In-Silico Trained AI for Enhanced T2 Spectrum Imaging and Myelin Water Fraction Mapping in Preclinical 7T MRI

Daniel Vallejo-Aldana, Alonso Ramirez-Manzanares, Erick Jorge Canales-Rodríguez, Thomas Yu, Abraham Cisneros-Mejorado, Luis Concha, Jonathan Rafael-Patino, Jean-Philippe Thiran

Abstract

This study introduces a Machine Learning (ML) approach for estimating the T 2 spectrum and myelin water fraction (MWF) using multi-echo T 2 (MET 2 ) data from preclinical 7T Magnetic Resonance Imaging (MRI) scanners. ML methods have shown promise in MWF estimation, outperforming Regularized Non-Negative Least Squares (RNNLS). However, existing ML methods were optimized for high signal-to-noise ratios (SNR) typical of 3T clinical MET 2 data with larger voxel sizes. We adapted the Model-Informed Machine Learning (MIML) method to handle challenges in preclinical 7T MRI, including reduced voxel sizes, elevated noise levels (SNR=30-60), and shifts in T 2 lobes. Results from in-silico simulated data demonstrate the superior performance of the proposed multi-layer-perceptron-based solution over RNNLS. Validation with ME T 2 data from two mice—a healthy control and a cuprizone-exposed pathological mouse—confirms the ML method’s success in identifying cuprizone-induced demyelination. Our study showcases the adaptability and enhanced performance of the MIML approach under challenging preclinical 7T MRI conditions, contributing to the advancement of MWF estimation methods in high-field MRI settings.
BibTeX
@inproceedings{vallejoaldana2024silico,
  author = {Daniel Vallejo-Aldana and Alonso Ramirez-Manzanares and Erick Jorge Canales-Rodríguez and Thomas Yu and Abraham Cisneros-Mejorado and Luis Concha and Jonathan Rafael-Patino and Jean-Philippe Thiran},
  title = {{In-Silico Trained AI for Enhanced T2 Spectrum Imaging and Myelin Water Fraction Mapping in Preclinical 7T MRI}},
  booktitle = {ISBI},
  year = {2024},
  doi = {10.1109/isbi56570.2024.10635246},
}

Loading related content...

Figure