Deep learning models for bone suppression in chest radiographs

Gusarev, Maxim; Kuleev, Ramil; Khan, Adil; Rivera, Adín Ramírez; Khattak, Asad Masood

doi:10.1109/cibcb.2017.8058543

Cited by 32 publications

(31 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we adopt dilated convolutions to expand the receptive field, but different from the work of Eslami et al, 14 we use gradually increased dilation factors instead of a fixed one. More importantly, downsampling operations are removed after the dilated convolution to overcome the problem of contextual information loss With the rise of Convolutional Neural Networks (CNNs) for image processing and analysis, a number of deep learning models have been developed to solve this problem, 3,10,11,14 given the strong capability of CNNs to model highly nonlinear mapping, such as the case in bone suppression. Many of them are based on an auto-encoder architecture, and optimize the model parameters to minimize pixel-wise differences between the target and predicted x-ray images.…”

Section: A Overviewmentioning

confidence: 99%

“…Many of them are based on an auto-encoder architecture, and optimize the model parameters to minimize pixel-wise differences between the target and predicted x-ray images. 10,11 More recent studies took advantage of generative adversarial network (GAN), 3 which optimizes the image generators by adding discriminators as assisting components during the training session. Moreover, study from Eslami et al 14 showed us an effective receptive field of the generator plays an essential role in improving cGAN for the bone suppression problem.…”

Section: A Overviewmentioning

confidence: 99%

“…Normalized mean squared error (NMSE) is also adopted, where the error is normalized by the signal power, as defined in Eq. (11). The symbols y and x indicate the ground truth and the generated image, respectively.…”

Section: B Evaluation Metricsmentioning

confidence: 99%

See 2 more Smart Citations

Dilated conditional GAN for bone suppression in chest radiographs with enforced semantic features

Zhou

Shen³

2020

Medical Physics

View full text Add to dashboard Cite

Purpose The purpose of this essay is to improve computer‐aided diagnosis of lung diseases by the removal of bone structures imagery such as ribs and clavicles, which may shadow a clinical view of lesions. This paper aims to develop an algorithm to suppress the imaging of bone structures within clinical x‐ray images, leaving a residual portrayal of lung tissue; such that these images can be used to better serve applications, such as lung nodule detection or pathology based on the radiological reading of chest x rays. Methods We propose a conditional Adversarial Generative Network (cGAN) (Mirza and Osindero, Conditional generative adversarial nets, 2014.) model, consisting of a generator and a discriminator, for the task of bone shadow suppression. The proposed model utilizes convolutional operations to expand the size of the receptive field of the generator without losing contextual information while downsampling the image. It is trained by enforcing both the pixel‐wise intensity similarity and the semantic‐level visual similarity between the generated x‐ray images and the ground truth, via optimizing an L‐1 loss of the pixel intensity values on the generator side and a feature matching loss on the discriminator side, respectively. Results The framework we propose is trained and tested on an open‐access chest radiograph dataset for benchmark. Results show that our model is capable of generating bone‐suppressed images of outstanding quality with a limited number of training samples (N = 272). Conclusions Our approach outperforms current state‐of‐the‐art bone suppression methods using x‐ray images. Instead of simply downsampling images at different scales, our proposed method mitigates the loss of contextual information by utilizing dilated convolutions, which gains a noticeable quality improvement for the outputs. On the other hand, our experiment shows that enforcing the semantic similarity between the generated and the ground truth images assists the adversarial training process and achieves better perceptual quality.

show abstract

Section: A Overviewmentioning

confidence: 99%

Section: A Overviewmentioning

confidence: 99%

See 1 more Smart Citation

Dilated conditional GAN for bone suppression in chest radiographs with enforced semantic features

Zhou

Shen³

2020

Medical Physics

View full text Add to dashboard Cite

show abstract

“…In another study, Gusarev et al proposed two deep learning architectures to perform bone suppression and creating a soft tissue image. Considering bones as a noise level that is affecting these chest images [20], they tried to minimize the presence of this noise (i.e., bone) while still preserving the sharpness of the image for the eventual organ segmentation. In [57], many of the noise suppressing methods reviewed where the main objective in all these methods is to remove as much of the noise as possible while preserving most of the relevant details in the image.…”

