Deep EndoVO: A Recurrent Convolutional Neural Network (RCNN) based Visual Odometry Approach for Endoscopic Capsule Robots
This work addresses the need for reliable pose estimation in minimally invasive diagnostic capsule robots, which is incremental as it applies existing deep learning techniques to a specific domain.
The authors tackled the problem of real-time pose estimation for actively controlled endoscopic capsule robots by proposing a monocular visual odometry method using deep Recurrent Convolutional Neural Networks (RCNNs), achieving high translational and rotational accuracies on a real pig stomach dataset.
Ingestible wireless capsule endoscopy is an emerging minimally invasive diagnostic technology for inspection of the GI tract and diagnosis of a wide range of diseases and pathologies. Medical device companies and many research groups have recently made substantial progresses in converting passive capsule endoscopes to active capsule robots, enabling more accurate, precise, and intuitive detection of the location and size of the diseased areas. Since a reliable real time pose estimation functionality is crucial for actively controlled endoscopic capsule robots, in this study, we propose a monocular visual odometry (VO) method for endoscopic capsule robot operations. Our method lies on the application of the deep Recurrent Convolutional Neural Networks (RCNNs) for the visual odometry task, where Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) are used for the feature extraction and inference of dynamics across the frames, respectively. Detailed analyses and evaluations made on a real pig stomach dataset proves that our system achieves high translational and rotational accuracies for different types of endoscopic capsule robot trajectories.