Deep learning for 3D vascular segmentation in hierarchical phase contrast tomography: a case study on kidney

Scientific Reports, vol. 14, pp. 27258

Abstract

Automated blood vessel segmentation is critical for biomedical image analysis, as vessel morphology changes are associated with numerous pathologies. Still, precise segmentation is difficult due to the complexity of vascular structures, anatomical variations across patients, the scarcity of annotated public datasets, and the quality of images. Our goal is to provide a foundation on the topic and identify a robust baseline model for application to vascular segmentation using a new imaging modality, Hierarchical Phase-Contrast Tomography (HiP-CT). We begin with an extensive review of current machine-learning approaches for vascular segmentation across various organs. Our work introduces a meticulously curated training dataset, verified by double annotators, consisting of vascular data from three kidneys imaged using HiP-CT as part of the Human Organ Atlas Project. HiP-CT pioneered at the European Synchrotron Radiation Facility in 2020, revolutionizes 3D organ imaging by offering a resolution of around 20 μm/voxel and enabling highly detailed localised zooms up to 1-2 μm/voxel without physical sectioning. We leverage the nnU-Net framework to evaluate model performance on this high-resolution dataset, using both known and novel samples, and implementing metrics tailored for vascular structures. Our comprehensive review and empirical analysis on HiP-CT data sets a new standard for evaluating machine learning models in high-resolution organ imaging. Our three experiments yielded Dice similarity coefficient (DSC) scores of 0.9523, 0.9410, and 0.8585, respectively. Nevertheless, DSC primarily assesses voxel-to-voxel concordance, overlooking several crucial characteristics of the vessels and should not be the sole metric for deciding the performance of vascular segmentation. Our results show that while segmentations yielded reasonably high scores-such as centerline DSC ranging from 0.82 to 0.88, certain errors persisted. Specifically, large vessels that collapsed due to the lack of hydrostatic pressure (HiP-CT is an ex vivo technique) were segmented poorly. Moreover, decreased connectivity in finer vessels and higher segmentation errors at vessel boundaries were observed. Such errors, particularly in significant vessels, obstruct the understanding of the structures by interrupting vascular tree connectivity. Our study establishes the benchmark across various evaluation metrics, for vascular segmentation of HiP-CT imaging data, an imaging technology that has the potential to substantively shift our understanding of human vascular networks.

Authors 13

1. Ekin Yağış corresponding

   University College London

   Affiliation as printed

   Department of Mechanical Engineering, University College London, London, UK. e.yagis@ucl.ac.uk

   Department of Mechanical Engineering, University College London, London, UK

2. Shahab Aslani

   University College London

   Affiliation as printed

   Centre for Medical Image Computing, University College London, London, UK

   Department of Mechanical Engineering, University College London, London, UK

3. Yashvardhan Jain

   Indiana University Bloomington · Indiana University

   Affiliation as printed

   Department of Intelligent Systems Engineering, Luddy School of Informatics, Computing, and Engineering, Indiana University, Bloomington, USA

4. Yang Zhou

   University College London

   Affiliation as printed

   Department of Mechanical Engineering, University College London, London, UK

5. Shahrokh Rahmani

   University College London · Imperial College London

   Affiliation as printed

   Department of Mechanical Engineering, University College London, London, UK

   National Heart and Lung Institute, Faculty of Medicine, Imperial College London, London, UK

6. Joseph Brunet

   European Synchrotron Radiation Facility · University College London

   Affiliation as printed

   Department of Mechanical Engineering, University College London, London, UK

   European Synchrotron Radiation Facility, Grenoble, France

7. Alexandre Bellier
   Affiliation as printed

   Laboratoire d'Anatomie Des Alpes Françaises, Grenoble, France

8. Christopher Werlein

   Medizinische Hochschule Hannover

   Affiliation as printed

   Institute of Pathology, Hannover Medical School, Carl-Neuberg-Straße 1, 30625, Hannover, Germany

9. Maximilian Ackermann

   University Medical Center of the Johannes Gutenberg University Mainz

   Affiliation as printed

   Institute of Functional and Clinical Anatomy, University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany

10. Danny David Jonigk Aachen Institute of Pathology

    RWTH Aachen University

    Affiliation as printed

    Institute of Pathology, RWTH Aachen University, Pauwelsstrasse 30, 52074, Aachen, Germany

11. Paul Tafforeau

    European Synchrotron Radiation Facility

    Affiliation as printed

    European Synchrotron Radiation Facility, Grenoble, France

12. Peter David Lee

    University College London

    Affiliation as printed

    Department of Mechanical Engineering, University College London, London, UK

13. Claire Walsh

    University College London

    Affiliation as printed

    Department of Mechanical Engineering, University College London, London, UK