Research Article
Enhanced Brain Tumor Delineation Using t-SNE and Machine learning algorithms
@INPROCEEDINGS{10.4108/eai.28-4-2025.2358081, author={M. S. Minu and N M Vedhinee and Prakratee Singh and V. Karthik}, title={Enhanced Brain Tumor Delineation Using t-SNE and Machine learning algorithms}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part II}, publisher={EAI}, proceedings_a={ICITSM PART II}, year={2025}, month={10}, keywords={feature space transformation for medical imaging t-sne dimensionality mapping neuro-oncological analysis manifold gaining knowledge of tumor recognition hybrid clustering-cnn architecture lcg boundary enhancement multi-degree segmentation pipeline}, doi={10.4108/eai.28-4-2025.2358081} }- M. S. Minu
N M Vedhinee
Prakratee Singh
V. Karthik
Year: 2025
Enhanced Brain Tumor Delineation Using t-SNE and Machine learning algorithms
ICITSM PART II
EAI
DOI: 10.4108/eai.28-4-2025.2358081
Abstract
Our approach aims to alleviate the challenges of LCG detection, as tumors are often difficult to distinguish from surrounding brain tissue. Unlike conventional approaches that apply CNNs directly to raw MRI images, we employ t-SNE to enhance differentiation between tumor and non-tumor regions prior to segmentation. Experiments using a modified U-Net architecture on the Kaggle LCG MRI dataset demonstrate improved tumor detection, particularly in low-contrast regions, compared to baseline CNN methods. This hybrid strategy, which integrates clustering with supervised deep learning, provides more robust MRI analysis and offers clinicians more reliable tools for tumor delineation and treatment planning.
Copyright © 2025–2025 EAI
