arXiv subjects

Ge Wang

Publications and source records attributed to Ge Wang.

At least 253 records · Page 14Linked to original sources

arXiv2018

Quantitative MRI: Absolute T1, T2 and Proton Density Parameters from Deep Learning

Quantitative MRI is highly desirable in terms of intrinsic tissue parameters such as T1, T2 and proton density. This approach promises to minimize diagnostic variability and differentiate normal and pathological tissues by comparing tissue parameters to the normal ranges. Also, absolute quantification can help segment MRI tissue images with better accuracy compared to traditional qualitative segmentation methods. Currently, there are several methods proposed to quantify tissue parameters; however, all of them require excessive scan time and thus are difficult to be applied in clinical applications. In this paper, we propose a novel machine learning approach for MRI quantification, which can dramatically decrease the scan time and improve image quality.

physics.med-ph↗

arXiv2018

A Synergized Pulsing-Imaging Network (SPIN)

Currently, the deep neural network is the mainstream for machine learning, and being actively developed for biomedical imaging applications with an increasing emphasis on tomographic reconstruction for MRI, CT, and other imaging modalities. Multiple deep-learning-based approaches were applied to MRI image reconstruction from k-space samples to final images. Each of these studies assumes a given pulse sequence that produces incomplete and/or inconsistent data in the Fourier space, and targets a trained neural network that recovers an underlying image as close as possible to the ground truth. For the first time, in this paper we view data acquisition and the image reconstruction as the two key parts of an integrated MRI process, and optimize both the pulse sequence and the reconstruction scheme seamlessly in the machine learning framework. Our pilot simulation results show an exemplary embodiment of our new MRI strategy. Clearly, this work can be extended to other imaging modalities and their combinations as well, such as ultrasound imaging, and also potentially simultaneous emission-transmission tomography aided by polarized radiotracers.

physics.med-ph↗

arXiv2018

Projection Decomposition for Dual-energy Computed Tomography

Dual-energy computed tomography (CT) is to reconstruct images of an object from two projection datasets generated from two distinct x-ray source energy spectra. It can provide more accurate attenuation quantification than conventional CT with a single x-ray energy spectrum. In the diagnostic energy range, x-ray energy-dependent attenuation can be approximated as a linear combination of photoelectric absorption and Compton scattering. Hence, two physical components of x-ray attenuation can be determined from two spectral informative projection datasets to achieve monochromatic imaging and material decomposition. In this paper, a projection decomposition method is proposed for the image reconstruction in dual-energy CT. This method combines both an analytical algorithm and a single-variable optimization method to solve the non-linear polychromatic x-ray integral model, allowing accurate quantification of photoelectric absorption and Compton scattering components. Numerical tests are performed to illustrate the merit of the proposed method by comparing with classical projection decomposition methods.

physics.med-ph↗

arXiv2018

Neural-networks-based Photon-Counting Data Correction: Pulse Pileup Effect

Compared with the start-of-art energy integration detectors (EIDs), photon-counting detectors (PCDs) with energy discrimination capabilities have demonstrated great potentials in various applications of medical x-ray radiography and computed tomography (CT). However, the advantages of PCDs may be compromised by various hardware-related degradation factors. For example, at high count rate, quasi-coincident photons may be piled up and result in not only lost counts but also distorted spectrums, which is called the pulse pileup effects. Considering the relative slow detection speed of current PCDs and high x-ray flux for clinical imaging, it is vital to develop an effective approach to compensate or correct for the pileup effect. The aim of this paper was to develop a model-free, end-to-end, and data-driven approach in the neural network framework. We first introduce the trigger threshold concept to regularize the correction process. Then, we design a fully-connected cascade artificial neural network and train it with measurements and true counts. After training, the network is used to recover raw photon-counting data for higher fidelity. To evaluate the proposed approach, Monte Carlo simulations are conducted. We simulate four types of PCDs combining the unipolar or bipolar pulse shape with the nonparalyzable or paralyzable detection mode. The simulation results demonstrate that the trained neural network decreases the relative error of count averages of raw data by an order of magnitude while maintaining a reasonable variance over a wide range of attenuation paths.

physics.med-ph↗

arXiv2018

3D Convolutional Encoder-Decoder Network for Low-Dose CT via Transfer Learning from a 2D Trained Network

