Sen Wang

All Publications

• Spectral optimization using fast kV switching and filtration for photon counting CT with realistic detector responses: a simulation study. Journal of medical imaging (Bellingham, Wash.) Wang, S., Yang, Y., Pal, D., Yin, Z., Maltz, J. S., Pelc, N. J., Wang, A. S. 2024; 11 (Suppl 1): S12805

  Abstract

  Photon counting CT (PCCT) provides spectral measurements for material decomposition. However, the image noise (at a fixed dose) depends on the source spectrum. Our study investigates the potential benefits from spectral optimization using fast kV switching and filtration to reduce noise in material decomposition.The effect of the input spectra on noise performance in both two-basis material decomposition and three-basis material decomposition was compared using Cramer-Rao lower bound analysis in the projection domain and in a digital phantom study in the image domain. The fluences of different spectra were normalized using the CT dose index to maintain constant dose levels. Four detector response models based on Si or CdTe were included in the analysis.For single kV scans, kV selection can be optimized based on the imaging task and object size. Furthermore, our results suggest that noise in material decomposition can be substantially reduced with fast kV switching. For two-material decomposition, fast kV switching reduces the standard deviation (SD) by ∼ 10 % . For three-material decomposition, greater noise reduction in material images was found with fast kV switching (26.2% for calcium and 25.8% for iodine, in terms of SD), which suggests that challenging tasks benefit more from the richer spectral information provided by fast kV switching.The performance of PCCT in material decomposition can be improved by optimizing source spectrum settings. Task-specific tube voltages can be selected for single kV scans. Also, our results demonstrate that utilizing fast kV switching can substantially reduce the noise in material decomposition for both two- and three-material decompositions, and a fixed Gd filter can further enhance such improvements for two-material decomposition.

  View details for DOI 10.1117/1.JMI.11.S1.S12805

  View details for PubMedID 39072221

  View details for PubMedCentralID PMC11272100

• Optimal weighting strategies for maximizing contrast-to-noise ratio in photon counting CT images. Medical physics Yang, Y., Wang, S., Stevens, G. M., Fan, J., Wang, A. S. 2024

  Abstract

  Photon counting detectors (PCDs) with energy discrimination capabilities have the potential to generate grayscale CT images with improved contrast-to-noise ratio (CNR) through optimal weighting of their spectral measurements.This study evaluates the CNR performance of grayscale CT projections and images generated from spectral measurements of PCDs using three energy-weighting strategies: pre-log weighting, post-log weighting, and material decomposition (MD) weighting. This study provides the expressions of optimal weights and maximum achievable CNR of these energy-weighting strategies, which only require the knowledge of detected bin counts and do not require information of PCD energy responses or imaging techniques.We defined and solved a generalized eigenvalue problem to obtain the maximum achievable CNR in the projection domain for low-contrast tasks using three energy-weighting strategies: pre-log weighting (weighted sum of energy bin counts), post-log weighting (weighted sum of line integrals), and MD weighting (weighted sum of basis material thicknesses, which is equivalent to virtual monoenergetic images [VMIs]). These expressions only contain energy bin counts from PCD measurements. We used a realistic PCD energy response model to simulate the detected bin counts and conducted Monte Carlo simulations of different contrast tasks and phantoms to evaluate the projection- and image-domain CNR performance of these energy-weighting strategies. Additionally, the total counts method (a special case of pre-log weighting with unity weights) was included for comparison. We also conducted Gammex head and body phantom scans on an edge-on-irradiated silicon PCCT prototype to evaluate the image-domain CNR performance of these energy-weighting strategies.The results show that pre-log, post-log, and MD weighting strategies generate approximately equal projection-domain maximum achievable CNR, with a difference of less than 2%, and outperform the total counts method. These three energy-weighting strategies also generate approximately equal image-domain maximum CNR when the contrast task is located at the center of a homogeneous phantom. Pre-log weighting generates the highest image-domain CNR for an off-center contrast task location or inhomogeneous phantoms while also outperforming the total counts method.We derived the expression of projection-domain maximum achievable CNR using three energy-weighting strategies. Our results suggest that using pre-log weighting strategies enables fast grayscale CT image generation with high CNR from spectral PCD measurements for inhomogeneous phantoms and off-center region of interests (ROIs).

