Artificial neural network for myelin water imaging

TL;DR

This paper demonstrates the development of neural networks that enable real-time, highly accurate myelin water imaging processing, significantly accelerating traditional methods and applicable to clinical and research settings.

Contribution

Introduces three neural networks for rapid, accurate myelin water imaging analysis, outperforming conventional methods in speed and maintaining high precision.

Findings

Neural networks achieved under 3% error in MWF estimation.

Processing time reduced to under 1 second, vastly faster than traditional methods.

High correlation (R2 > 0.97) with conventional MWI results.

Abstract

Purpose: To demonstrate the application of artificial-neural-network (ANN) for real-time processing of myelin water imaging (MWI). Methods: Three neural networks, ANN-IMWF, ANN-IGMT2, and ANN-II, were developed to generate MWI. ANN-IMWF and ANN-IGMT2 were designed to output myelin water fraction (MWF) and geometric mean T2 (GMT2,IEW), respectively whereas ANN-II generates a T2 distribution. For the networks, gradient and spin echo data from 18 healthy controls (HC) and 26 multiple sclerosis patients (MS) were utilized. Among them, 10 HC and 12 MS had the same scan parameters and were used for training (6 HC and 6 MS), validation (1 HC and 1 MS), and test sets (3 HC and 5 HC). The remaining data had different scan parameters and were applied to exam the effects of the scan parameters. The network results were compared with those of conventional MWI in the white matter mask and regions of interest (ROI). Results: The networks produced highly accurate results, showing averaged normalized root-mean-squared error under 3% for MWF and 0.4% for GMT2,IEW in the white matter mask of the test set. In the ROI analysis, the differences between ANNs and conventional MWI were less than 0.1% in MWF and 0.1 ms in GMT2,IEW (no statistical difference and R2 > 0.97). Datasets with different scan parameters showed increased errors. The average processing time was 0.68 sec in ANNs, gaining 11,702 times acceleration in the computational speed (conventional MWI: 7,958 sec). Conclusion: The proposed neural networks demonstrate the feasibility of real-time processing for MWI with high accuracy.