Low-dose computed tomography (CT) has attracted a major attention in the medical imaging field, since CT-associated x-ray radiation carries health risks for patients. The reduction of CT radiation dose, however, compromises the signal-to-noise ratio, and may compromise the image quality and the diagnostic performance. Recently, deep-learning-based algorithms have achieved promising results in low-dose CT denoising, especially convolutional neural network (CNN) and generative adversarial network (GAN). This article introduces a Contracting Path-based Convolutional Encoder-decoder (CPCE) network in 2D and 3D configurations within the GAN framework for low-dose CT denoising. A novel feature of our approach is that an initial 3D CPCE denoising model can be directly obtained by extending a trained 2D CNN and then fine-tuned to incorporate 3D spatial information from adjacent slices. Based on the transfer learning from 2D to 3D, the 3D network converges faster and achieves a better denoising performance than that trained from scratch. By comparing the CPCE with recently published methods based on the simulated Mayo dataset and the real MGH dataset, we demonstrate that the 3D CPCE denoising model has a better performance, suppressing image noise and preserving subtle structures.

cs.CV↗

arXiv2018

Low Dose CT Image Denoising Using a Generative Adversarial Network with Wasserstein Distance and Perceptual Loss

In this paper, we introduce a new CT image denoising method based on the generative adversarial network (GAN) with Wasserstein distance and perceptual similarity. The Wasserstein distance is a key concept of the optimal transform theory, and promises to improve the performance of the GAN. The perceptual loss compares the perceptual features of a denoised output against those of the ground truth in an established feature space, while the GAN helps migrate the data noise distribution from strong to weak. Therefore, our proposed method transfers our knowledge of visual perception to the image denoising task, is capable of not only reducing the image noise level but also keeping the critical information at the same time. Promising results have been obtained in our experiments with clinical CT images.

cs.CV↗

arXiv2018

LEARN: Learned Experts' Assessment-based Reconstruction Network for Sparse-data CT

Compressive sensing (CS) has proved effective for tomographic reconstruction from sparsely collected data or under-sampled measurements, which are practically important for few-view CT, tomosynthesis, interior tomography, and so on. To perform sparse-data CT, the iterative reconstruction commonly use regularizers in the CS framework. Currently, how to choose the parameters adaptively for regularization is a major open problem. In this paper, inspired by the idea of machine learning especially deep learning, we unfold a state-of-the-art "fields of experts" based iterative reconstruction scheme up to a number of iterations for data-driven training, construct a Learned Experts' Assessment-based Reconstruction Network ("LEARN") for sparse-data CT, and demonstrate the feasibility and merits of our LEARN network. The experimental results with our proposed LEARN network produces a competitive performance with the well-known Mayo Clinic Low-Dose Challenge Dataset relative to several state-of-the-art methods, in terms of artifact reduction, feature preservation, and computational speed. This is consistent to our insight that because all the regularization terms and parameters used in the iterative reconstruction are now learned from the training data, our LEARN network utilizes application-oriented knowledge more effectively and recovers underlying images more favorably than competing algorithms. Also, the number of layers in the LEARN network is only 12, reducing the computational complexity of typical iterative algorithms by orders of magnitude.

physics.med-ph↗

arXiv2018

Monochromatic CT Image Reconstruction from Current-Integrating Data via Deep Learning

In clinical CT, the x-ray source emits polychromatic x-rays, which are detected in the current-integrating mode. This physical process is accurately described by an energy-dependent non-linear integral model on the basis of the Beer-Lambert law. However, the non-linear model is too complicated to be directly solved for the image reconstruction, and is often approximated to a linear integral model in the form of the Radon transform, basically ignoring energy-dependent information. This model approximation would generate inaccurate quantification of attenuation image and significant beam-hardening artifacts. In this paper, we develop a deep-learning-based CT image reconstruction method to address the mismatch of computing model to physical model. Our method learns a nonlinear transformation from big data to correct measured projection data to accurately match the linear integral model, realize monochromatic imaging and overcome beam hardening effectively. The deep-learning network is trained and tested using clinical dual-energy dataset to demonstrate the feasibility of the proposed methodology. Results show that the proposed method can achieve a high accuracy of the projection correction with a relative error of less than 0.2%.

physics.med-ph↗

arXiv2018

Learning from Pseudo-Randomness With an Artificial Neural Network - Does God Play Pseudo-Dice?