  View details for DOI 10.1002/mp.17489

  View details for PubMedID 39447021

• Emulating Low-Dose PCCT Image Pairs with Independent Noise for Self-Supervised Spectral Image Denoising. IEEE transactions on medical imaging Wang, S., Yang, Y., Stevens, G. M., Yin, Z., Wang, A. S. 2024; PP

  Abstract

  Photon counting CT (PCCT) acquires spectral measurements and enables generation of material decomposition (MD) images that provide distinct advantages in various clinical situations. However, noise amplification is observed in MD images, and denoising is typically applied. Clean or high-quality references are rare in clinical scans, often making supervised learning (Noise2Clean) impractical. Noise2Noise is a self-supervised counterpart, using noisy images and corresponding noisy references with zero-mean, independent noise. PCCT counts transmitted photons separately, and raw measurements are assumed to follow a Poisson distribution in each energy bin, providing the possibility to create noise-independent pairs. The approach is to use binomial selection to split the counts into two low-dose scans with independent noise. We prove that the reconstructed spectral images inherit the noise independence from counts domain through noise propagation analysis and also validated it in numerical simulation and experimental phantom scans. The method offers the flexibility to split measurements into desired dose levels while ensuring the reconstructed images share identical underlying features, thereby strengthening the model's robustness for input dose levels and capability of preserving fine details. In both numerical simulation and experimental phantom scans, we demonstrated that Noise2Noise with binomial selection outperforms other common self-supervised learning methods based on different presumptive conditions.

  View details for DOI 10.1109/TMI.2024.3449817

  View details for PubMedID 39196747

• Retrospective Tube Current Modulation Optimization of Individualized Organ-Level CT Dose and Image Quality Wang, S., Medrano, M., Imran, A., Stevens, G., Tse, J., Wang, A. S., Fahrig, R., Sabol, J. M., Li, K. SPIE-INT SOC OPTICAL ENGINEERING. 2024

  View details for DOI 10.1117/12.3006870

  View details for Web of Science ID 001223517100036

• Denoising X-Ray Images with Deep Learning: Impact of Spatially Correlated Noise Ku, A., Wang, S., Wang, A., Fahrig, R., Sabol, J. M., Li, K. SPIE-INT SOC OPTICAL ENGINEERING. 2024

  View details for DOI 10.1117/12.3006556

  View details for Web of Science ID 001223517100027

• Optimal Weighting Strategies for Maximizing Contrast-to-Noise Ratio in Photon Counting CT Images Yang, Y., Wang, S., Stevens, G., Fan, J., Wang, A. S., Fahrig, R., Sabol, J. M., Li, K. SPIE-INT SOC OPTICAL ENGINEERING. 2024

  View details for DOI 10.1117/12.3006847

  View details for Web of Science ID 001223517100001

• Personalized, Scout-Based Dose Estimation for Prospective Optimization of CT Tube Current Modulation Medrano, M., Wang, S., Imran, A., Stevens, G., Tse, J., Wang, A., Fahrig, R., Sabol, J. M., Li, K. SPIE-INT SOC OPTICAL ENGINEERING. 2024

  View details for DOI 10.1117/12.3006268

  View details for Web of Science ID 001223517100043

• Empirical optimization of energy bin weights for compressing measurements with realistic photon counting x-ray detectors. Medical physics Yang, Y., Wang, S., Pal, D., Yin, Z., Pelc, N. J., Wang, A. S. 2023

  Abstract

  BACKGROUND: Photon counting detectors (PCDs) provide higher spatial resolution, improved contrast-to-noise ratio (CNR), and energy discriminating capabilities. However, the greatly increased amount of projection data in photon counting computed tomography (PCCT) systems becomes challenging to transmit through the slip ring, process, and store.PURPOSE: This study proposes and evaluates an empirical optimization algorithm to obtain optimal energy weights for energy bin data compression. This algorithm is universally applicable to spectral imaging tasks including 2 and 3 material decomposition (MD) tasks and virtual monoenergetic images (VMIs). This method is simple to implement while preserving spectral information for the full range of object thicknesses and is applicable to different PCDs, for example, silicon detectors and CdTe detectors.METHODS: We used realistic detector energy response models to simulate the spectral response of different PCDs and an empirical calibration method to fit a semi-empirical forward model for each PCD. We numerically optimized the optimal energy weights by minimizing the average relative Cramer-Rao lower bound (CRLB) due to the energy-weighted bin compression, for MD and VMI tasks over a range of material area density rho A , m ${\rho }_{A,m}$ (0-40g/cm2 water, 0-2.16g/cm2 calcium). We used Monte Carlo simulation of a step wedge phantom and an anthropomorphic head phantom to evaluate the performance of this energy bin compression method in the projection domain and image domain, respectively.RESULTS: The results show that for 2 MD, the energy bin compression method can reduce PCCT data size by 75% and 60%, with an average variance penalty of less than 17% and 3% for silicon and CdTe detectors, respectively. For 3 MD tasks with a K-edge material (iodine), this method can reduce the data size by 62.5% and 40% with an average variance penalty of less than 12% and 13% for silicon and CdTe detectors, respectively.CONCLUSIONS: We proposed an energy bin compression method that is broadly applicable to different PCCT systems and object sizes, with high data compression ratio and little loss of spectral information.

