Successful data representation is a fundamental factor in machine learning based medical imaging analysis. Deep Learning (DL) has taken an essential role in robust representation learning. However, the inability of deep models to generalize to unseen data can quickly overfit intricate patterns. Thereby, the importance of implementing strategies to aid deep models in discovering useful priors from data to learn their intrinsic properties. Our model, which we call a dual role network (DRN), uses a dependency maximization approach based on Least Squared Mutual Information (LSMI). LSMI leverages dependency measures to ensure representation invariance and local smoothness. While prior works have used information theory dependency measures like mutual information, these are known to be computationally expensive due to the density estimation step. In contrast, our proposed DRN with LSMI formulation does not require the density estimation step and can be used as an alternative to approximate mutual information. Experiments on the CT based COVID-19 Detection and COVID-19 Severity Detection Challenges of the 2nd COV19D competition [24] demonstrate the effectiveness of our method compared to the baseline method of such competition.
Diaz Berenguer, A , Mukherjee, T , Da, Y , Bossa Bossa, MN , Kvasnytsia, M , Vandemeulebroucke, J , Deligiannis, N & Sahli, H 2023, Representation Learning with Information Theory to Detect COVID-19 and its Severity . in L Karlinsky, T Michaeli & K Nishino (eds), Lecture Notes in Computer Science: Computer Vision – ECCV 2022 Workshops. vol. 13807, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13807 LNCS, Springer, Cham, pp. 605-620.
Diaz Berenguer, A. , Mukherjee, T. , Da, Y. , Bossa Bossa, M. N. , Kvasnytsia, M. , Vandemeulebroucke, J. , Deligiannis, N. , & Sahli, H. (2023). Representation Learning with Information Theory to Detect COVID-19 and its Severity . In L. Karlinsky, T. Michaeli, & K. Nishino (Eds.), Lecture Notes in Computer Science: Computer Vision – ECCV 2022 Workshops (Vol. 13807, pp. 605-620). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 13807 LNCS). Springer, Cham.
@inbook{32ee10bccde244d4ae5bca9ae9bbdfba,
title = " Representation Learning with Information Theory to Detect COVID-19 and its Severity " ,
abstract = " Successful data representation is a fundamental factor in machine learning based medical imaging analysis. Deep Learning (DL) has taken an essential role in robust representation learning. However, the inability of deep models to generalize to unseen data can quickly overfit intricate patterns. Thereby, the importance of implementing strategies to aid deep models in discovering useful priors from data to learn their intrinsic properties. Our model, which we call a dual role network (DRN), uses a dependency maximization approach based on Least Squared Mutual Information (LSMI). LSMI leverages dependency measures to ensure representation invariance and local smoothness. While prior works have used information theory dependency measures like mutual information,these are known to be computationally expensive due to the density estimation step. In contrast, our proposed DRN with LSMI formulation does not require the density estimation step and can be used as an alternative to approximate mutual information. Experiments on the CT based COVID-19 Detection and COVID-19 Severity Detection Challenges of the 2nd COV19D competition [24] demonstrate the effectiveness of our method compared to the baseline method of such competition. " ,
keywords = " Representation learning, mutual information, COVID-19 detection " ,
author = " {Diaz Berenguer}, Abel and Tanmoy Mukherjee and Yifei Da and {Bossa Bossa}, {Mat{'i}as Nicol{'a}s} and Maryna Kvasnytsia and Jef Vandemeulebroucke and Nikos Deligiannis and Hichem Sahli " ,
note = " Funding Information: Acknowledgement. We want to thank the organizers of the 2nd COV19D Competition occurring in the ECCV 2022 Workshop: AI-enabled Medical Image Analysis - Digital Pathology & Radiology/COVID19 for providing access to extensive and high-quality data to benchmark our model. This research has been partially financed by the European Union under the Horizon 2020 Research and Innovation programme under grant agreement 101016131 (ICOVID). Publisher Copyright: { extcopyright} 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG. " ,
year = " 2023 " ,
month = feb,
day = " 23 " ,
doi = " 10.1007/978-3-031-25082-8_41 " ,
language = " English " ,
isbn = " 978-3-031-25081-1 " ,
volume = " 13807 " ,
series = " Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) " ,
publisher = " Springer, Cham " ,
pages = " 605620 " ,
editor = " Leonid Karlinsky and Tomer Michaeli and Ko Nishino " ,
booktitle = " Lecture Notes in Computer Science " ,
}