Bachelor and Master projects at MRISM Lab – SPRING 2026
Optimization and evaluation of a DL reconstruction model for 3D T1ρ/T2 mapping of the knee cartilage
Project description:
Background: Relaxometry on the transverse plane (T2) and in the rotating frame (T1ρ) play an important role in the study of brain pathologies as well as cartilage osteoarthritis. However, conventional spin-echo-based sequences for T2/ T1ρ mapping are limited to 2D or 3D-slab imaging with anisotropic resolutions due to high energy deposition and long acquisition time. In this project, we investigate magnetization preparation approaches (T2 or T1ρ) combined with fast gradient echo sequences that allow for volumetric and isotropic high-resolution mapping of the entire organ of interest.
Our current technique employs incoherent undersampled data acquisition and compressed sensing (CS) image reconstruction. To further reduce acquisition times and/or improve robustness and signal-to-noise, we have recently implemented a deep learning (DL) framework for reconstruction. The network has been trained on different gradient echo images of the brain and finetuned at different field strengths. This already showed improved SNR of knee T1rho maps at 3 T and promises to reduce acquisition times by few minutes, making the technique more suitable for clinical workflow.
Goals:
- To characterize the performance of the current DL framework in comparison to CS in a standardized phantom (ISMRM-NIST) and in healthy subjects (e.g. linear regression and Bland-Altman analysis of T2/T1rho estimates).
- To finetune the current DL model with QuantoRAGE images of knee and hip cartilage.
- To characterize the finetuned DL model in comparison to current DL model and CS in phantom and healthy subjects.
- (Optional) To investigate other accelerated acquisition strategies (GRAPPA/CAIPIRINHA) in combination with DL reconstruction.
The student will have the opportunity to:
- Learn about CS and DL reconstructions
- Learn about acceleration strategies for acquisition
- Have access to source DL recon for finetuning
- Acquire MR data in phantoms and in humans
- Perform comparison analysis of different methods (statistics)
Requirements:
- Knowledge of signal and image processing.
- Basic knowledge of MR physics.
- Basic knowledge of Machine Learning.
- MATLAB, Python programming. C/C++ is an advantage.
- Independent worker / Problem solving attitude
- Good command of English.
Duration: 6 months
Supervisors:
Day-to-day supervision and location: Gabriele Bonanno, PhD , Translational Imaging Center of sitem-insel, Bern
Academic supervisors: Jonathan Stelter, Prof. Dimitrios Karampinos
The project will be carried out in close collaboration with the Acquisition and Reconstruction groups of the Swiss Innovation Hub, Siemens Healthineers, Lausanne.
How to apply: please send your CV and cover letter to [email protected]
References:
- Bonanno G, et al. Program number 1475. Intl. Soc. Mag. Reson. Med. 29 (2021)
https://cds.ismrm.org/protected/21MPresentations/abstracts/1475.html - Forman C, et al. High-resolution 3D whole-heart coronary MRA: a study on the combination of data acquisition in multiple breath-holds and 1D residual respiratory motion compensation. Magn Reson Mater Phy 2014;27:435-443.
- Nezafat R et al., B1-Insensitive T2 Preparation for Improved Coronary Magnetic Resonance Angiography at 3 T. Magnetic Resonance in Medicine 55:858–864 (2006)
- Sharafi A, Xia D, Chang G, Regatte RR. Biexponential T1ρ relaxation mapping of human knee cartilagein vivoat 3 T. NMR Biomed. 2017;30(10):e3760. 10.1002/nbm.3760
- Wetzl J, Forman C, Wintersperger BJ, et al. High‐resolution dynamic CE‐MRA of the thorax enabled by iterative TWIST reconstruction. Magn Reson Med. 2017;77:833–
- Bathla, G. et al. Deep Learning–Based Reconstruction of 3D T1 SPACE Vessel Wall Imaging Provides Improved Image Quality with Reduced Scan Times: A Preliminary Study. J. Neuroradiol. (2024) doi:10.3174/ajnr.A8382.
- Yaman, Burhaneddin, Seyed Amir Hossein Hosseini, Steen Moeller, Jutta Ellermann, Kâmil Uğurbil, and Mehmet Akçakaya. “Self‐supervised learning of physics‐guided reconstruction neural networks without fully sampled reference data.” Magnetic resonance in medicine 84, no. 6 (2020): 3172-3191.
