loss functions, Binary Cross-Entropy (BCE) and Focal loss, specifically in handling tumor regions within the dataset. Results: The two proposed loss functions were evaluated using 5-fold cross-validation, and the average precision, recall, and F1-score were 76.16%, 71.9%, and 74.52 for BCE loss and 82.92%, 79.32%, and 81% for the Focal loss on the test data, respectively. Moreover, the accuracy for BCE loss was 99.03% and 99.44% for the Focal loss. Conclusion: We recommend using BCE loss cautiously in classification tasks without data imbalance and emphasize the adoption of Focal loss for more accurate and reliable results in brain tumor segmentation. © 2025 Elsevier B.V., All rights reserved.
Related Docs
View other Related Docs
1. Fully Automated Gross Tumor Volume Delineation From Pet in Head and Neck Cancer Using Deep Learning Algorithms, Clinical Nuclear Medicine (2021)
2. Mri-Based Machine Learning for Determining Quantitative and Qualitative Characteristics Affecting the Survival of Glioblastoma Multiforme, Magnetic Resonance Imaging (2022)
3. Machine Learning-Based Overall Survival Prediction in Gbm Patients Using Mri Radiomics, 2022 IEEE NSS/MIC RTSD - IEEE Nuclear Science Symposium# Medical Imaging Conference and Room Temperature Semiconductor Detector Conference (2022)
Experts (# of related papers)
View all Related Experts
Seyed Maysam Alimohamadi (1)
Mostafa Farzin (1)
Other Related Docs
4. Segmentation of Gbm in Mri Images Using an Efficient Speed Function Based on Level Set Method, Proceedings - 2017 10th International Congress on Image and Signal Processing# BioMedical Engineering and Informatics# CISP-BMEI 2017 (2017)
5. Exploring Two Methods of Cbv Estimation in Two Groups of Grade Iii Gliomas With Different Appearance on Post-Contrast T1 Images, Journal of Biomedical Physics and Engineering (2020)
6. Diagnostic Accuracy of Deep Learning Models in Predicting Glioma Molecular Markers: A Systematic Review and Meta-Analysis, Diagnostics (2025)
7. Brain Tumor Segmentation Using Multimodal Mri and Convolutional Neural Network, 2022 30th International Conference on Electrical Engineering# ICEE 2022 (2022)
8. Radiomics Model Based on Computed Tomography Images for Prediction of Radiation-Induced Optic Neuropathy Following Radiotherapy of Brain and Head and Neck Tumors, Heliyon (2025)
9. Evaluating Machine Learning Models for Post-Surgery Treatment Response Assessment in Glioblastoma Multiforme: A Comparative Study of Gray Level Co-Occurrence Matrix (Glcm), Curvelet, and Combined Radiomics Features Selected by Multiple Algorithms, BMC Medical Imaging (2025)
10. Deep Learning-Based Automated Delineation of Head and Neck Malignant Lesions From Pet Images, 2020 IEEE Nuclear Science Symposium and Medical Imaging Conference# NSS/MIC 2020 (2020)
11. Enhancing Classification of Active and Non-Active Lesions in Multiple Sclerosis: Machine Learning Models and Feature Selection Techniques, BMC Medical Imaging (2024)
12. Lung Cancer Management: Revolutionizing Patient Outcomes Through Machine Learning and Artificial Intelligence, Cancer Reports (2025)
13. Fully Automated Clinical Target Volume Segmentation for Glioblastoma Radiotherapy Using a Deep Convolutional Neural Network, Polish Journal of Radiology (2023)
14. Enhancing Radiomics Robustness Using Bayesian Penalized Likelihood Pet Reconstruction: Application to Phantom and Non-Small Cell Lung Cancer Patient Studies, BMC Medical Imaging (2025)
15. Information Fusion for Fully Automated Segmentation of Head and Neck Tumors From Pet and Ct Images, Medical Physics (2024)
16. Current Trends in Glioma Tumor Segmentation: A Survey of Deep Learning Modules, Physica Medica (2025)
17. Cochlea Ct Radiomics Predicts Chemoradiotherapy Induced Sensorineural Hearing Loss in Head and Neck Cancer Patients: A Machine Learning and Multi-Variable Modelling Study, Physica Medica (2018)
18. Artificial Intelligence for Tumor [18F]Fdg Pet Imaging: Advancements and Future Trends - Part Ii, Seminars in Nuclear Medicine (2025)
19. Mri Radiomic Features Harmonization: A Multi-Center Phantom Study, 2022 IEEE NSS/MIC RTSD - IEEE Nuclear Science Symposium# Medical Imaging Conference and Room Temperature Semiconductor Detector Conference (2022)
20. Diagnostic Accuracy of Radiomics and Artificial Intelligence Models in Diagnosing Lymph Node Metastasis in Head and Neck Cancers: A Systematic Review and Meta-Analysis, Neuroradiology (2025)