Title: Automatic part segmentation for full newborn skeleton-articulated geometries using geometric deep learning and 3D point cloud

Authors: Morgane Ferrandini; Duc-Phong Nguyen; Duyen Hien Le-Nguyen; Vi-Do Tran; Hoai-Danh Vo; Tan-Nhu Nguyen; Tien-Tuan Dao

Addresses: Laboratoire de Mécanique, Multiphysique, Multiéchelle, Univ. Lille, CNRS, Centrale Lille, UMR 9013, LaMcube, F-59000 Lille, France ' Centre de Recherche Royallieu, CNRS, Biomechanics and Bioengineering, Université de Technologie de Compiègne, CS 60 319 – 60 203, Compiègne Cedex, France ' Laboratoire de Mécanique, Multiphysique, Multiéchelle, Univ. Lille, CNRS, Centrale Lille, UMR 9013, LaMcube, F-59000 Lille, France ' Ho Chi Minh City University of Technology and Education, Ho Chi Minh City, Vietnam ' University of Science, Ho Chi Minh City, Vietnam; Vietnam National University, Ho Chi Minh City, Vietnam ' School of Biomedical Engineering, International University, Ho Chi Minh City, Vietnam; Vietnam National University, Ho Chi Minh City, Vietnam ' Laboratoire de Mécanique, Multiphysique, Multiéchelle, Univ. Lille, CNRS, Centrale Lille, UMR 9013, LaMcube, F-59000 Lille, France

Abstract: The development of the maternal pelvis model including a detailed foetal model with articulated joints is of great clinical relevance. The objective of the present study is to propose an automatic and fast segmentation workflow of the full newborn skeleton using geometric deep learning. Computed tomography scans of 124 newborn were retrieved and manually segmented. PointNet++, a geometric deep learning algorithm, was trained to perform automatic segmentation on the 3D point clouds of the reconstructed skeletons. This method was compared with the k-means clustering approach. The PointNet++ model provided highly accurate results, with an accuracy of 95.7 ± 4.7% and an IoU of 91.7 ± 7.9%, while k-means clustering provided unsatisfactory results (Accuracy = 74.6 ± 3.7% and IoU = 59.8 ± 4.7%). This study provided a powerful and accurate automatic segmentation workflow for the full newborn skeleton.

Keywords: skeleton part segmentation; newborn bones; geometric deep learning; CT scan; 3D point cloud.

DOI: 10.1504/IJBET.2024.141577

International Journal of Biomedical Engineering and Technology, 2024 Vol.46 No.2, pp.101 - 115

Received: 10 Dec 2023
Accepted: 11 Feb 2024
Published online: 24 Sep 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article