3D multirater RCNN for multimodal multiclass detection and characterisation of extremely small objects

Carole H. Sudre, Beatriz Gomez Anson, Silvia Ingala, Chris D. Lane, Daniel Jimenez, Lukas Haider, Thomas Varsavsky, Lorna Smith, Sébastien Ourselin, Rolf H. Jäger, M. Jorge Cardoso
Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning, PMLR 102:447-456, 2019.

Abstract

Extremely small objects (ESO) have become observable on clinical routine magnetic resonance imaging acquisitions, thanks to a reduction in acquisition time at higher resolution. Despite their small size (usually <10 voxels per object for an image of more than $10^6$ voxels), these markers reflect tissue damage and need to be accounted for to investigate the complete phenotype of complex pathological pathways. In addition to their very small size, variability in shape and appearance leads to high labelling variability across human raters, resulting in a very noisy gold standard. Such objects are notably present in the context of cerebral small vessel disease where enlarged perivascular spaces and lacunes, commonly observed in the ageing population, are thought to be associated with acceleration of cognitive decline and risk of dementia onset. In this work, we redesign the RCNN model to scale to 3D data, and to jointly detect and characterise these important markers of age-related neurovascular changes. We also propose training strategies enforcing the detection of extremely small objects, ensuring a tractable and stable training process.

Cite this Paper


BibTeX
@InProceedings{pmlr-v102-sudre19a, title = {3D multirater RCNN for multimodal multiclass detection and characterisation of extremely small objects}, author = {Sudre, {Carole H.} and {Gomez Anson}, Beatriz and Ingala, Silvia and Lane, {Chris D.} and Jimenez, Daniel and Haider, Lukas and Varsavsky, Thomas and Smith, Lorna and Ourselin, S{\'e}bastien and { J{\"a}ger}, {Rolf H.} and Cardoso, {M. Jorge}}, booktitle = {Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning}, pages = {447--456}, year = {2019}, editor = {Cardoso, M. Jorge and Feragen, Aasa and Glocker, Ben and Konukoglu, Ender and Oguz, Ipek and Unal, Gozde and Vercauteren, Tom}, volume = {102}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v102/sudre19a/sudre19a.pdf}, url = {https://proceedings.mlr.press/v102/sudre19a.html}, abstract = {Extremely small objects (ESO) have become observable on clinical routine magnetic resonance imaging acquisitions, thanks to a reduction in acquisition time at higher resolution. Despite their small size (usually <10 voxels per object for an image of more than $10^6$ voxels), these markers reflect tissue damage and need to be accounted for to investigate the complete phenotype of complex pathological pathways. In addition to their very small size, variability in shape and appearance leads to high labelling variability across human raters, resulting in a very noisy gold standard. Such objects are notably present in the context of cerebral small vessel disease where enlarged perivascular spaces and lacunes, commonly observed in the ageing population, are thought to be associated with acceleration of cognitive decline and risk of dementia onset. In this work, we redesign the RCNN model to scale to 3D data, and to jointly detect and characterise these important markers of age-related neurovascular changes. We also propose training strategies enforcing the detection of extremely small objects, ensuring a tractable and stable training process.} }
Endnote
%0 Conference Paper %T 3D multirater RCNN for multimodal multiclass detection and characterisation of extremely small objects %A Carole H. Sudre %A Beatriz Gomez Anson %A Silvia Ingala %A Chris D. Lane %A Daniel Jimenez %A Lukas Haider %A Thomas Varsavsky %A Lorna Smith %A Sébastien Ourselin %A Rolf H. Jäger %A M. Jorge Cardoso %B Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2019 %E M. Jorge Cardoso %E Aasa Feragen %E Ben Glocker %E Ender Konukoglu %E Ipek Oguz %E Gozde Unal %E Tom Vercauteren %F pmlr-v102-sudre19a %I PMLR %P 447--456 %U https://proceedings.mlr.press/v102/sudre19a.html %V 102 %X Extremely small objects (ESO) have become observable on clinical routine magnetic resonance imaging acquisitions, thanks to a reduction in acquisition time at higher resolution. Despite their small size (usually <10 voxels per object for an image of more than $10^6$ voxels), these markers reflect tissue damage and need to be accounted for to investigate the complete phenotype of complex pathological pathways. In addition to their very small size, variability in shape and appearance leads to high labelling variability across human raters, resulting in a very noisy gold standard. Such objects are notably present in the context of cerebral small vessel disease where enlarged perivascular spaces and lacunes, commonly observed in the ageing population, are thought to be associated with acceleration of cognitive decline and risk of dementia onset. In this work, we redesign the RCNN model to scale to 3D data, and to jointly detect and characterise these important markers of age-related neurovascular changes. We also propose training strategies enforcing the detection of extremely small objects, ensuring a tractable and stable training process.
APA
Sudre, C.H., Gomez Anson, B., Ingala, S., Lane, C.D., Jimenez, D., Haider, L., Varsavsky, T., Smith, L., Ourselin, S., Jäger, R.H. & Cardoso, M.J.. (2019). 3D multirater RCNN for multimodal multiclass detection and characterisation of extremely small objects. Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 102:447-456 Available from https://proceedings.mlr.press/v102/sudre19a.html.

Related Material