مقالات پذیرفته شده در نهمین کنگره بین المللی زیست پزشکی
Artificial Intelligence in Glioblastoma: Emerging Approaches for Diagnosis, Therapeutic Decision-Making, and Response Prediction
Artificial Intelligence in Glioblastoma: Emerging Approaches for Diagnosis, Therapeutic Decision-Making, and Response Prediction
Ilnaz Rahimmanesh,1,*
1. Applied Physiology Research Center, Cardiovascular Research Institute, Isfahan University of Medical Sciences, Isfahan, Iran
Introduction: Glioblastoma multiforme (GBM) is the most aggressive primary brain tumor, characterized by infiltrative growth, significant heterogeneity, and resistance to standard therapies. While essential, traditional strategies remain insufficient for achieving long-term survival. This review explores how recent progress in artificial intelligence (AI) and machine learning (ML) is introducing a transformative paradigm in neuro-oncology.
Methods: This analysis synthesizes findings from recent literature on the application of AI and ML in GBM management. It examines methodologies including deep learning and radiomics for imaging analysis (MRI), multi-omics integration (genomics, transcriptomics), digital pathology using convolutional neural networks, and predictive modeling for treatment response. The development and validation of these models across various datasets are discussed.
Results: AI-driven strategies have demonstrated significant advances. In diagnostics, radiomics and deep learning algorithms excel at differentiating GBM from other lesions, characterizing subtypes, and non-invasively predicting molecular markers (e.g., IDH, MGMT). For treatment, AI models inform personalized therapy by predicting responses to temozolomide and targeted agents, optimizing radiotherapy planning, and automating tumor segmentation. Furthermore, AI enables superior monitoring by differentiating true progression from pseudoprogression and providing high-resolution insights into the tumor microenvironment through digital pathology. AI also shows promise in stratifying patients for immunotherapy and accelerating drug discovery.
Conclusion: Artificial intelligence is redefining the landscape of GBM management by enabling earlier diagnosis, precise treatment planning, and reliable prediction of therapeutic responses. Despite challenges related to validation, generalizability, and integration into clinical workflows, AI-driven strategies represent a powerful complement to traditional approaches. The convergence of AI with large-scale multimodal data holds immense promise for advancing personalized cancer therapy and improving patient outcomes in neuro-oncology.