Targeted Neural Architectures in Multi-Objective Frameworks for Complete Glioma Characterization from Multimodal MRI
Shravan Venkatraman, Pandiyaraju V, Abeshek A, Aravintakshan S A, Pavan Kumar S, Kannan A, Madhan S
TL;DR
This work tackles complete glioma characterization from multimodal MRI by jointly localizing, segmenting, and grading tumors within a unified, multi-objective deep learning framework. It deploys a targeted architecture suite: SGA-LinkNet for precise localization, SE-ResNet101–LinkNet for high-fidelity segmentation, and SE-ResNet152 with Adaptive Boosting for robust grading, all fed by a hierarchical preprocessing pipeline (MHF, AFRC, DCS, LGCE, SFN). On BraTS2020, the approach achieves IoU around 0.96 for segmentation and an accuracy of approximately 98.53% for glioma grading, outperforming several state-of-the-art baselines. The framework promises improved diagnostic accuracy and supports potential clinical impact in early diagnosis and personalized treatment planning, with avenues for real-time deployment and integration of additional patient-specific data.
Abstract
Brain tumors result from abnormal cell growth in brain tissue. If undiagnosed, they cause neurological deficits, including cognitive impairment, motor dysfunction, and sensory loss. As tumors grow, intracranial pressure increases, potentially leading to fatal complications such as brain herniation. Early diagnosis and treatment are crucial to controlling these effects and slowing tumor progression. Deep learning (DL) and artificial intelligence (AI) are increasingly used to assist doctors in early diagnosis through magnetic resonance imaging (MRI) scans. Our research proposes targeted neural architectures within multi-objective frameworks that can localize, segment, and classify the grade of these gliomas from multimodal MRI images to solve this critical issue. Our localization framework utilizes a targeted architecture that enhances the LinkNet framework with an encoder inspired by VGG19 for better multimodal feature extraction from the tumor along with spatial and graph attention mechanisms that sharpen feature focus and inter-feature relationships. For the segmentation objective, we deployed a specialized framework using the SeResNet101 CNN model as the encoder backbone integrated into the LinkNet architecture, achieving an IoU Score of 96%. The classification objective is addressed through a distinct framework implemented by combining the SeResNet152 feature extractor with Adaptive Boosting classifier, reaching an accuracy of 98.53%. Our multi-objective approach with targeted neural architectures demonstrated promising results for complete glioma characterization, with the potential to advance medical AI by enabling early diagnosis and providing more accurate treatment options for patients.