@article{Krauze_2022, title={Using Artificial Intelligence and Magnetic Resonance Imaging to Address Limitations in Response Assessment in Glioma}, url={https://mediterraneanjournals.com/index.php/oi/article/view/616}, abstractNote={<p>Gliomas are rapidly progressive, neurologically devastating, nearly uniformly fatal brain tumors. In WHO grade IV tumors like glioblastoma, the standard of care involves maximal surgical resection followed by concurrent radiation therapy (RT) and temozolomide (TMZ) chemotherapy followed by adjuvant TMZ. This results in overall survival (OS) of less than 30% at two years. Currently, tumor progression assessment is based on clinician assessment and MRI interpretation using Response Assessment in Neuro-Oncology (RANO) criteria. These criteria classify response as complete, partial, stable, or progression. This approach, however, suffers from significant limitations due to the difficulty in interpreting MRI findings on T1 gad and T2 FLAIR sequences, lack of concurrent correlation with radiation therapy fields, inconsistent follow-up imaging, concurrent administration of steroids, and systemic management, including immunotherapy. The neuro-oncology field struggles with classifying true progression vs. pseudoprogression vs. pseudoresponse with progression guidelines actively evolving. The lack of consensus on the definition of progression impairs the ability to initiate earlier management upon progression, judge the impact of therapies, and optimize and personalize management. Due to the pivotal role of imaging, radiology is at the center of the question of optimizing and advancing response criteria [1-5]. The hypothesis is that MRI images of patients with glioma, when subjected to change over time analysis (at diagnosis, prior to and post-radiation therapy), can identify features predictive of treatment failure helping guide patient management in the clinic. Likely a combination of imaging and biospecimen-driven biomarkers is needed. Given the large amount of data generated by both approaches, success in this space hinges on leveraging computational approaches and artificial intelligence algorithms validated using large-scale publicly available data sets to disentangle the complexity and heterogeneity inherent in glioma progression.</p>}, journal={Oncology Insights}, author={Krauze, Andra}, year={2022}, month={Apr.} }