Abstract #3677

Automated tumour definition in diffusion imaging from Gaussian mixture modelling of intrinsically-registered PET data in breast cancer: pilot study

Maren Marie Andreassen¹, Tone Bathen^1,2, Pål Erik Goa^3,4, Steinar Lundgren^5,6, Roja Hedayati^5,6, Torill Sjøbakk¹, Hans Petter Eikesdal^7,8, and Neil P. Jerome^1,4

¹Department of Circulation and Medical Imaging, NTNU, Norwegian University of Science and Technology, Trondheim, Norway, ²St.Olavs Hospital, Trondheim University Hospital, Trondheim, Norway, ³Department of Physics, NTNU, Norwegian University of Science and Technology, Trondheim, Norway, ⁴Department of Radiology and Nuclear Medicine, Trondheim University Hospital, Trondheim, Norway, ⁵Department of Clinical and Molecular Medicine, NTNU, Norwegian University of Science and Technology, Trondheim, Norway, ⁶Department of Oncology, Trondheim University Hospital, Trondheim, Norway, ⁷Section of Oncology, Department of Clinical Science, University of Bergen, Bergen, Norway, ⁸Department of Oncology, Haukeland University Hospital, Bergen, Norway

Simultaneous positron emission tomography and magnetic resonance imaging (PET/MRI) is an emerging technique in breast cancer practice, allowing collection of morphologic parameters in addition to real-time metabolism. Though feasibility has been demonstrated, the best approach to utilize PET/MRI has not yet been validated. This study focus on evaluating a Gaussian mixture model (GMM) based segmentation technique from PET images with intrinsic MRI registration as a proxy for regions-of-interest (ROIs) manually drawn on post contrast images. The application of the method has been evaluated in a neoadjuvant treatment response assessment setting using apparent diffusion coefficient (ADC) values.

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

Join Here