Abstract #2330

Discrimination of breast cancer from healthy breast tissues using a tri-exponential model

Maren Marie Sjaastad Andreassen¹, Ana Elvira Rodriguez-Zoto², Cristopher Charles Conlin², Igor Vidic³, Grace Sora Ahn⁴, Neil Peter Jerome^1,5, Agnes Østlie⁵, Tone Frost Bathen^1,5, Pål Erik Goa^3,5, Rebecca Rakow-Penner², and Anders Martin Dale^2,6

¹Department of Circulation and Medical Imaging, Norwegian University of Science and Technology, Trondheim, Norway, ²Department of Radiology, University of California, San Diego, La Jolla, CA, United States, ³Department of Physics, Norwegian University of Science and Technology, Trondheim, Norway, ⁴School of Medicine, University of California, San Diego, La Jolla, CA, United States, ⁵Department of Radiology and Nuclear Medicine, Norwegian University of Science and Technology, Trondheim, Norway, ⁶Department of Neurosciences, University of California, San Diego, La Jolla, CA, United States

The purpose of this work was to develop a method to discriminate breast cancer from healthy breast tissues using signal contribution maps from multi-b diffusion MRI acquisitions. Signal contributions were estimated using a tri-exponential model, with the ADC values for three distinct compartments assumed fixed across voxels and subjects. A linear discriminant function was constructed using the estimated signal contributions from the two lowest ADC components. Average ROC AUC for discriminating cancer from healthy breast tissues was 0.99 (CI95% = 0.98-1.00), superior to that of independent signal contributions, maximum b-value volume and conventional DWI estimates (ADC and apparent diffusion kurtosis).

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

Join Here