Dynamic multi-scale CNN forest learning for automatic cervical cancer segmentation

Nesrine Bnouni (Lead / Corresponding author), Islem Rekik, Mohamed Salah Rhim, Najoua Essoukri Ben Amara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)
126 Downloads (Pure)

Abstract

Deep-learning based labeling methods have gained unprecedented popularity in different computer vision and medical image segmentation tasks. However, to the best of our knowledge, these have not been used for cervical tumor segmentation. More importantly, while the majority of innovative deep-learning works using convolutional neural networks (CNNs) focus on developing more sophisticated and robust architectures (e.g., ResNet, U-Net, GANs), there is very limited work on how to aggregate different CNN architectures to improve their relational learning at multiple levels of CNN-to-CNN interactions. To address this gap, we introduce a Dynamic Multi-Scale CNN Forest (CK+1DMF), which aims to address three major issues in medical image labeling and ensemble CNN learning: (1) heterogeneous distribution of MRI training patches, (2) a bi-directional flow of information between two consecutive CNNs as opposed to cascading CNNs—where information passes in a directional way from current to the next CNN in the cascade, and (3) multiscale anatomical variability across patients. To solve the first issue, we group training samples into K clusters, then design a forest with (K+ 1) trees: a principal tree of CNNs trained using all data samples and subordinate trees, each trained using a cluster of samples. As for the second and third issues, we design each dynamic multiscale tree (DMT) in the forest such that each node in the tree nests a CNN architecture. Two successive CNN nodes in the tree pass bidirectional contextual maps to progressively improve the learning of their relational non-linear mapping. Besides, as we traverse a path from the root node to a leaf node in the tree, the architecture of each CNN node becomes shallower to take in smaller training patches. Our CK+1DMF significantly (p < 0.05) outperformed several conventional and ensemble CNN architectures, including conventional CNN (improvement by 10.3%) and CNN-based DMT (improvement by 5%).

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging
Subtitle of host publication9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
EditorsMingxia Liu, Heung-Il Suk, Yinghuan Shi
Place of PublicationSwitzerland
PublisherSpringer Verlag
Pages19-27
Number of pages9
Volume11046
ISBN (Electronic)9783030009199
ISBN (Print)9783030009182
DOIs
Publication statusPublished - 2018
Event9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: 16 Sep 201816 Sep 2018

Publication series

NameLecture Notes in Computer Science
PublisherSpringer
Volume11046
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period16/09/1816/09/18

    Fingerprint

Cite this

Bnouni, N., Rekik, I., Rhim, M. S., & Amara, N. E. B. (2018). Dynamic multi-scale CNN forest learning for automatic cervical cancer segmentation. In M. Liu, H-I. Suk, & Y. Shi (Eds.), Machine Learning in Medical Imaging: 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings (Vol. 11046, pp. 19-27). (Lecture Notes in Computer Science; Vol. 11046). Springer Verlag. https://doi.org/10.1007/978-3-030-00919-9_3