Advanced deep learning models based on neutrosophic logic for the analysis of brain tumor medical images

Abstract

The categorization of medical photographs poses considerable difficulties owing to noise, 
uncertainty, and ambiguous information. Conventional deep learning models frequently 
encounter difficulties in addressing this issue, resulting in diminished diagnostic precision, 
particularly in the context of low-quality or ambiguous situations. This work presents a hybrid 
methodology that combines Neutrosophic Set (NS) theory with deep learning models to 
improve magnetic resonance imaging (MRI) picture classification in uncertain settings. NS 
theory delineates three domains: True (T), Indeterminate (I) and False (F) to address picture 
uncertainty and noise, hence enhancing deep learning models' capacity to analyze complex, 
ambiguous visual data. To assess the methodology, four advanced deep learning models 
MobileNet, VGG16, DenseNet121 and InceptionV3 were employed, and their efficacy was 
analyzed on brain tumor medical image datasets. The findings demonstrate that models trained 
on NS-transformed data, especially DenseNet and inception, produce enhanced results relative 
to those trained on the original data, attaining notably higher accuracy, precision, and recall. 
This illustrates that integrating NS theory into deep learning models markedly improves their 
capacity to categorize uncertain and noisy MRI pictures, offering a reliable method for 
enhancing diagnostic accuracy in medical imaging. 

DOI: 10.5281/zenodo.15207970