Conference Paper2022Cited by 1

Cascaded DNNs for detecting the position and orientation of Left Ventricle from 3D CT Scans

Magnus Caspersen, Md Sayed Tanveer, ASM Shihavuddin, MM Mahbubul Syeed, Md Hasan Maruf, Ashraful Amin, Faisal M Uddin

2022 IEEE 14th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP)

IEEE, pp. 1–5

  • Good Health and Well-being — Relevance: 94%
DOI: 10.1109/ivmsp54334.2022.9816249URL

Abstract

Accurate and automated localization of the left ventricle (LV) of heart and identification of its’ orientation are crucial for effective diagnosis of various heart diseases. This work presents a fully automated deep learning based pipeline for generating a short-axis transformation of transversal 3D CT scans. The proposed novel solution consists of three independent convolutional neural networks (CNN) in cascaded sequential phases. In phase one, a binary classification network is used to determine a suitable 2D slice that clearly shows the left ventricle from the 3D image stack. In phase two, a different CNN is used to predict the location of 3 landmarks on the suitable 2D slice. From these landmarks a general orientation of the left ventricle is computed in the xy plane. In the last phase, a third CNN is used to predict the location of 3 landmarks on the orthogonal image. With all 6 points, a 3D orientation of the heart can be calculated. The proposed method performs as good as human accuracy, considering inter-observer variability, and performed equally well in terms of End-Diastolic Volume (EDV) and Left Ventricular Mass (LVM) calculation.

Keywords

  • left ventricular mass
  • cascaded convolutional neural network
  • interobserver variability
  • end-diastolic volume
  • landmark detection
  • three-dimensional computed tomography
  • cardiac image analysis

References

  1. 1.Nicole L. Stout, Daniel Santa Mina, Kathleen D. Lyons, Karen Robb, Julie K. Silver. (2020). A systematic review of rehabilitation and exercise recommendations in oncology guidelines. CA A Cancer Journal for Clinicians, 71(2), 149–175[10.3322/caac.21639]
  2. 2.Christian Payer, Darko Štern, Horst Bischof, Martin Urschler. (2016). Regressing Heatmaps for Multiple Landmark Localization Using CNNs. Lecture notes in computer science, 230–238[10.1007/978-3-319-46723-8_27]
  3. 3.Joseph C. Del Paggio, John S. Berry, Wilma M. Hopman, Elizabeth A. Eisenhauer, Vinay Prasad, Bishal Gyawali, Christopher M. Booth. (2021). Evolution of the Randomized Clinical Trial in the Era of Precision Oncology. JAMA Oncology, 7(5), 728[10.1001/jamaoncol.2021.0379]
  4. 4.Qian Tao, Wenjun Yan, Yuanyuan Wang, Elisabeth H. M. Paiman, Denis P. Shamonin, Pankaj Garg, Sven Plein, Lu Huang, Liming Xia, Marek Sramko, Jarsolav Tintera, Albert de Roos, Hildo J. Lamb, Rob J. van der Geest. (2018). Deep Learning–based Method for Fully Automatic Quantification of Left Ventricle Function from Cine MR Images: A Multivendor, Multicenter Study. Radiology, 290(1), 81–88[10.1148/radiol.2018180513]
  5. 5.Jun Zhang, Mingxia Liu, Dinggang Shen. (2017). Detecting Anatomical Landmarks From Limited Medical Imaging Data Using Two-Stage Task-Oriented Deep Neural Networks. IEEE Transactions on Image Processing, 26(10), 4753–4764[10.1109/tip.2017.2721106]
  6. 6.Gongning Luo, Suyu Dong, Kuanquan Wang, Wangmeng Zuo, Shaodong Cao, Henggui Zhang. (2017). Multi-Views Fusion CNN for Left Ventricular Volumes Estimation on Cardiac MR Images. IEEE Transactions on Biomedical Engineering, 65(9), 1924–1934[10.1109/tbme.2017.2762762]
  7. 7.ASM Shihavuddin, Mohammad Rifat Ahmmad Rashid, Md Hasan Maruf, Muhammad Abul Hasan, Mohammad Asif ul Haq, Ratil H. Ashique, Ahmed Al Mansur. (2021). Image based surface damage detection of renewable energy installations using a unified deep learning approach. Energy Reports, 7, 4566–4576[10.1016/j.egyr.2021.07.045]
  8. 8.Anabel Gómez-Ríos, Siham Tabik, Julián Luengo, A.S.M. Shihavuddin, Francisco Herrera. (2019). Coral species identification with texture or structure images using a two-level classifier based on Convolutional Neural Networks. Knowledge-Based Systems, 184, 104891[10.1016/j.knosys.2019.104891]
  9. 9.Enrique Bermejo, Kei Taniguchi, Yoshinori Ogawa, Rubén Martos, Andrea Valsecchi, Pablo Mesejo, Oscar Ibáñez, Kazuhiko Imaizumi. (2021). Automatic landmark annotation in 3D surface scans of skulls: Methodological proposal and reliability study. Computer Methods and Programs in Biomedicine, 210, 106380[10.1016/j.cmpb.2021.106380]
  10. 10.Amin Suzani, Abtin Rasoulian, Alexander Seitel, Sidney Fels, Robert N. Rohling, Purang Abolmaesumi. (2015). Deep learning for automatic localization, identification, and segmentation of vertebral bodies in volumetric MR images. Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE, 9415, 941514[10.1117/12.2081542]
  11. 11.Jonas Björk, Ulf Strömberg, Annika Rosengren, Kjell Toren, Björn Fagerberg, Anna Grimby-Ekman, Göran M.L. BergströM. (2017). Predicting participation in the population-based Swedish cardiopulmonary bio-image study (SCAPIS) using register data. Scandinavian Journal of Public Health, 45(17_suppl), 45–49[10.1177/1403494817702326]
  12. 12.Mahbubul Syeed, Juho Lindman, Imed Hammouda. (2017). Measuring Perceived Trust in Open Source Software Communities. IFIP advances in information and communication technology, 49–54[10.1007/978-3-319-57735-7_5]
  13. 13.M. M. Mahbubul Syeed, Alexander Lokhman, Tommi Mikkonen, Imed Hammouda. (2015). Pluggable Systems as Architectural Pattern. , 1–6[10.1145/2797433.2797477]

Cited By

  1. 1.Sayeed Sohail. (2024). ADAPT: A Parameter Optimised Segmentation Algorithm for Estimating Seeds. [10.31219/osf.io/uk9m8]
← All publications