Journal: BMC Medical Imaging
Article Title: Neural networks with personalized training for improved MOLLI T 1 mapping
doi: 10.1186/s12880-025-01769-z
Figure Lengend Snippet: Block diagram for personalized-training and using the neural network for estimating T 1 values with MOLLI. Personalized-training (solid lines) was performed by simulating the MOLLI pulse sequence with its parameters and the physiological ECG triggers so as to generate a customized database from the MRI simulator for the specific MRI acquisition. Once the network was trained with volunteer-specific simulated MR data then it could be used (dashed lines) for generating T 1 maps from MOLLI images acquired by the scanner
Article Snippet: MOLLI pulse sequences at a 1.5T Philips Achieva scanner (Philips Healthcare, Best, Netherlands) with two different acquisition schemes were used: 5(3)3, the clinical MOLLI pulse sequence for native myocardial T 1 mapping with an additional three heartbeat pause between inversions.
Techniques: Blocking Assay, Sequencing