Fully Automatic and Real-Time Catheter Segmentation in X-Ray Fluoroscopy

Ambrosini, Pierre; Ruijters, Daniël; Niessen, Wiro J.; Moelker, Adriaan; Walsum, Theo van

doi:10.1007/978-3-319-66185-8_65

Cited by 59 publications

(57 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Consequently, substantial amounts of clinical data must be collected and annotated to enable machine learning for fluoroscopy-guided procedures. Despite clear opportunities, in particular for prediction tasks, very little work has considered learning in this context [4,5,6,7]. A promising approach to tackling the above challenges is in silico fluoroscopy generation from diagnostic 3D CT, most commonly referred to as digitally reconstructed radiographs (DRRs) [4,5].…”

Section: Introductionmentioning

confidence: 99%

DeepDRR – A Catalyst for Machine Learning in Fluoroscopy-Guided Procedures

Unberath

Zaech

Lee

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Machine learning-based approaches outperform competing methods in most disciplines relevant to diagnostic radiology. Interventional radiology, however, has not yet benefited substantially from the advent of deep learning, in particular because of two reasons: 1) Most images acquired during the procedure are never archived and are thus not available for learning, and 2) even if they were available, annotations would be a severe challenge due to the vast amounts of data. When considering fluoroscopy-guided procedures, an interesting alternative to true interventional fluoroscopy is in silico simulation of the procedure from 3D diagnostic CT. In this case, labeling is comparably easy and potentially readily available, yet, the appropriateness of resulting synthetic data is dependent on the forward model. In this work, we propose DeepDRR, a framework for fast and realistic simulation of fluoroscopy and digital radiography from CT scans, tightly integrated with the software platforms native to deep learning. We use machine learning for material decomposition and scatter estimation in 3D and 2D, respectively, combined with analytic forward projection and noise injection to achieve the required performance. On the example of anatomical landmark detection in X-ray images of the pelvis, we demonstrate that machine learning models trained on DeepDRRs generalize to unseen clinically acquired data without the need for re-training or domain adaptation. Our results are promising and promote the establishment of machine learning in fluoroscopy-guided procedures.

show abstract

Section: Introductionmentioning

confidence: 99%

DeepDRR – A Catalyst for Machine Learning in Fluoroscopy-Guided Procedures

Unberath

Zaech

Lee

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Further, in order to demonstrate the effectiveness of the recurrent module, we trained another network termed w/oR with the recurrent module removed under the exact same settings. This network resembles the typical UNet-style network used in [2] .…”

Section: Methodsmentioning

confidence: 99%

“…A very recent work used a fully convolutional neural network for detection of peripherally inserted central catheter (PICC) tip position on adult chest X-ray images [16]. A similar approach was taken by Ambrosini et al [2] to detect catheter in X-ray fluoroscopy but using a UNet-style [18] network. Both methods require human to manually annotate catheter locations for supervised training.…”

Section: Related Workmentioning

confidence: 99%

Automatic Catheter and Tube Detection in Pediatric X-ray Images Using a Scale-Recurrent Network and Synthetic Data

et al. 2019

View full text Add to dashboard Cite

Catheters are commonly inserted life supporting devices. X-ray images are used to assess the position of a catheter immediately after placement as serious complications can arise from malpositioned catheters. Previous computer vision approaches to detect catheters on X-ray images either relied on low-level cues that are not sufficiently robust or only capable of processing a limited number or type of catheters. With the resurgence of deep learning, supervised training approaches are begining to showing promising results. However, dense annotation maps are required, and the work of a human annotator is hard to scale. In this work, we proposed a simple way of synthesizing catheters on X-ray images and a scale recurrent network for catheter detection. By training on adult chest X-rays, the proposed network exhibits promising detection results on pediatric chest/abdomen X-rays in terms of both precision and recall.

show abstract

“…Notably, this method is also computationally expensive and the net response for an image at full scale (512 × 512) took approximately 1000 ms. More importantly, the majority of recent methods usually address the guidewire detection problem only without precise segmentation of the guidewire tip which is necessary for robot control. Recently, Chen and Wang [32] and Ambrosini et al [33] introduced segmentation approaches using a U-net architecture [34] for tracking the guidewire in ultrasound images and X-ray fluoroscopic images, respectively. These methods achieved significant improvements compared to the feature-based methods in terms of guidewire extraction.…”

Section: B Deep Learning-based Methodsmentioning

confidence: 99%

Real-Time Tracking of Guidewire Robot Tips Using Deep Convolutional Neural Networks on Successive Localized Frames

2019

View full text Add to dashboard Cite

Studies are proceeded to stabilize cardiac surgery using thin micro-guidewires and catheter robots. To control the robot to a desired position and pose, it is necessary to accurately track the robot tip in real time but tracking and accurately delineating the thin and small tip is challenging. To address this problem, a novel image analysis-based tracking method using deep convolutional neural networks (CNN) has been proposed in this paper. The proposed tracker consists of two parts; (1) a detection network for rough detection of the tip position and (2) a segmentation network for accurate tip delineation near the tip position. To learn a robust real-time tracker, we extract small image patches, including the tip in successive frames and then learn the informative spatial and motion features for the segmentation network. During inference, the tip bounding box is first estimated in the initial frame via the detection network, thereafter tip delineation is consecutively performed through the segmentation network in the following frames. The proposed method enables accurate delineation of the tip in real time and automatically restarts tracking via the detection network when tracking fails in challenging frames. Experimental results show that the proposed method achieves better tracking accuracy than existing methods, with a considerable real-time speed of 19ms. INDEX TERMS Convolutional neural networks, micro-robot tracking, guidewire tracking, patch-wise segmentation.

show abstract

Fully Automatic and Real-Time Catheter Segmentation in X-Ray Fluoroscopy

Cited by 59 publications

References 17 publications

DeepDRR – A Catalyst for Machine Learning in Fluoroscopy-Guided Procedures

DeepDRR – A Catalyst for Machine Learning in Fluoroscopy-Guided Procedures

Automatic Catheter and Tube Detection in Pediatric X-ray Images Using a Scale-Recurrent Network and Synthetic Data

Real-Time Tracking of Guidewire Robot Tips Using Deep Convolutional Neural Networks on Successive Localized Frames

Contact Info

Product

Resources

About