Abstract #2261

Artificial neural network derived myelin water fraction map with multi-echo gradient echo signal: brain development from infants to adults.

Hyun Gi Kim¹, Jae Eun Song², Dongyeob Han³, Jee Young Kim¹, Se Won Oh¹, and Dong-Hyun Kim²

¹Radiology, The Catholic University of Korea, Seoul, Korea, Republic of, ²Yonsei University, Seoul, Korea, Republic of, ³Siemens Healthcare, Seoul, Korea, Republic of

Myelin water fraction (MWF) values were obtained by an artificial neural network (ANN-MWF) and complex model fitting (CF-MWF) with 3D multi-echo gradient-echo (mGRE) signal. Linear regression test showed high correlation between ANN-MWF and CF-MWF values (R² = 0.802, p < .001). Bland-Altman plot showed higher ANN-MWF values compared to CF-MWF values in the areas with high MWF values. The ANN-MWF values in the white matter showed a high association with age (R² = 0.821, p =.005).

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

Join Here