eprintid: 10135778 rev_number: 20 eprint_status: archive userid: 608 dir: disk0/10/13/57/78 datestamp: 2021-10-06 13:17:57 lastmod: 2022-10-11 15:34:27 status_changed: 2021-10-06 13:17:57 type: proceedings_section metadata_visibility: show creators_name: Lin, H creators_name: Zhou, Y creators_name: Slator, P creators_name: Alexander, D title: Generalised Super Resolution for Quantitative MRI Using Self-supervised Mixture of Experts ispublished: pub subjects: MOOR divisions: UCL divisions: B04 divisions: C05 divisions: F48 divisions: F42 keywords: Self supervision, Mixture of experts, Quantitative MRI, Generalised super resolution, Pseudo labels note: This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. abstract: Multi-modal and multi-contrast imaging datasets have diverse voxel-wise intensities. For example, quantitative MRI acquisition protocols are designed specifically to yield multiple images with widely-varying contrast that inform models relating MR signals to tissue characteristics. The large variance across images in such data prevents the use of standard normalisation techniques, making super resolution highly challenging. We propose a novel self-supervised mixture-of-experts (SS-MoE) paradigm for deep neural networks, and hence present a method enabling improved super resolution of data where image intensities are diverse and have large variance. Unlike the conventional MoE that automatically aggregates expert results for each input, we explicitly assign an input to the corresponding expert based on the predictive pseudo error labels in a self-supervised fashion. A new gater module is trained to discriminate the error levels of inputs estimated by Multiscale Quantile Segmentation. We show that our new paradigm reduces the error and improves the robustness when super resolving combined diffusion-relaxometry MRI data from the Super MUDI dataset. Our approach is suitable for a wide range of quantitative MRI techniques, and multi-contrast or multi-modal imaging techniques in general. It could be applied to super resolve images with inadequate resolution, or reduce the scanning time needed to acquire images of the required resolution. The source code and the trained models are available at https://github.com/hongxiangharry/SS-MoE. date: 2021-09-21 date_type: published publisher: Springer official_url: https://doi.org/10.1007/978-3-030-87231-1_5 oa_status: green full_text_type: other language: eng primo: open primo_central: open_green verified: verified_manual elements_id: 1891826 doi: 10.1007/978-3-030-87231-1_5 isbn_13: 978-3-030-87230-4 lyricists_name: Alexander, Daniel lyricists_name: Slator, Patrick lyricists_name: Zhou, Yukun lyricists_id: DALEX06 lyricists_id: PJSLA44 lyricists_id: YZHOH46 actors_name: Slator, Patrick actors_id: PJSLA44 actors_role: owner full_text_status: public series: Lecture Notes in Computer Science (LNCS) publication: In: de Bruijne M. et al. (eds) Medical Image Computing and Computer Assisted Intervention – MICCAI 2021 volume: 12906 place_of_pub: Cham, Switzerland pagerange: 44-54 event_title: Medical Image Computing and Computer Assisted Intervention – MICCAI 2021: 24th International Conference institution: MICCAI 2021 book_title: Medical Image Computing and Computer Assisted Intervention – MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part VI editors_name: De Bruijne, M editors_name: Cattin, PC editors_name: Cotin, S editors_name: Padoy, N editors_name: Speidel, S editors_name: Zheng, Y editors_name: Essert, C citation: Lin, H; Zhou, Y; Slator, P; Alexander, D; (2021) Generalised Super Resolution for Quantitative MRI Using Self-supervised Mixture of Experts. In: De Bruijne, M and Cattin, PC and Cotin, S and Padoy, N and Speidel, S and Zheng, Y and Essert, C, (eds.) Medical Image Computing and Computer Assisted Intervention – MICCAI 2021: 24th International Conference, Strasbourg, France, September 27–October 1, 2021, Proceedings, Part VI. (pp. pp. 44-54). Springer: Cham, Switzerland. Green open access document_url: https://discovery.ucl.ac.uk/id/eprint/10135778/1/MICCAI21-200.pdf