Iterative training of robust k‐space interpolation networks for improved image reconstruction with limited scan specific training samples
Please always quote using this URN: urn:nbn:de:bvb:20-opus-312306
- 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 parallelTo 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.…
Author: | Peter Dawood, Felix Breuer, Jannik Stebani, Paul Burd, István Homolya, Johannes Oberberger, Peter M. Jakob, Martin Blaimer |
---|---|
URN: | urn:nbn:de:bvb:20-opus-312306 |
Document Type: | Journal article |
Faculties: | Fakultät für Physik und Astronomie / Physikalisches Institut |
Medizinische Fakultät / Medizinische Klinik und Poliklinik I | |
Language: | English |
Parent Title (English): | Magnetic Resonance in Medicine |
Year of Completion: | 2023 |
Volume: | 89 |
Issue: | 2 |
First Page: | 812 |
Last Page: | 827 |
Source: | Magnetic Resonance in Medicine 2023, 89(2):812-827. DOI: 10.1002/mrm.29482 |
DOI: | https://doi.org/10.1002/mrm.29482 |
Dewey Decimal Classification: | 6 Technik, Medizin, angewandte Wissenschaften / 61 Medizin und Gesundheit / 610 Medizin und Gesundheit |
Tag: | GRAPPA; RAKI; complex‐valued machine learning; data augmentation; deep learning; parallel imaging |
Release Date: | 2023/06/27 |
Licence (German): | CC BY-NC: Creative-Commons-Lizenz: Namensnennung, Nicht kommerziell 4.0 International |