Robust automatic detection and removal of fiducial projections in fluoroscopy images: An integrated solution

Zhang, Xuan; Zheng, Guoyan

doi:10.1016/j.medengphy.2008.11.009

Cited by 12 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Image resolution is directly related to the physical size of each pixel in the image, so it could significantly affect the accuracy of the camera calibration . The resolution of the original images is 1560 × 1440, it was reduced by multiple factors of 0.7 to get a variety of realistic resolution levels for intraoperative fluoroscopic images, and the minimum resolution was 375 × 346.…”

Section: Methodsmentioning

confidence: 99%

Influence of the quality of intraoperative fluoroscopic images on the spatial positioning accuracy of a CAOS system

Wang

Zhu

et al. 2018

Robotics Computer Surgery

View full text Add to dashboard Cite

The quality of intraoperative fluoroscopic images is a key issue in the spatial positioning accuracy of a CAOS system. Although the 2 typical positioning methods have very similar mathematical principles, they showed different sensitivities to different image quality parameters. The result of this research may help to create a realistic standard for intraoperative fluoroscopic images for CAOS systems.

show abstract

Section: Methodsmentioning

confidence: 99%

Influence of the quality of intraoperative fluoroscopic images on the spatial positioning accuracy of a CAOS system

Wang

Zhu

et al. 2018

Robotics Computer Surgery

View full text Add to dashboard Cite

show abstract

“…To the best of our knowledge, no prior studies address this application. The most relevant prior studies have either focused on removing small fiducial projections on X-rays 28 or provided inpainting solutions for tasks and anatomies other than spine X-rays (e.g., a chest X-ray inpainting method reported in Sogancioglu et al 29 and Belli et al 30 ).…”

Section: Introductionmentioning

confidence: 99%

Deep learning‐based X‐ray inpainting for improving spinal 2D‐3D registration

Esfandiari

Weidert

Kövesházi

et al. 2021

Robotics Computer Surgery

View full text Add to dashboard Cite

Background Two‐dimensional (2D)‐3D registration is challenging in the presence of implant projections on intraoperative images, which can limit the registration capture range. Here, we investigate the use of deep‐learning‐based inpainting for removing implant projections from the X‐rays to improve the registration performance. Methods We trained deep‐learning‐based inpainting models that can fill in the implant projections on X‐rays. Clinical datasets were collected to evaluate the inpainting based on six image similarity measures. The effect of X‐ray inpainting on capture range of 2D‐3D registration was also evaluated. Results The X‐ray inpainting significantly improved the similarity between the inpainted images and the ground truth. When applying inpainting before the 2D‐3D registration process, we demonstrated significant recovery of the capture range by up to 85%. Conclusion Applying deep‐learning‐based inpainting on X‐ray images masked by implants can markedly improve the capture range of the associated 2D‐3D registration task.

show abstract

“…The preinstalled basic particle-labeling (BPL) method consists of three steps: (1) convert the grayscale or color image to a black and white binary image using a predetermined threshold value, (2) extract all particulate objects in the image, and (3) label particles that fulfill target parameters (size/shape). Template matching/pattern matching is another conventional method that seeks target objects in test images through comparison with a template image (23, 24). Objects in the test image are labeled as targets, if they exhibit similarity to the template as determined by an assigned threshold value.…”

Section: Introductionmentioning

confidence: 99%

Quantification of Calcified Particles in Human Valve Tissue Reveals Asymmetry of Calcific Aortic Valve Disease Development

Yabusaki

Hutcheson

Vyas

et al. 2016

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

Recent studies indicated that small calcified particles observable by scanning electron microscopy (SEM) may initiate calcification in cardiovascular tissues. We hypothesized that if the calcified particles precede gross calcification observed in calcific aortic valve disease (CAVD), they would exhibit a regional asymmetric distribution associated with CAVD development, which always initiates at the base of aortic valve leaflets adjacent to the aortic outflow in a region known as the fibrosa. Testing this hypothesis required counting the calcified particles in histological sections of aortic valve leaflets. SEM images, however, do not provide high contrast between components within images, making the identification and quantification of particles buried within tissue extracellular matrix difficult. We designed a new unique pattern-matching based technique to allow for flexibility in recognizing particles by creating a gap zone in the detection criteria that decreased the influence of non-particle image clutter in determining whether a particle was identified. We developed this flexible pattern particle-labeling (FpPL) technique using synthetic test images and human carotid artery tissue sections. A conventional image particle counting method (preinstalled in ImageJ) did not properly recognize small calcified particles located in noisy images that include complex extracellular matrix structures and other commonly used pattern-matching methods failed to detect the wide variation in size, shape, and brightness exhibited by the particles. Comparative experiments with the ImageJ particle counting method demonstrated that our method detected significantly more (p < 2 × 10−7) particles than the conventional method with significantly fewer (p < 0.0003) false positives and false negatives (p < 0.0003). We then applied the FpPL technique to CAVD leaflets and showed a significant increase in detected particles in the fibrosa at the base of the leaflets (p < 0.0001), supporting our hypothesis. The outcomes of this study are twofold: (1) development of a new image analysis technique that can be adapted to a wide range of applications and (2) acquisition of new insight on potential early mediators of calcification in CAVD.

show abstract

Robust automatic detection and removal of fiducial projections in fluoroscopy images: An integrated solution

Cited by 12 publications

References 41 publications

Influence of the quality of intraoperative fluoroscopic images on the spatial positioning accuracy of a CAOS system

Influence of the quality of intraoperative fluoroscopic images on the spatial positioning accuracy of a CAOS system

Deep learning‐based X‐ray inpainting for improving spinal 2D‐3D registration

Quantification of Calcified Particles in Human Valve Tissue Reveals Asymmetry of Calcific Aortic Valve Disease Development

Contact Info

Product

Resources

About