Deep reinforcement learning for cerebral anterior vessel tree extraction from 3D CTA images

Deep reinforcement learning for cerebral anterior vessel tree extraction from 3D CTA images

Su, Jiahang (Erasmus MC)
Li, Shuai (Erasmus MC)
Wolff, Lennard (Erasmus MC)
van Zwam, Wim (Maastricht UMC)
Niessen, W.J. (TU Delft ImPhys/Vos group; TU Delft ImPhys/Computational Imaging; Erasmus MC)
van der Lugt, Aad (Erasmus MC)
van Walsum, T. (TU Delft Biomechanical Engineering; Erasmus MC)

Biomechanical Engineering

2023

Extracting the cerebral anterior vessel tree of patients with an intracranial large vessel occlusion (LVO) is relevant to investigate potential biomarkers that can contribute to treatment decision making. The purpose of our work is to develop a method that can achieve this from routinely acquired computed tomography angiography (CTA) and computed tomography perfusion (CTP) images. To this end, we regard the anterior vessel tree as a set of bifurcations and connected centerlines. The method consists of a proximal policy optimization (PPO) based deep reinforcement learning (DRL) approach for tracking centerlines, a convolutional neural network based bifurcation detector, and a breadth-first vessel tree construction approach taking the tracking and bifurcation detection results as input. We experimentally determine the added values of various components of the tracker. Both DRL vessel tracking and CNN bifurcation detection were assessed in a cross validation experiment using 115 subjects. The anterior vessel tree formation was evaluated on an independent test set of 25 subjects, and compared to interobserver variation on a small subset of images. The DRL tracking result achieves a median overlapping rate until the first error (1.8 mm off the reference standard) of 100, [46, 100] % on 8032 vessels over 115 subjects. The bifurcation detector reaches an average recall and precision of 76% and 87% respectively during the vessel tree formation process. The final vessel tree formation achieves a median recall of 68% and precision of 70%, which is in line with the interobserver agreement.

3D CTA
Bifurcation detection
Brain vessel
CNN
Deep reinforcement learning
Tracking

http://resolver.tudelft.nl/uuid:a45a1aa5-3877-4265-af34-623425eaad8f

https://doi.org/10.1016/j.media.2022.102724

1361-8415

Medical Image Analysis, 84

Institutional Repository

journal article

© 2023 Jiahang Su, Shuai Li, Lennard Wolff, Wim van Zwam, W.J. Niessen, Aad van der Lugt, T. van Walsum