Inspired by the fact that the neural network, as the mainstream for machine learning, has brought successes in many application areas, here we propose to use this approach for decoding hidden correlation among pseudo-random data and predicting events accordingly. With a simple neural network structure and a typical training procedure, we demonstrate the learning and prediction power of the neural network in extremely random environment. Finally, we postulate that the high sensitivity and efficiency of the neural network may allow to critically test if there could be any fundamental difference between quantum randomness and pseudo randomness, which is equivalent to the question: Does God play dice?

cs.LG↗

arXiv2017

Development of a focused-X-ray luminescence tomography (FXLT) system

Biophotonics is an active research area in molecular imaging, genetic diagnosis and prognosis, with direct applicability in precision medicine. However, long-standing challenges of biophotonics are well known due to low signal-to-noise ratio and poor image quality, mainly due to strong optical scattering especially in deep tissues. Recently, X-ray luminescence computed tomography (XLCT) has emerged as a hybrid molecular imaging modality and shown great promises in overcoming the optical scattering in deep tissues. However, its high spatial resolution capacity has not been fully implemented yet. In this paper, with a superfine focused X-ray beam we design a focused-X-ray luminescence tomography (FXLT) system for spatial resolution better than 150 micrometers and molecular sensitivity of 2 micromolar. First, we describe our system design. Then, we demonstrate that the molecular sensitivity of FXLT is about 5 micromolar considering the emitted visible photons from background. Furthermore, we analyze the measurement time per scan from measured photons numbers with a fiber bundle-PMT setup, report numerical and experimental results. Finally, we specify the imaging system performance based on numerical simulations and physical measurements.

physics.med-ph↗

arXiv2017

General Backpropagation Algorithm for Training Second-order Neural Networks

The artificial neural network is a popular framework in machine learning. To empower individual neurons, we recently suggested that the current type of neurons could be upgraded to 2nd order counterparts, in which the linear operation between inputs to a neuron and the associated weights is replaced with a nonlinear quadratic operation. A single 2nd order neurons already has a strong nonlinear modeling ability, such as implementing basic fuzzy logic operations. In this paper, we develop a general backpropagation (BP) algorithm to train the network consisting of 2nd-order neurons. The numerical studies are performed to verify of the generalized BP algorithm.

cs.LG↗

arXiv2017

Reconstruction of optical parameters for molecular tomographic imaging

Optical molecular tomographic imaging is to reconstruct the concentration distribution of photon-molecular probes in a small animal from measured photon fluence rates. The localization and quantification of molecular probes is related to tissue optical properties. The optical parameters can be reconstructed by the optimization method, in which the objective function is not convex, and there are multiple local optimal solutions. The gradient-based optimization algorithm could converge prematurely to an inaccurate solution if the initial guess is not close enough to the true solution. In this paper, we develop a hybrid algorithm combined global optimization and gradient-based optimization methods to reconstruct optical parameters. Differential evolution is an effective global optimizer, and could yield an estimated solution near the global optimal region to eliminate the need of initial guess of the optical parameters. The gradient-based optimization method is further performed with initial values obtained from DE to obtain local optimal solution for more accurate reconstruction of optical parameters. This proposed reconstruction method is evaluated by comprehensive numerical simulation experiments. The results show that the proposed method is stable and accurate for the reconstruction of optical parameters.

physics.bio-ph↗

arXiv2017

Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)

Given the potential X-ray radiation risk to the patient, low-dose CT has attracted a considerable interest in the medical imaging field. The current main stream low-dose CT methods include vendor-specific sinogram domain filtration and iterative reconstruction, but they need to access original raw data whose formats are not transparent to most users. Due to the difficulty of modeling the statistical characteristics in the image domain, the existing methods for directly processing reconstructed images cannot eliminate image noise very well while keeping structural details. Inspired by the idea of deep learning, here we combine the autoencoder, the deconvolution network, and shortcut connections into the residual encoder-decoder convolutional neural network (RED-CNN) for low-dose CT imaging. After patch-based training, the proposed RED-CNN achieves a competitive performance relative to the-state-of-art methods in both simulated and clinical cases. Especially, our method has been favorably evaluated in terms of noise suppression, structural preservation and lesion detection.

physics.med-ph↗

arXiv2017

A New Type of Neurons for Machine Learning

In machine learning, the use of an artificial neural network is the mainstream approach. Such a network consists of layers of neurons. These neurons are of the same type characterized by the two features: (1) an inner product of an input vector and a matching weighting vector of trainable parameters and (2) a nonlinear excitation function. Here we investigate the possibility of replacing the inner product with a quadratic function of the input vector, thereby upgrading the 1st order neuron to the 2nd order neuron, empowering individual neurons, and facilitating the optimization of neural networks. Also, numerical examples are provided to illustrate the feasibility and merits of the 2nd order neurons. Finally, further topics are discussed.

