Automatic View Planning with Multi-scale Deep Reinforcement Learning Agents

By A. Alansary, Loic Le Folgoc, G. Vaillant, Ozan Oktay, Yuanwei Li, Wenjia Bai, Jonathan Passerat-Palmbach, Ricardo Guerrero, Konstantinos Kamnitsas, Benjamin Hou and 4 more. 2018

61 citations recorded by the source corpus.

We propose a fully automatic method to find standardized view planes in 3D image acquisitions. Standard view images are important in clinical practice as they provide a means to perform biometric measurements from similar anatomical regions. These views are often constrained to the native orientation of a 3D image acquisition. Navigating through target anatomy to find the required view plane is tedious and operator-dependent. For this task, we employ a multi-scale reinforcement learning (RL) agent framework and extensively evaluate several Deep Q-Network (DQN) based strategies. RL enables a natural learning paradigm by interaction with the environment, which can be used to mimic experienced operators. We evaluate our results using the distance between the anatomical landmarks and detected planes, and the angles between their normal vector and target. The proposed algorithm is assessed on the mid-sagittal and anterior-posterior commissure planes of brain MRI, and the 4-chamber long-axis plane commonly used in cardiac MRI, achieving accuracy of 1.53mm, 1.98mm and 4.84mm, respectively.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Chao Yu
2. Jiming Liu
3. Shamim Nemati
4. Chen Qin
5. Jinming Duan
6. Huaqi Qiu
7. Daniel Rueckert
8. Wenjia Bai

In Computer Vision and Pattern Recognition

1. Shamim Nemati
2. Chao Yu
3. Jiming Liu
4. Wenjia Bai
5. Walid Abdullah Al
6. Daniel Rueckert
7. Chen Qin
8. Jinming Duan

Citations in the corpus

• Reinforcement Learning in Healthcare: A Survey (2019, methodology): “evaluated different kinds of DRL methods (DQN, Double DQN (DDQN), Duel DQN, and Duel DDQN) [12] for anatomical landmark localization in 3D fetal UI [259], and automatic standard view plane detection [260].”
• Evaluating Reinforcement Learning Agents for Anatomical Landmark Detection (2019, unknown): “Because of the recent success of employing Q-learning in several medical imaging applications [5, 9, 25, 26, 28, 29, 30, 31], we adopt in this paper the common strategy of Q-learning based methods as a solution for the RL problem formulation of landmark detection.”
• Deep Learning for Cardiac Image Segmentation: A Review (2019, background): “, 2016), view planning (Alansary et al., 2018), motion correction (Dangi et al.”
• Partial Policy-Based Reinforcement Learning for Anatomical Landmark Localization in 3D Medical Images (2018, methodology): “[31] employed a multi-scale RL agent to sequentially update the view-plane parameters to obtain the standard view-planes clinically defined for accurate measurements of specific organs.”