# Joint Supervised and Self-supervised Learning for MRI Reconstruction

arXiv: https://arxiv.org/abs/2311.15856 · HTML: https://mlsanity.com/paper/2311.15856/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 45 references that are in the map.
- First posted on arXiv 27 Nov 2023.

## Builds on (most cited first, 18 of 18)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [fastMRI: An Open Dataset and Benchmarks for Accelerated MRI](https://mlsanity.com/paper/1811.08839/) (2018)
- [Learning a Variational Network for Reconstruction of Accelerated MRI Data](https://mlsanity.com/paper/1704.00447/) (2017)
- [Noise2Self: Blind Denoising by Self-Supervision](https://mlsanity.com/paper/1901.11365/) (2019)
- [Self-Supervised MRI Reconstruction with Unrolled Diffusion Models](https://mlsanity.com/paper/2306.16654/) (2023)
- [Test-Time Training Can Close the Natural Distribution Shift Performance Gap in Deep Learning Based Compressed Sensing](https://mlsanity.com/paper/2204.07204/) (2022)
- [Recurrent Variational Network: A Deep Learning Inverse Problem Solver applied to the task of Accelerated MRI Reconstruction](https://mlsanity.com/paper/2111.09639/) (2021)
- [DSFormer: A Dual-domain Self-supervised Transformer for Accelerated Multi-contrast MRI Reconstruction](https://mlsanity.com/paper/2201.10776/) (2022)
- [A theoretical framework for self-supervised MR image reconstruction using sub-sampling via variable density Noisier2Noise](https://mlsanity.com/paper/2205.10278/) (2022)
- [Self-Supervised Learning for MRI Reconstruction with a Parallel Network Training Framework](https://mlsanity.com/paper/2109.12502/) (2021)
- [Self-Score: Self-Supervised Learning on Score-Based Models for MRI Reconstruction](https://mlsanity.com/paper/2209.00835/) (2022)
- [A review and experimental evaluation of deep learning methods for MRI reconstruction](https://mlsanity.com/paper/2109.08618/) (2021)
- [Dual-Domain Self-Supervised Learning for Accelerated Non-Cartesian MRI Reconstruction](https://mlsanity.com/paper/2302.09244/) (2023)
- [On Retrospective k-space Subsampling schemes For Deep MRI Reconstruction](https://mlsanity.com/paper/2301.08365/) (2023)
- [vSHARP: variable Splitting Half-quadratic Admm algorithm for Reconstruction of inverse-Problems](https://mlsanity.com/paper/2309.09954/) (2023)
- [Deep Cardiac MRI Reconstruction with ADMM](https://mlsanity.com/paper/2310.06628/) (2023)
- [Zero-Shot Self-Supervised Joint Temporal Image and Sensitivity Map Reconstruction via Linear Latent Space](https://mlsanity.com/paper/2303.02254/) (2023)
- [Optimization methods for MR image reconstruction (long version)](https://mlsanity.com/paper/1903.03510/) (2019)

## Built on by (most cited first, 4 of 4)

- [End-to-end Adaptive Dynamic Subsampling and Reconstruction for Cardiac MRI](https://mlsanity.com/paper/2403.10346/) (2024)
- [Deep End-to-end Adaptive k-Space Sampling, Reconstruction, and Registration for Dynamic MRI](https://mlsanity.com/paper/2411.18249/) (2024)
- [Equivariant Multiscale Learned Invertible Reconstruction for Cone Beam CT: From Simulated to Real Data](https://mlsanity.com/paper/2512.21180/) (2025)
- [Joint Supervised and Self-Supervised Training with Acquisition-Robust Techniques for Accelerated 4D Flow MRI Reconstruction](https://mlsanity.com/paper/2609.38644/) (2026)

## Abstract (arXiv)

Magnetic Resonance Imaging (MRI) represents an important diagnostic modality; however, its inherently slow acquisition process poses challenges in obtaining fully-sampled k-space data under motion. In the absence of fully-sampled acquisitions, serving as ground truths, training deep learning algorithms in a supervised manner to predict the underlying ground truth image becomes challenging. To address this limitation, self-supervised methods have emerged as a viable alternative, leveraging available subsampled k-space data to train deep neural networks for MRI reconstruction. Nevertheless, these approaches often fall short when compared to supervised methods. We propose Joint Supervised and Self-supervised Learning (JSSL), a novel training approach for deep learning-based MRI reconstruction algorithms aimed at enhancing reconstruction quality in cases where target datasets containing fully-sampled k-space measurements are unavailable. JSSL operates by simultaneously training a model in a self-supervised learning setting, using subsampled data from the target dataset(s), and in a supervised learning manner, utilizing datasets with fully-sampled k-space data, referred to as proxy datasets. We demonstrate JSSL's efficacy using subsampled prostate or cardiac MRI data as the target datasets, with fully-sampled brain and knee, or brain, knee and prostate k-space acquisitions, respectively, as proxy datasets. Our results showcase substantial improvements over conventional self-supervised methods, validated using common image quality metrics. Furthermore, we provide theoretical motivations for JSSL and establish "rule-of-thumb" guidelines for training MRI reconstruction models. JSSL effectively enhances MRI reconstruction quality in scenarios where fully-sampled k-space data is not available, leveraging the strengths of supervised learning by incorporating proxy datasets.