1Faculty of Computer Science, Universitas Dian Nuswantoro, Indonesia
2School of Mechanical Engineering, Pusan National University, South Korea
3Anugrah General Hospital, Indonesia
BibTex Citation Data :
@article{JMASIF82619, author = {Muhammad Naufal and Harun Al Azies and Farrikh Alzami and Novita Kurnia Ningrum and Rivaldo Mersis Brilianto and Pratidina Kusuma Dewi}, title = {Enhancing Brain Tumor Classification on Mildly Imbalanced Datasets Using Categorical Focal Cross-Entropy}, journal = {Jurnal Masyarakat Informatika}, volume = {17}, number = {2}, year = {2026}, keywords = {Class Imbalance; Convolutional Neural Network; Focal Loss; Brain Tumor Classification;}, abstract = { Brain tumor is a dangerous disease that requires accurate diagnosis, but medical datasets often suffer from class imbalance, resulting in bias in deep learning-based classification models. This study proposes using Categorical Focal Cross Entropy (CFCE) in a Convolutional Neural Network (CNN) to address this issue by comparing it with Categorical Cross-Entropy (CCE). CFCE is designed to emphasize minority class samples and hard-to-classify examples, thereby reducing the dominance of the majority class. Experiments were conducted on a brain tumor dataset with class imbalance, where the CNN model with CFCE achieved 84.57% accuracy, 83.57% precision, 85.27% recall, and 83.97% F1-score, outperforming the model with CCE (81.30% accuracy, 81.77% precision, 82.83% recall, and 81.15% F1-score). These results show that Focal Loss effectively improves the classification performance on imbalanced data, with a better ability to detect brain tumors, especially in the minority class. This study contributes to developing a more robust and reliable deep learning-based diagnosis system for medical applications. }, issn = {2777-0648}, pages = {218--228} doi = {10.14710/jmasif.17.2.82619}, url = {https://ejournal.undip.ac.id/index.php/jmasif/article/view/82619} }
Refworks Citation Data :
Brain tumor is a dangerous disease that requires accurate diagnosis, but medical datasets often suffer from class imbalance, resulting in bias in deep learning-based classification models. This study proposes using Categorical Focal Cross Entropy (CFCE) in a Convolutional Neural Network (CNN) to address this issue by comparing it with Categorical Cross-Entropy (CCE). CFCE is designed to emphasize minority class samples and hard-to-classify examples, thereby reducing the dominance of the majority class. Experiments were conducted on a brain tumor dataset with class imbalance, where the CNN model with CFCE achieved 84.57% accuracy, 83.57% precision, 85.27% recall, and 83.97% F1-score, outperforming the model with CCE (81.30% accuracy, 81.77% precision, 82.83% recall, and 81.15% F1-score). These results show that Focal Loss effectively improves the classification performance on imbalanced data, with a better ability to detect brain tumors, especially in the minority class. This study contributes to developing a more robust and reliable deep learning-based diagnosis system for medical applications.
Article Metrics:
Last update:
Last update: 2026-10-02 17:02:41
The authors who submit the manuscript must understand that the article's copyright belongs to the author(s) if accepted for publication. However, the author(s) grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. Authors should also understand that their article (and any additional files, including data sets, and analysis/computation data) will become publicly available once published under that license. By submitting the manuscript to Jmasif, the author(s) agree with this policy. No special document approval is required.
The author(s) guarantee that:
The author(s) retain all rights to the published work, such as (but not limited to) the following rights:
Suppose the article was prepared jointly by more than one author. Each author submitting the manuscript warrants that all co-authors have given their permission to agree to copyright and license notices (agreements) on their behalf and notify co-authors of the terms of this policy. Jmasif will not be held responsible for anything arising because of the writer's internal dispute. Jmasif will only communicate with correspondence authors.
Authors should also understand that their articles (and any additional files, including data sets and analysis/computation data) will become publicly available once published. The license of published articles (and additional data) will be governed by a Creative Commons Attribution-ShareAlike 4.0 International License. Jmasif allows users to copy, distribute, display and perform work under license. Users need to attribute the author(s) and Jmasif to distribute works in journals and other publication media. Unless otherwise stated, the author(s) is a public entity as soon as the article is published.
