Bias-Reduced Neural Networks for Parameter Estimation in Quantitative MRI

Abstract

Purpose: To develop neural network (NN)-based quantitative MRI parameter estimators with minimal bias and a variance close to the Cramér-Rao bound. Theory and Methods: We generalize the mean squared error loss to control the bias and variance of the NN's estimates, which involves averaging over multiple noise realizations of the same measurements during training. Bias and variance properties of the resulting NNs are studied for two neuroimaging applications. Results: In simulations, the proposed strategy reduces the estimates' bias throughout parameter space and achieves a variance close to the Cramér-Rao bound. In vivo, we observe good concordance between parameter maps estimated with the proposed NNs and traditional estimators, such as non-linear least-squares fitting, while state-of-the-art NNs show larger deviations. Conclusion: The proposed NNs have greatly reduced bias compared to those trained using the mean squared error and offer significantly improved computational efficiency over traditional estimators with comparable or better accuracy.

References in corpus (9)

• A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
• Reconciling modern machine learning practice and the bias-variance trade-off
• MR fingerprinting Deep RecOnstruction NEtwork (DRONE)
• Low Rank Alternating Direction Method of Multipliers Reconstruction for MR Fingerprinting
• Bayesian MRI Reconstruction with Joint Uncertainty Estimation using Diffusion Models
• Bias in parametric estimation: reduction and useful side-effects
• Hybrid-State Free Precession in Nuclear Magnetic Resonance
• Generalized Bloch model: a theory for pulsed magnetization transfer
• Learning to Estimate Without Bias

Cited by in corpus (2)

• Unconstrained quantitative magnetization transfer imaging: disentangling T1 of the free and semi-solid spin pools
• Cramér-Rao Bound Optimized Subspace Reconstruction in Quantitative MRI