Section: B Task 2: Bone and Rib Suppressionmentioning

confidence: 99%

Image-to-Images Translation for Multi-Task Organ Segmentation and Bone Suppression in Chest X-Ray Radiography

Eslami

Tabarestani

Albarqouni

et al. 2020

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Chest X-ray radiography is one of the earliest medical imaging technologies and remains one of the most widelyused for the diagnosis, screening and treatment follow up of diseases related to lungs and heart. The literature in this field of research reports many interesting studies dealing with the challenging tasks of bone suppression and organ segmentation but performed separately, limiting any learning that comes with the consolidation of parameters that could optimize both processes. Although image processing could facilitate computer aided diagnosis, machine learning seems more amenable in dealing with the many parameters one would have to contend with to yield an near optimal classification or decision-making process. This study, and for the first time, introduces a multitask deep learning model that generates simultaneously the bonesuppressed image and the organ segmented image, minimizing as a consequence the number of parameters the model has to deal with and optimizing the processing time as well; while at the same time exploiting the interplay in these parameters so as to benefit the performance of both tasks. The design architecture of this model, which relies on a conditional generative adversarial network, reveals the process on how we managed to modify the well-established pix2pix network to fit the need for multitasking and hence extending the standard image-to-image network to the new image-to-images architecture. Dilated convolutions are also used to improve the results through a more effective receptive field assessment. A comparison of the proposed approach to state-of-the-art algorithms is provided to gauge the merits of the proposed approach.

show abstract

“…The k-nearest neighbor regression method with optimized local features has been also proposed to perform the separation of bone and tissue components in chest radiographs [14]. Deep learning models transform standard CXR images into soft-tissue images by treating bones as noise through autoencoder-like model and multi-layer neural model for bone suppression [16]. An improved performance of bone suppression method was obtained by learning the mapping between the gradients of the CXRs and the corresponding bone images [4].…”

Section: Introductionmentioning

confidence: 99%

Bone Suppression of Chest Radiographs With Cascaded Convolutional Networks in Wavelet Domain

et al. 2019

View full text Add to dashboard Cite

Bone suppression of chest radiographs (CXRs) is potentially useful for diagnosing lung diseases for radiologists and computer-aided diagnosis. This paper presents a cascaded convolutional network model in wavelet domain (Wavelet-CCN) for bone suppression in single conventional CXR. Wavelet coefficients are sparse and suitable as the output of convolutional network. The convolutional networks are trained to predict the wavelet coefficients of bone images from the wavelet coefficients of CXRs, using real two-exposure dual energy subtraction (DES) CXRs as training data. By combining the multilevel wavelet decomposition and a cascaded refinement framework, the Wavelet-CCN model can work automatically with a multi-scale approach and progressively refine the prediction in terms of accuracy and spatial resolution. Compared with previous work of CamsNet model which preforms bone prediction in gradient domain, the Wavelet-CCN model predicts the wavelet coefficients to reconstruct bone images and can avoid the inconsistent background intensity caused by 2D integration of gradients. The predicted bone image is subtracted from the original CXR to produce a soft-tissue image. The Wavelet-CCN model and its variants with different wavelet basis are evaluated on a dataset that consists of 504 cases of real two-exposure DES CXRs (404 cases for training and 100 cases for test). Experimental results show that among all the variants and different wavelet bases, the Wavelet-CCN model with Haar wavelet performs best. The average peak signal-to-noise ratio and structural similarity index of the soft-tissue images produced by the proposed Wavelet-CCN model are both higher than those of the previous CamsNet model in gradient domain, reaching values of 39.4 (±0.94) dB and 0.977 (±0.004), respectively. The results also demonstrate that the Wavelet-CCN model can process the CXRs acquired by four types of X-ray machines. INDEX TERMS Bone suppression, convolutional networks, chest radiographs, wavelet transform.

show abstract

Deep learning models for bone suppression in chest radiographs

Cited by 32 publications

References 16 publications

Dilated conditional GAN for bone suppression in chest radiographs with enforced semantic features

Dilated conditional GAN for bone suppression in chest radiographs with enforced semantic features

Image-to-Images Translation for Multi-Task Organ Segmentation and Bone Suppression in Chest X-Ray Radiography

Bone Suppression of Chest Radiographs With Cascaded Convolutional Networks in Wavelet Domain

Contact Info

Product

Resources

About