?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=Generalised+Super+Resolution+for+Quantitative+MRI+Using+Self-supervised+Mixture+of+Experts&rft.creator=Lin%2C+H&rft.creator=Zhou%2C+Y&rft.creator=Slator%2C+P&rft.creator=Alexander%2C+D&rft.description=Multi-modal+and+multi-contrast+imaging+datasets+have+diverse+voxel-wise+intensities.+For+example%2C+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%2C+making+super+resolution+highly+challenging.+We+propose+a+novel+self-supervised+mixture-of-experts+(SS-MoE)+paradigm+for+deep+neural+networks%2C+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%2C+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%2C+and+multi-contrast+or+multi-modal+imaging+techniques+in+general.+It+could+be+applied+to+super+resolve+images+with+inadequate+resolution%2C+or+reduce+the+scanning+time+needed+to+acquire+images+of+the+required+resolution.+The+source+code+and+the+trained+models+are+available+at+https%3A%2F%2Fgithub.com%2Fhongxiangharry%2FSS-MoE.&rft.subject=Self+supervision%2C+Mixture+of+experts%2C+Quantitative+MRI%2C+Generalised+super+resolution%2C+Pseudo+labels&rft.publisher=Springer&rft.contributor=De+Bruijne%2C+M&rft.contributor=Cattin%2C+PC&rft.contributor=Cotin%2C+S&rft.contributor=Padoy%2C+N&rft.contributor=Speidel%2C+S&rft.contributor=Zheng%2C+Y&rft.contributor=Essert%2C+C&rft.date=2021-09-21&rft.type=Proceedings+paper&rft.publisher=MICCAI+2021&rft.language=eng&rft.source=+++++In%3A+De+Bruijne%2C+M+and+Cattin%2C+PC+and+Cotin%2C+S+and+Padoy%2C+N+and+Speidel%2C+S+and+Zheng%2C+Y+and+Essert%2C+C%2C+(eds.)+Medical+Image+Computing+and+Computer+Assisted+Intervention+%E2%80%93+MICCAI+2021%3A+24th+International+Conference%2C+Strasbourg%2C+France%2C+September+27%E2%80%93October+1%2C+2021%2C+Proceedings%2C+Part+VI.++(pp.+pp.+44-54).++Springer%3A+Cham%2C+Switzerland.+(2021)+++++&rft.format=text&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10135778%2F1%2FMICCAI21-200.pdf&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10135778%2F&rft.rights=open