Model-informed machine learning for multi-component T2 relaxometry

• Published in: Medical image analysis Vol. 69; p. 1
• Main Authors: Yu, Thomas, Canales-Rodríguez, Erick Jorge, Pizzolato, Marco, Piredda, Gian Franco, Hilbert, Tom, Fischi-Gomez, Elda, Weigel, Matthias, Barakovic, Muhamed, Bach Cuadra, Meritxell, Granziera, Cristina, Kober, Tobias, Thiran, Jean-Philippe
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.04.2021; Elsevier BV
• Subjects: [formula omitted] relaxometry; Algorithms; Biomarkers; Brain; Learning algorithms; Lesions; Localization; Machine learning; Magnetic resonance imaging; Multilayers; Multiple sclerosis; Myelin; Myelin water imaging; Tissue analysis; Tissues; Myelin water imaging; Machine learning; T2 relaxometry
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2020.101940

•Recovering T2 distributions from MRI data is difficult, especially due to noise.•Noise robust recovery possible through machine learning on synthetic data.•Synthetic data is derived from biophysical models.•Validated on in-vivo/ex-vivo scans including a subject with multiple sclerosis. [Display omitted] Recovering the T2 distribution from multi-echo T2 magnetic resonance (MR) signals is challenging but has high potential as it provides biomarkers characterizing the tissue micro-structure, such as the myelin water fraction (MWF). In this work, we propose to combine machine learning and aspects of parametric (fitting from the MRI signal using biophysical models) and non-parametric (model-free fitting of the T2 distribution from the signal) approaches to T2 relaxometry in brain tissue by using a multi-layer perceptron (MLP) for the distribution reconstruction. For training our network, we construct an extensive synthetic dataset derived from biophysical models in order to constrain the outputs with a priori knowledge of in vivo distributions. The proposed approach, called Model-Informed Machine Learning (MIML), takes as input the MR signal and directly outputs the associated T2 distribution. We evaluate MIML in comparison to a Gaussian Mixture Fitting (parametric) and Regularized Non-Negative Least Squares algorithms (non-parametric) on synthetic data, an ex vivo scan, and high-resolution scans of healthy subjects and a subject with Multiple Sclerosis. In synthetic data, MIML provides more accurate and noise-robust distributions. In real data, MWF maps derived from MIML exhibit the greatest conformity to anatomical scans, have the highest correlation to a histological map of myelin volume, and the best unambiguous lesion visualization and localization, with superior contrast between lesions and normal appearing tissue. In whole-brain analysis, MIML is 22 to 4980 times faster than the non-parametric and parametric methods, respectively.