cs.NE↗

arXiv2017

CT Image Reconstruction in a Low Dimensional Manifold

Regularization methods are commonly used in X-ray CT image reconstruction. Different regularization methods reflect the characterization of different prior knowledge of images. In a recent work, a new regularization method called a low-dimensional manifold model (LDMM) is investigated to characterize the low-dimensional patch manifold structure of natural images, where the manifold dimensionality characterizes structural information of an image. In this paper, we propose a CT image reconstruction method based on the prior knowledge of the low-dimensional manifold of CT image. Using the clinical raw projection data from GE clinic, we conduct comparisons for the CT image reconstruction among the proposed method, the simultaneous algebraic reconstruction technique (SART) with the total variation (TV) regularization, and the filtered back projection (FBP) method. Results show that the proposed method can successfully recover structural details of an imaging object, and achieve higher spatial and contrast resolution of the reconstructed image than counterparts of FBP and SART with TV.

physics.med-ph↗

arXiv2017

Optical Tomographic Imaging for Breast Cancer Detection

Diffuse optical breast imaging utilizes near-infrared (NIR) light propagation through tissues to assess the optical properties of tissue for the identification of abnormal tissue. This optical imaging approach is sensitive, cost-effective, and does not involve any ionizing radiation. However, the image reconstruction of diffuse optical tomography (DOT) is a nonlinear inverse problem and suffers from severe ill-posedness, especially in the cases of strong noise and incomplete data. In this paper, a novel image reconstruction method is proposed for the detection of breast cancer. This method split the image reconstruction problem into the localization of abnormal tissues and quantification of absorption variations. The localization of abnormal tissues is performed based on a new well-posed optimization model, which can be solved via differential evolution optimization method to achieve a stable image reconstruction. The quantification of abnormal absorption variations is then determined in localized regions of relatively small extents, which are potentially tumors. Consequently, the number of unknown absorption variables can be greatly reduced to overcome the underdetermined nature of diffuse optical tomography (DOT), allowing for accurate and stable reconstruction of the abnormal absorption variations in the breast. Numerical simulation experiments show that the image reconstruction method is stable and accurate for the identification of abnormal tissues, and robust against measurement noise of data.

physics.med-ph↗

arXiv2017

CT Image Denoising with Perceptive Deep Neural Networks

Increasing use of CT in modern medical practice has raised concerns over associated radiation dose. Reduction of radiation dose associated with CT can increase noise and artifacts, which can adversely affect diagnostic confidence. Denoising of low-dose CT images on the other hand can help improve diagnostic confidence, which however is a challenging problem due to its ill-posed nature, since one noisy image patch may correspond to many different output patches. In the past decade, machine learning based approaches have made quite impressive progress in this direction. However, most of those methods, including the recently popularized deep learning techniques, aim for minimizing mean-squared-error (MSE) between a denoised CT image and the ground truth, which results in losing important structural details due to over-smoothing, although the PSNR based performance measure looks great. In this work, we introduce a new perceptual similarity measure as the objective function for a deep convolutional neural network to facilitate CT image denoising. Instead of directly computing MSE for pixel-to-pixel intensity loss, we compare the perceptual features of a denoised output against those of the ground truth in a feature space. Therefore, our proposed method is capable of not only reducing the image noise levels, but also keeping the critical structural information at the same time. Promising results have been obtained in our experiments with a large number of CT images.

cs.CV↗

arXiv2016

Deep Learning for the Classification of Lung Nodules

Deep learning, as a promising new area of machine learning, has attracted a rapidly increasing attention in the field of medical imaging. Compared to the conventional machine learning methods, deep learning requires no hand-tuned feature extractor, and has shown a superior performance in many visual object recognition applications. In this study, we develop a deep convolutional neural network (CNN) and apply it to thoracic CT images for the classification of lung nodules. We present the CNN architecture and classification accuracy for the original images of lung nodules. In order to understand the features of lung nodules, we further construct new datasets, based on the combination of artificial geometric nodules and some transformations of the original images, as well as a stochastic nodule shape model. It is found that simplistic geometric nodules cannot capture the important features of lung nodules.

q-bio.QM↗