We proposed an automatic evaluation model for estimating the degree of motion artifacts in high-resolution multi-echo gradient echo images for nigrosome-1 visualization in the substantia nigra. A combination of a convolutional neural network and a long short-term memory was used to develop the automatic motion evaluation model. The results demonstrated that the proposed model could be useful tools for N1 visualization for diagnosing Parkinson’s disease.
This abstract and the presentation materials are available to members only; a login is required.