- Batson, Joshua, and Loic Royer. “Noise2self: Blind denoising by self-supervision.” In International Conference on Machine Learning, pp. 524-533. PMLR, 2019.
Background: Quantitative Magnetic Resonance Imaging (qMRI) aims to estimate tissue-specific parameters beyond conventional image contrast, enabling improved characterization of biological tissue. In particular, MRI relaxometry focuses on the quantification of nuclear magnetic resonance relaxation times, such as T1 and T2. Accurate relaxometry critically depends on the design of the MR pulse sequence and its robustness to confounding effects, including B0 and B1 inhomogeneities.
The Bloch equations allow to calculate the time evolution of the macroscopic magnetization in an MRI experiment. Bloch simulations are an important tool for pulse sequence optimization, sensitivity analysis, and the development of accurate quantification pipelines. However, full Bloch simulations can be computationally expensive, and their accuracy and efficiency strongly depend on solver implementations, model assumptions, and hardware acceleration.
In this project, we systematically evaluate different open-source Bloch simulation frameworks in the context of a previously developed abdominal T1 and T2 mapping sequence. The frameworks are compared under varying modeling assumptions, with the aim of improving MRI acquisition design and quantitative parameter estimation.
Goals:
- To compare different open-source Bloch simulation frameworks with respect to computational efficiency and accuracy for MRI relaxometry
- To analyze the impact of modeling assumptions on the simulation fidelity
- (Optional) To use Bloch simulations to refine and optimize MR pulse sequence design using the open-source Pulseq framework
The student will have the opportunity to:
- Learn about MR pulse sequence design and relaxometry techniques
- Learn about Bloch simulations and numerical solvers
- Contribute directly to ongoing research projects and be integrated in the MRISM lab
- Work on a reseach-oriented software project involving software engineering practices
Requirements:
- Basic knowledge of MR physics.
- Strong programming skills in Python. Julia is an advantage
- Independent working style with a strong problem-solving mindset
- Good command of English.
Duration: Spring Semester 2026
Supervisors: Jonathan Stelter, Prof. Dimitrios Karampinos
How to apply: Please apply through the SB student projects webpage
Background:
Magnetic resonance imaging (MRI) employs arrays of multiple receive coils to improve signal-to-noise ratio and enable accelerated data acquisition. Each receive coil exhibits a spatially varying sensitivity profile that must be accounted when combining multi-coil data for image reconstruction. Parallel imaging techniques exploit these coil sensitivity variations to reduce acquisition time and form the foundation of state-of-the-art accelerated reconstruction methods, including iterative regularized reconstructions and many deep learning–based methods.
Accurate estimation of coil sensitivities and, alternatively, k-space interpolation kernels is critical for robust and artifact-free reconstruction. Self-calibrated methods estimate this information directly from the acquired data using an auto-calibration region in k-space. Prominent examples include low-resolution sensitivity estimation, ESPIRiT, and k-space–based approaches such as GRAPPA.
In this project, we implement different self-calibrated parallel imaging techniques and compare them, with a focus on the fidelity of the estimated coil sensitivity maps and the resulting MR signal phase. The influence of the different sensitivity estimation strategies on the image quality is evaluated within an accelerated iterative reconstruction framework and compared with k-space–based GRAPPA reconstruction.
Goals:
- To (partially) implement and compare different self-calibrated parallel imaging techniques for MR image reconstruction (low-resolution sensitivity estimation, ESPIRiT, GRAPPA)
- To evaluate the influence of different coil sensitivity estimation methods on the image quality of parallel imaging–accelerated reconstructions
The student will have the opportunity to:
- Learn about MR image reconstruction and coil sensitivity estimation
- Learn about parallel imaging acceleration techniques
- Contribute directly to ongoing research projects and be integrated in the MRISM lab
- Work on a reseach-oriented software project involving software engineering practices
Requirements:
- Basic knowledge of MR physics and image processing.
- Strong programming skills in Python. Julia is an advantage
- Independent working style with a strong problem-solving mindset
- Good command of English
Duration: Spring Semester 2026
Supervisors: Jonathan Stelter, Yutong Luo, Prof. Dimitrios Karampinos
How to apply: Please apply through the SB student projects webpage