GAMBAS: Generalised-Hilbert Mamba for Super-resolution of Paediatric Ultra-Low-Field MRI

Levente Baljer, Ula Briski, Robert Leech, Niall J Bourke, Kirsten A Donald, Layla E Bradford, Simone R Williams, Sadia Parkar, Sidra Kaleem, Salman Osmani, Sean CL Deoni, Steven CR Williams, Rosalyn J Moran, Emma C. Robinson, František Váša
Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, PMLR 301:82-99, 2026.

Abstract

Magnetic resonance imaging (MRI) is critical for neurodevelopmental research, however access to high-field (HF) systems in low- and middle-income countries is severely hindered by their cost. Ultra-low-field (ULF) systems mitigate such issues of access inequality, however their diminished signal-to-noise ratio limits their applicability for research and clinical use. Deep-learning approaches can enhance the quality of scans acquired at lower field strengths at no additional cost. For example, Convolutional neural networks (CNNs) fused with transformer modules have demonstrated a remarkable ability to capture both local information and long-range context. Unfortunately, the quadratic complexity of transformers leads to an undesirable trade-off between long-range sensitivity and local precision. We propose a hybrid CNN and state-space model (SSM) architecture featuring a novel 3D to 1D serialisation (GAMBAS), which learns long-range context without sacrificing spatial precision. We exhibit improved performance compared to other state-of-the-art medical image-to-image translation models. Our code is made publicly available at https://github.com/levente-1/GAMBAS.

Cite this Paper

BibTeX
@InProceedings{pmlr-v301-baljer26a, title = {GAMBAS: Generalised-Hilbert Mamba for Super-resolution of Paediatric Ultra-Low-Field MRI}, author = {Baljer, Levente and Briski, Ula and Leech, Robert and Bourke, Niall J and Donald, Kirsten A and Bradford, Layla E and Williams, Simone R and Parkar, Sadia and Kaleem, Sidra and Osmani, Salman and Deoni, Sean CL and Williams, Steven CR and Moran, Rosalyn J and Robinson, Emma C. and V\'a\v{s}a, Franti\v{s}ek}, booktitle = {Proceedings of The 8th International Conference on Medical Imaging with Deep Learning}, pages = {82--99}, year = {2026}, editor = {Tasdizen, Tolga and Elhabian, Shireen and Summers, Ronald and Chen, Chen and Koch, Lisa and Zhuang, Yan}, volume = {301}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v301/main/assets/baljer26a/baljer26a.pdf}, url = {https://proceedings.mlr.press/v301/baljer26a.html}, abstract = {Magnetic resonance imaging (MRI) is critical for neurodevelopmental research, however access to high-field (HF) systems in low- and middle-income countries is severely hindered by their cost. Ultra-low-field (ULF) systems mitigate such issues of access inequality, however their diminished signal-to-noise ratio limits their applicability for research and clinical use. Deep-learning approaches can enhance the quality of scans acquired at lower field strengths at no additional cost. For example, Convolutional neural networks (CNNs) fused with transformer modules have demonstrated a remarkable ability to capture both local information and long-range context. Unfortunately, the quadratic complexity of transformers leads to an undesirable trade-off between long-range sensitivity and local precision. We propose a hybrid CNN and state-space model (SSM) architecture featuring a novel 3D to 1D serialisation (GAMBAS), which learns long-range context without sacrificing spatial precision. We exhibit improved performance compared to other state-of-the-art medical image-to-image translation models. Our code is made publicly available at https://github.com/levente-1/GAMBAS.} }
Endnote
%0 Conference Paper %T GAMBAS: Generalised-Hilbert Mamba for Super-resolution of Paediatric Ultra-Low-Field MRI %A Levente Baljer %A Ula Briski %A Robert Leech %A Niall J Bourke %A Kirsten A Donald %A Layla E Bradford %A Simone R Williams %A Sadia Parkar %A Sidra Kaleem %A Salman Osmani %A Sean CL Deoni %A Steven CR Williams %A Rosalyn J Moran %A Emma C. Robinson %A František Váša %B Proceedings of The 8th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Tolga Tasdizen %E Shireen Elhabian %E Ronald Summers %E Chen Chen %E Lisa Koch %E Yan Zhuang %F pmlr-v301-baljer26a %I PMLR %P 82--99 %U https://proceedings.mlr.press/v301/baljer26a.html %V 301 %X Magnetic resonance imaging (MRI) is critical for neurodevelopmental research, however access to high-field (HF) systems in low- and middle-income countries is severely hindered by their cost. Ultra-low-field (ULF) systems mitigate such issues of access inequality, however their diminished signal-to-noise ratio limits their applicability for research and clinical use. Deep-learning approaches can enhance the quality of scans acquired at lower field strengths at no additional cost. For example, Convolutional neural networks (CNNs) fused with transformer modules have demonstrated a remarkable ability to capture both local information and long-range context. Unfortunately, the quadratic complexity of transformers leads to an undesirable trade-off between long-range sensitivity and local precision. We propose a hybrid CNN and state-space model (SSM) architecture featuring a novel 3D to 1D serialisation (GAMBAS), which learns long-range context without sacrificing spatial precision. We exhibit improved performance compared to other state-of-the-art medical image-to-image translation models. Our code is made publicly available at https://github.com/levente-1/GAMBAS.
APA
Baljer, L., Briski, U., Leech, R., Bourke, N.J., Donald, K.A., Bradford, L.E., Williams, S.R., Parkar, S., Kaleem, S., Osmani, S., Deoni, S.C., Williams, S.C., Moran, R.J., Robinson, E.C. & Váša, F.. (2026). GAMBAS: Generalised-Hilbert Mamba for Super-resolution of Paediatric Ultra-Low-Field MRI. Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 301:82-99 Available from https://proceedings.mlr.press/v301/baljer26a.html.

Related Material