Young Neurosurgeons and Rapid Fire Abstracts-Glioblastoma MRI Analytics Using Deep Learning

Video Transcription

Video Summary

In this video, Elie Veliani, a second-year medical student at the Icahn School of Medicine at Mount Sinai, presents his recent work on glioblastoma MRI analytics using deep learning. Glioblastoma is a common and aggressive brain tumor with a poor prognosis. The goal of the work was to develop a machine learning pipeline that can analyze preoperative MRI scans and predict genetic characteristics of the tumor. The researchers used a dataset of MRI scans and genome information from glioma patients and developed a segmentation algorithm that accurately identified tumor regions. The algorithm also performed well in predicting genetic biomarkers associated with tumor grade, IDH status, 1p19q code deletions, EGFR amplification, and MGMT promoter methylation. The findings suggest that MRI analysis can effectively characterize glioma biomarkers. Future work includes incorporating additional radiomics characteristics and validating the pipeline in clinical workflows. The research was funded by the Neurosurgery Research and Education Foundation.

Asset Subtitle

Aly Al-Amyn Valliani

Keywords

glioblastoma

MRI analytics

deep learning

genetic characteristics

tumor segmentation