Reproducible Evaluation of Data Augmentation and Loss Functions for Brain Tumor Segmentation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autore principale: B, Saumya
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918157796507648
author B, Saumya
author_facet B, Saumya
contents Brain tumor segmentation is crucial for diagnosis and treatment planning, yet challenges such as class imbalance and limited model generalization continue to hinder progress. This work presents a reproducible evaluation of U-Net segmentation performance on brain tumor MRI using focal loss and basic data augmentation strategies. Experiments were conducted on a publicly available MRI dataset, focusing on focal loss parameter tuning and assessing the impact of three data augmentation techniques: horizontal flip, rotation, and scaling. The U-Net with focal loss achieved a precision of 90%, comparable to state-of-the-art results. By making all code and results publicly available, this study establishes a transparent, reproducible baseline to guide future research on augmentation strategies and loss function design in brain tumor segmentation.
format Preprint
id arxiv_https___arxiv_org_abs_2510_08617
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Reproducible Evaluation of Data Augmentation and Loss Functions for Brain Tumor Segmentation
B, Saumya
Computer Vision and Pattern Recognition
Machine Learning
Brain tumor segmentation is crucial for diagnosis and treatment planning, yet challenges such as class imbalance and limited model generalization continue to hinder progress. This work presents a reproducible evaluation of U-Net segmentation performance on brain tumor MRI using focal loss and basic data augmentation strategies. Experiments were conducted on a publicly available MRI dataset, focusing on focal loss parameter tuning and assessing the impact of three data augmentation techniques: horizontal flip, rotation, and scaling. The U-Net with focal loss achieved a precision of 90%, comparable to state-of-the-art results. By making all code and results publicly available, this study establishes a transparent, reproducible baseline to guide future research on augmentation strategies and loss function design in brain tumor segmentation.
title Reproducible Evaluation of Data Augmentation and Loss Functions for Brain Tumor Segmentation
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2510.08617