Abstract #2661

Automated Segmentation of Ewing’s Sarcoma using Diffusion Weighted Imaging

Amit Mehndiratta^1,2, Abhimanyu Sahai¹, Esha Baidya Kayal ¹, Jayendra Tiru Alampally³, Sameer Bakhshi⁴, Devasenathipathy K³, and Raju Sharma³

¹Center for Biomedical Engineering, Indian Institute of Technology Delhi, New Delhi, India, ²Department of Biomedical Engineering, All Indian Institute of Medical Sciences, New Delhi, India, ³Department of Radiology, All India Institute of Medical Sciences, New Delhi, India, ⁴BRA IRCH, All India Institute of Medical Sciences, New Delhi, India

Accurate demarcation of tumors on DWI MRimages could play a crucial role in diagnosis and prognosis when using quantitative image analysis like ADC or IVIM. Manual demarcation of tumour on each slice of a 3D stack is usually not feasible. Automated or semi-automated methods of segmentation are thus desirable specifically for DWimages that can be used to identify the tumor region, optimizing on both speed and accuracy. Our results reveals that semi-automated algorithms based on both Otsu-threshold or Active-Contours based region growing perform tumour segmentation with acceptable level of accuracy in diffusion MRimages and reduce time and manual effort required.

This abstract and the presentation materials are available to members only; a login is required.

Join Here