Abstract #3204

Using numerical simulations to compare and evaluate different mathematical models for analyzing dynamic contrast enhanced MRI data

Dianning He^1,2, Wei Qian^1,3, Lisheng Xu^1,4, and Xiaobing Fan²

¹Sino-Dutch Biomedical and Information Engineering School, Northeastern University, Shenyang, China, People's Republic of, ²Radiology, University of Chicago, Chicago, IL, United States, ³Electrical and Computer Engineering, University of Texas at El Paso, El Paso, TX, United States, ⁴Key Laboratory of Medical Image Computing, Ministry of Education, Shenyang, China, People's Republic of

Numerical simulations were performed to study analytical mathematical models for fitting tissue contrast agent concentration curves obtained from DCE-MRI. Randomly generated K^trans and v_e were used to calculate the curves using the Tofts model. A total of five analytical mathematical models, empirical mathematical model, modified logistic model, modified sigmoidal function, Weibull model, and extended phenomenological universalities were compared and evaluated in terms of how well they fitted to 100 curves. Statistical analysis showed that the empirical mathematical model provided the best fit out of those models. The analytical mathematical models were different despite having the same number of parameters.

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

Join Here