Multiorgan structures detection using deep convolutional neural networks

Onieva, Jorge Onieva; Serrano, Germán González; Young, Thomas P.; Washko, George R.; Ledesma-Carbayo, María J.; Estépar, Raúl San Jośe

doi:10.1117/12.2293761

Cited by 4 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning-based methods have outperformed many traditional image processing methods (Fu et al 2019), achieving the-start-of-art performances in many image processing tasks such as object detection (Onieva et al 2018, Xu et al 2019, classification (Anthimopoulos et al 2016, Shen et al 2017 and image segmentations (Fu et al 2018b, Cardenas et al 2019, Harms et al 2019, Jeong et al 2019. Recently, a thorough review on deep learning-based registration algorithms was published by Haskins et al who divided the deep learning-based image registration methods into three categories: deep iterative registration, supervised transformation estimation and unsupervised transformation estimation (Haskins et al 2019).…”

Section: Introductionmentioning

confidence: 99%

4D-CT deformable image registration using multiscale unsupervised deep learning

Lei

Wang

et al. 2020

Phys. Med. Biol.

View full text Add to dashboard Cite

Deformable image registration (DIR) of 4D-CT images is important in multiple radiation therapy applications including motion tracking of soft tissue or fiducial markers, target definition, image fusion, dose accumulation and treatment response evaluations. It is very challenging to accurately and quickly register 4D-CT abdominal images due to its large appearance variances and bulky sizes. In this study, we proposed an accurate and fast multi-scale DIR network (MS-DIRNet) for abdominal 4D-CT registration. MS-DIRNet consists of a global network (GlobalNet) and local network (LocalNet). GlobalNet was trained using down-sampled whole image volumes while LocalNet was trained using sampled image patches. MS-DIRNet consists of a generator and a discriminator. The generator was trained to directly predict a deformation vector field (DVF) based on the moving and target images. The generator was implemented using convolutional neural networks with multiple attention gates. The discriminator was trained to differentiate the deformed images from the target images to provide additional DVF regularization. The loss function of MS-DIRNet includes three parts which are image similarity loss, adversarial loss and DVF regularization loss. The MS-DIRNet was trained in a completely unsupervised manner meaning that ground truth DVFs are not needed. Different from traditional DIRs that calculate DVF iteratively, MS-DIRNet is able to calculate the final DVF in a single forward prediction which could significantly expedite the DIR process. The MS-DIRNet was trained and tested on 25 patients' 4D-CT datasets using five-fold cross validation. For registration accuracy evaluation, target registration errors (TREs) of MS-DIRNet were compared to clinically used software. Our results showed that the MS-DIRNet with an average TRE of 1.2 ± 0.8 mm outperformed the commercial software with an average TRE of 2.5 ± 0.8 mm in 4D-CT abdominal DIR, demonstrating the superior performance of our method in fiducial marker tracking and overall soft tissue alignment.

show abstract

Section: Introductionmentioning

confidence: 99%

4D-CT deformable image registration using multiscale unsupervised deep learning

Lei

Wang

et al. 2020

Phys. Med. Biol.

View full text Add to dashboard Cite

show abstract

Cycle GAN-Based Data Augmentation For Multi-Organ Detection In CT Images Via Yolo

Hammami¹,

Friboulet²,

Kéchichian³

2020

2020 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

We propose a deep learning solution to the problem of object detection in 3D CT images, i.e. the localization and classification of multiple structures. Supervised learning methods require large annotated datasets that are usually difficult to acquire. We thus develop a Cycle Generative Adversarial Network (CycleGAN) + You Only Look Once (YOLO) combined method for CT data augmentation using MRI source images to train a YOLO detector. This results in a fast and accurate detection with a mean average distance of 7.95 ± 6.2 mm, which is significantly better than detection without data augmentation. We show that the approach compares favorably to state-of-the-art detection methods for medical images.

show abstract

Data augmentation for multi-organ detection in medical images

Hammami

Friboulet

Kéchichian

2020

2020 Tenth International Conference on Image Processing Theory, Tools and Applications (IPTA)

View full text Add to dashboard Cite

We propose a deep learning solution to the problem of object detection in 3D medical images, i.e. the localization and classification of multiple structures. Supervised learning methods require large annotated datasets that are usually difficult to acquire. We thus develop a Cycle Generative Adversarial Network (CycleGAN) and You Only Look Once (YOLO) combined method for data augmentation from one modality to another via CycleGAN and organ detection from generated images via YOLO. This results in a fast and accurate detection with a mean average distance of 7.95 mm for CT modality and 16.18 mm for MRI modality, which is significantly better than detection without data augmentation. We show that the approach compares favorably to state-of-the-art detection methods for medical images on CT data.

show abstract

Multiorgan structures detection using deep convolutional neural networks

Cited by 4 publications

References 12 publications

4D-CT deformable image registration using multiscale unsupervised deep learning

4D-CT deformable image registration using multiscale unsupervised deep learning

Cycle GAN-Based Data Augmentation For Multi-Organ Detection In CT Images Via Yolo

Data augmentation for multi-organ detection in medical images

Contact Info

Product

Resources

About