Iterative Neutrosophic Refinement for Reliable Tumor Segmentation in Brain MRI Images

Sofia Jennifer J; Kalaivani C; Saraswathi S

Authors

Sofia Jennifer J Department of Information Technology, Sri Sivasubramaniya Nadar College of Engineering, Chennai 603 110, India;
Kalaivani C Department of Mathematics, Sri Sivasubramaniya Nadar College of Engineering, Chennai 603 110, India;
Saraswathi S Department of Computer Science and Engineering, Sri Sivasubramaniya Nadar College of Engineering, Chennai 603 110, India;

Keywords:

Brain tumor, Refined neutrosophic sets, Tumor segmentation, Medical imaging.

Abstract

. Accurate segmentation of brain tumors in MRI scans is important for diagnosis and treatment,
but noise, overlapping tissues, low contrast, and unclear boundaries make it difficult for traditional image
processing methods. Our idea is to present a robust method for brain tumor segmentation using Refined
Neutrosophic Set (NRS). This approach builds on the traditional neutrosophic framework by breaking down
truth, indeterminacy, and falsity into multiple subcomponents for better handling of uncertainty . The input
MRI images is transformed into the neutrosophic refined set domain, and then iterative refinement of truth
and indeterminacy is applied. This method effectively handles uncertainty, reduces ambiguity, and produces
clearer, more accurate tumor boundaries. Experimental results on benchmark MRI datasets demonstrate that
the NRS-based method achieves higher Dice similarity scores and lower Hausdroff distance.

DOI 10.5281/zenodo.18882902