0939

View Abstract / View Presentation

Radiomics Based Classification of Ependymoma and High Grade Glioma Using Multimodal MRI

Apoorva Safai¹, Sumeet Shinde¹, Manali Jadhav¹, Tanay Chougule¹, Abhilasha Indoria², Manoj Kumar², Vani Santosh², Shumyla Jabeen², Manish Beniwal², Subhash Konar², Jitender Saini², and Madhura Ingalhalikar¹
¹Symbiosis International University, Pune, India, ²National Institute of Mental Health and Neurosciences, Bangalore, India

Quantitative radiomic markers such as texture and first order statistics from multimodal MRI can capture intricate and complementary information and thus aid in a robust multiclass tumor classification of STEE and HGG subtypes.

Fig-2:Processing pipeline implemented for radiomics analysis and classification of tumor subgroups

Fig-5: Feature importances obtained using SVM coefficient scores on multimodal feature set