Dynamic Multi-scale CNN Forest Learning for Automatic Cervical Cancer Segmentation

  • Nesrine BnouniEmail author
  • Islem Rekik
  • Mohamed Salah Rhim
  • Najoua Essoukri Ben Amara
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11046)


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%).

Supplementary material

473959_1_En_3_MOESM1_ESM.pdf (370 kb)
Supplementary material 1 (pdf 370 KB)


  1. 1.
    World Health Organization: Comprehensive Cervical Cancer Control: A Guide to Essential Practice, 2nd edn, p. 2014. WHO, Geneva (2017)Google Scholar
  2. 2.
    Bnouni, N., Mechi, O., Rekik, I., Rhim, M.S., Essoukri Ben Amara, N.: Semi-automatic lymph node segmentation and classification using cervical cancer MR imaging. In: 2018 4th International Conference on Advanced Technologies for Signal and Image Processing (ATSIP), pp. 1–6. IEEE (2018)Google Scholar
  3. 3.
    Torheim, T., Malinen, E., Hole, K.H., Lund, K.V., Indahl, U.G., Lyng, H., et al.: Autodelineation of cervical cancers using multiparametric magnetic resonance imaging and machine learning. Acta Oncol. 56(6), 806–812 (2017)CrossRefGoogle Scholar
  4. 4.
    Guo, Y., Gao, Y., Shen, D.: Deformable MR prostate segmentation via deep feature learning and sparse patch matching. In: Deep Learning for Medical Image Analysis, pp. 197–222 (2017)CrossRefGoogle Scholar
  5. 5.
    Gao, Y., Shao, Y., Lian, J., Wang, A.Z., Chen, R.C., Shen, D.: Accurate segmentation of CT male pelvic organs via regression-based deformable models and multi-task random forests. IEEE Trans. Med. Imaging 35(6), 1532–1543 (2016)CrossRefGoogle Scholar
  6. 6.
    Choi, J.Y., Kim, D.H., Plataniotis, K.N., Ro, Y.M.: Classifier ensemble generation and selection with multiple feature representations for classification applications in computer-aided detection and diagnosis on mammography. Expert Syst. Appl. 46, 106–121 (2016)CrossRefGoogle Scholar
  7. 7.
    Bahrami, Khosro, Shi, Feng, Rekik, Islem, Shen, Dinggang: Convolutional neural network for reconstruction of 7T-like images from 3T MRI using appearance and anatomical features. In: Carneiro, Gustavo, Mateus, Diana, Peter, Loïc, Bradley, Andrew, Tavares, João Manuel R.S., Belagiannis, Vasileios, Papa, João Paulo, Nascimento, Jacinto C., Loog, Marco, Lu, Zhi, Cardoso, Jaime S., Cornebise, Julien (eds.) LABELS/DLMIA -2016. LNCS, vol. 10008, pp. 39–47. Springer, Cham (2016). Scholar
  8. 8.
    Alansary, A., et al.: Fast fully automatic segmentation of the human placenta from motion corrupted MRI. In: International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 589–597. Springer, Cham (2016)CrossRefGoogle Scholar
  9. 9.
    Bahrami, K., Rekik, I., Shi, F., Shen, D.: Joint reconstruction and segmentation of 7T-like MR images from 3T MRI based on cascaded convolutional neural networks. In: International Conference on Medical Image Computing and Computer-Assisted Intervention, pp. 764–772. Springer, Cham (2017)CrossRefGoogle Scholar
  10. 10.
    Wolterink, J.M., Leiner, T., de Vos, B.D., van Hamersvelt, R.W., Viergever, M.A., Išgum, I.: Automatic coronary artery calcium scoring in cardiac CT angiography using paired convolutional neural networks. Med. Image Anal. 34, 123–136 (2016)CrossRefGoogle Scholar
  11. 11.
    Havaei, M., Davy, A., Warde-Farley, D., Biard, A., Courville, A., Bengio, Y., et al.: Brain tumor segmentation with deep neural networks. Med. Image Anal. 35, 18–31 (2017)CrossRefGoogle Scholar
  12. 12.
    Kamnitsas, K., et al. (2017). Ensembles of multiple models and architectures for robust brain tumour segmentation. arXiv preprint arXiv:1711.01468
  13. 13.
    Bnouni, N., Ben Amor, H., Rekik, I., Rhim, M. S., Solaiman, B., Essoukri Ben Amara, N.: Boosting CNN learning by ensemble image preprocessing methods for cervical cancer MR image segmentation. In: International Conference on Sensors, Systems Signals and Advanced Technologies (SSS) (2018)Google Scholar
  14. 14.
    Dolz, J., Desrosiers, C., Wang, L., Yuan, J., Shen, D., Ayed, I.B.: Deep CNN ensembles and suggestive annotations for infant brain MRI segmentation. arXiv preprint arXiv:1712.05319(2017)
  15. 15.
    Ju, C., Bibaut, A., van der Laan, M.J.: The relative performance of ensemble methods with deep convolutional neural networks for image classification. arXiv preprint arXiv:1704.01664 (2017)
  16. 16.
    Guan, Y., Plötz, T.: Ensembles of deep lstm learners for activity recognition using wearables. Proc. ACM Interact. Mob. Wearable Ubiquitous Technol. 1(2), 11 (2017)CrossRefGoogle Scholar
  17. 17.
    Amiri, S., Mahjoub, M.A., Rekik, I.: Tree-based ensemble classifier learning for automatic brain glioma segmentation. Neurocomputing (2018). ISSN:0925-2312Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  • Nesrine Bnouni
    • 1
    • 2
    Email author
  • Islem Rekik
    • 2
  • Mohamed Salah Rhim
    • 3
  • Najoua Essoukri Ben Amara
    • 1
  1. 1.LATIS- Laboratory of Advanced Technology and Intelligent Systems, ENISoSousse UniversitySousseTunisia
  2. 2.BASIRA Lab, CVIP, School of Science and Engineering, ComputingUniversity of DundeeDundeeUK
  3. 3.Department of Gynecology Obstetrics, Faculty of Medicine of MonastirMonastirTunisia

Personalised recommendations