Iterative training of robust k‐space interpolation networks for improved image reconstruction with limited scan specific training samples. Issue 2 (13th October 2022)

• Record Type: Journal Article
• Authors: Dawood, Peter; Breuer, Felix; Stebani, Jannik; Burd, Paul; Homolya, István; Oberberger, Johannes; Jakob, Peter M.; Blaimer, Martin
• Is Part Of: Magnetic resonance in medicine. Volume 89:Issue 2(2023)
• Page Start: 812
• Page End: 827
• Publication Date: 2022-10-13
• Subjects: complex‐valued machine learning -- data augmentation -- deep learning -- GRAPPA -- parallel imaging -- RAKI; Nuclear magnetic resonance -- Periodicals; Electron paramagnetic resonance -- Periodicals; 616.07548
• DOI: 10.1002/mrm.29482
• Languages: English
• ISSNs: 0740-3194

Abstract

Abstract : Purpose: To evaluate an iterative learning approach for enhanced performance of robust artificial‐neural‐networks for k‐space interpolation (RAKI), when only a limited amount of training data (auto‐calibration signals [ACS]) are available for accelerated standard 2D imaging. Methods: In a first step, the RAKI model was tailored for the case of limited training data amount. In the iterative learning approach (termed iterative RAKI [iRAKI]), the tailored RAKI model is initially trained using original and augmented ACS obtained from a linear parallel imaging reconstruction. Subsequently, the RAKI convolution filters are refined iteratively using original and augmented ACS extracted from the previous RAKI reconstruction. Evaluation was carried out on 200 retrospectively undersampled in vivo datasets from the fastMRI neuro database with different contrast settings. Results: For limited training data (18 and 22 ACS lines for R = 4 and R = 5, respectively), iRAKI outperforms standard RAKI by reducing residual artifacts and yields better noise suppression when compared to standard parallel imaging, underlined by quantitative reconstruction quality metrics. Additionally, iRAKI shows better performance than both GRAPPA and standard RAKI in case of pre‐scan calibration with varying contrast between training‐ and undersampled data. Conclusion: RAKI benefits from the iterative learning approach, which preserves the noise suppression feature, but requires less original training data for the accurate reconstruction of standard 2D images thereby improving net acceleration.