Consecutive Context Perceive Generative Adversarial Networks for Serial Sections Inpainting

Zhang, Siqi; Wang, Lei; Zhang, Jie; Gu, Ling; Jiang, Xiran; Zhai, Xuguang; Sha, Xianzheng; Chang, Shijie

doi:10.1109/access.2020.3031973

Cited by 6 publications

(2 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The task to synthesize missing image slices itself is also novel. Neural networks have been used to inpaint a missing patch inside a 2D medical image slice, [28][29][30] which is a considerably less challenging problem that is analogous to interpolation with known boundary conditions around the missing patch. In contrast, synthesizing data in a cropped image is analogous to extrapolation with undefined boundary conditions.…”

Section: Discussionmentioning

confidence: 99%

Using neural networks to extend cropped medical images for deformable registration among images with differing scan extents

et al. 2021

View full text Add to dashboard Cite

Purpose Missing or discrepant imaging volume is a common challenge in deformable image registration (DIR). To minimize the adverse impact, we train a neural network to synthesize cropped portions of head and neck CT's and then test its use in DIR. Methods Using a training dataset of 409 head and neck CT's, we trained a generative adversarial network to take in a cropped 3D image and output an image with synthesized anatomy in the cropped region. The network used a 3D U‐Net generator along with Visual Geometry Group (VGG) deep feature losses. To test our technique, for each of the 53 test volumes, we used Elastix to deformably register combinations of a randomly cropped, full, and synthetically full volume to a single cropped, full, and synthetically full target volume. We additionally tested our method's robustness to crop extent by progressively increasing the amount of cropping, synthesizing the missing anatomy using our network, and then performing the same registration combinations. Registration performance was measured using 95% Hausdorff distance across 16 contours. Results We successfully trained a network to synthesize missing anatomy in superiorly and inferiorly cropped images. The network can estimate large regions in an incomplete image, far from the cropping boundary. Registration using our estimated full images was not significantly different from registration using the original full images. The average contour matching error for full image registration was 9.9 mm, whereas our method was 11.6, 12.1, and 13.6 mm for synthesized‐to‐full, full‐to‐synthesized, and synthesized‐to‐synthesized registrations, respectively. In comparison, registration using the cropped images had errors of 31.7 mm and higher. Plotting the registered image contour error as a function of initial preregistered error shows that our method is robust to registration difficulty. Synthesized‐to‐full registration was statistically independent of cropping extent up to 18.7 cm superiorly cropped. Synthesized‐to‐synthesized registration was nearly independent, with a −0.04 mm of change in average contour error for every additional millimeter of cropping. Conclusions Different or inadequate in scan extent is a major cause of DIR inaccuracies. We address this challenge by training a neural network to complete cropped 3D images. We show that with image completion, the source of DIR inaccuracy is eliminated, and the method is robust to varying crop extent.

show abstract

Section: Discussionmentioning

confidence: 99%

Using neural networks to extend cropped medical images for deformable registration among images with differing scan extents

et al. 2021

View full text Add to dashboard Cite

show abstract

“…These difficulties have recently received a lot of attention in the field of medical image inpainting. Local deformations in medical modalities are widespread due to numerous causes such as metallic implants, foreign objects, or specular reflections during image acquisitions; hence inpainting techniques are becoming more popular in medical image analysis [8][9][10]. Completing these missing or distorted regions is critical for improving post-processing tasks like segmentation and classification.…”

Section: -2508/ C Audt 2023•http://wwwaudtorgmentioning

confidence: 99%

Review on Image Inpainting using Intelligence Mining Techniques

V.,

MPhil,

et al. 2023

Advanced Ultrasound in Diagnosis and Therapy

View full text Add to dashboard Cite

Inpainting is a technique for fixing or removing undesired areas of an image. Methods:In present scenario, image plays a vital role in every aspect such as business images, satellite images, and medical images and so on. Results and Conclusion:This paper presents a comprehensive review of past traditional image inpainting methods and the present state-of-the-art deep learning methods and also detailed the strengths and weaknesses of each to provide new insights in the field.

show abstract

Recent Methods on Medical Image Inpainting and Multi-task Learning Using Deep Learning Models

Rakibe,

Patil

2024

Smart Innovation, Systems and Technologies

View full text Add to dashboard Cite

Consecutive Context Perceive Generative Adversarial Networks for Serial Sections Inpainting

Cited by 6 publications

References 41 publications

Using neural networks to extend cropped medical images for deformable registration among images with differing scan extents

Using neural networks to extend cropped medical images for deformable registration among images with differing scan extents

Review on Image Inpainting using Intelligence Mining Techniques

Recent Methods on Medical Image Inpainting and Multi-task Learning Using Deep Learning Models

Contact Info

Product

Resources

About