AUTHOR=Gaviraghi Marta , Ricciardi Antonio , Palesi Fulvia , Brownlee Wallace , Vitali Paolo , Prados Ferran , Kanber Baris , Gandini Wheeler-Kingshott Claudia A. M. TITLE=A generalized deep learning network for fractional anisotropy reconstruction: Application to epilepsy and multiple sclerosis JOURNAL=Frontiers in Neuroinformatics VOLUME=Volume 16 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/neuroinformatics/articles/10.3389/fninf.2022.891234 DOI=10.3389/fninf.2022.891234 ISSN=1662-5196 ABSTRACT=Fractional anisotropy (FA) is a quantitative map sensitive to microstructural properties of tissue in-vivo that is extensively used to study the healthy and pathological brain. This map is classically calculated by model fitting (standard method) and requires many diffusion-weighted (DW) images for data quality and unbiased readings, hence needing acquisition times of several minutes. Here we adapted the U-net architecture to be generalized and to obtain good quality FA from DW volumes acquired in one minute. Our network requires 10 input DW volumes (hence fast acquisition), is robust to the direction of application of the diffusion gradients (hence generalised) and preserves/improves maps quality (hence good quality maps). We trained the network on the human connectome project (HCP) data, using standard model fitting on the entire set of DW directions to extract FA (ground truth). We addressed the generalization problem i.e., we trained the network to be applicable, without re-training, to clinical datasets acquired on different scanners with different DW imaging protocols. The network was applied to two different clinical datasets to assess FA quality and sensitivity to pathology in temporal lobe epilepsy and multiple sclerosis, respectively. For HCP data, when compared to the ground truth FA, the FA obtained from 10 DW volumes using the network was significantly better (p<10-4) than the FA obtained using the standard pipeline. For the clinical datasets, the network FA retained the same microstructural characteristics as the FA calculated with all DW volumes using the standard method. At subject level, the comparison between white matter (WM) ground truth FA values and network FA showed the same distribution; at group level, statistical differences of WM values detected in the clinical datasets with the ground truth FA were reproduced when using values from the network FA, i.e. the network retained sensitivity to pathology. In conclusion, the proposed network provides a clinically available method to obtain FA from a generic set of 10 DW volumes acquirable in one minute, augmenting data quality compared to direct model fitting, reducing the possibility of bias from a sub-sampled data and retaining FA pathological sensitivity, which is very attractive for clinical applications.