  View details for DOI 10.1002/mp.16590

  View details for PubMedID 37401203

• Personalized CT organ noise estimation from scout images SPIE Medical Imaging: Physics of Medical Imaging Imran, A., Pal, D., Wang, S., Dutta, S., Zucker, E., Wang, A. 2022

  View details for DOI 10.1117/12.2610986

• Fast kV Switching for Improved Material Decomposition with Photon Counting X-ray Detectors Wang, S., Yang, Y., Pal, D., Pelc, N. J., Wang, A. S., Zhao, W., Yu, L. SPIE-INT SOC OPTICAL ENGINEERING. 2022

  View details for DOI 10.1117/12.2611601

  View details for Web of Science ID 000836294000014

• Empirical Optimization of Energy Bin Weights for Compressing Measurements with Photon Counting X-ray Detectors Yang, Y., Wang, S., Pal, D., Pelc, N. J., Wang, A. S., Zhao, W., Yu, L. SPIE-INT SOC OPTICAL ENGINEERING. 2022

  View details for DOI 10.1117/12.2611555

  View details for Web of Science ID 000836294000013

• Multimodal Contrastive Learning for Prospective Personalized Estimation of CT Organ Dose Imran, A., Wang, S., Pal, D., Dutta, S., Zucker, E., Wang, A., Wang, L., Dou, Q., Fletcher, P. T., Speidel, S., Li, S. SPRINGER INTERNATIONAL PUBLISHING AG. 2022: 634-643

  View details for DOI 10.1007/978-3-031-16431-6_60

  View details for Web of Science ID 000867524300060

• Personalized CT Organ Dose Estimation from Scout Images International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) Imran, A., Wang, S., Pal, D., Dutta, S., Patel, B., Zucker, E., Wang, A. 2021

  View details for DOI 10.1007/978-3-030-87202-1_47

• A systematic evaluation and optimization of automatic detection of ulcers in wireless capsule endoscopy on a large dataset using deep convolutional neural networks. Physics in medicine and biology Wang, S., Xing, Y., Zhang, L., Gao, H., Zhang, H. 2019; 64 (23): 235014

  Abstract

  Compared with conventional gastroscopy which is invasive and painful, wireless capsule endoscopy (WCE) can provide noninvasive examination of gastrointestinal (GI) tract. The WCE video can effectively support physicians to reach a diagnostic decision while a huge number of images need to be analyzed (more than 50 000 frames per patient). In this paper, we propose a computer-aided diagnosis method called second glance (secG) detection framework for automatic detection of ulcers based on deep convolutional neural networks that provides both classification confidence and bounding box of lesion area. We evaluated its performance on a large dataset that consists of 1504 patient cases (the largest WCE ulcer dataset to our best knowledge, 1076 cases with ulcers, 428 normal cases). We use 15 781 ulcer frames from 753 ulcer cases and 17 138 normal frames from 300 normal cases for training. Validation dataset consists of 2040 ulcer frames from 108 cases and 2319 frames from 43 normal cases. For test, we use 4917 ulcer frames from 215 ulcer cases and 5007 frames from 85 normal cases. Test results demonstrate the 0.9469 ROC-AUC of the proposed secG detection framework outperforms state-of-the-art detection frameworks including Faster-RCNN (0.9014) and SSD-300 (0.8355), which implies the effectiveness of our method. From the ulcer size analysis, we find the detection of ulcers is highly related to the size. For ulcers with size larger than 1% of the full image size, the sensitivity exceeds 92.00%. For ulcers that are smaller than 1% of the full image size, the sensitivity is around 85.00%. The overall sensitivity, specificity and accuracy are 89.71%, 90.48% and 90.10%, at a threshold value of 0.6706, which implies the potential of the proposed method to suppress oversights and to reduce the burden of physicians.

  View details for DOI 10.1088/1361-6560/ab5086

  View details for PubMedID 31645019

• Deep Convolutional Neural Network for Ulcer Recognition in Wireless Capsule Endoscopy: Experimental Feasibility and Optimization COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE Wang, S., Xing, Y., Zhang, L., Gao, H., Zhang, H. 2019; 2019: 7546215

  Abstract

  Wireless capsule endoscopy (WCE) has developed rapidly over the last several years and now enables physicians to examine the gastrointestinal tract without surgical operation. However, a large number of images must be analyzed to obtain a diagnosis. Deep convolutional neural networks (CNNs) have demonstrated impressive performance in different computer vision tasks. Thus, in this work, we aim to explore the feasibility of deep learning for ulcer recognition and optimize a CNN-based ulcer recognition architecture for WCE images. By analyzing the ulcer recognition task and characteristics of classic deep learning networks, we propose a HAnet architecture that uses ResNet-34 as the base network and fuses hyper features from the shallow layer with deep features in deeper layers to provide final diagnostic decisions. 1,416 independent WCE videos are collected for this study. The overall test accuracy of our HAnet is 92.05%, and its sensitivity and specificity are 91.64% and 92.42%, respectively. According to our comparisons of F1, F2, and ROC-AUC, the proposed method performs better than several off-the-shelf CNN models, including VGG, DenseNet, and Inception-ResNet-v2, and classical machine learning methods with handcrafted features for WCE image classification. Overall, this study demonstrates that recognizing ulcers in WCE images via the deep CNN method is feasible and could help reduce the tedious image reading work of physicians. Moreover, our HAnet architecture tailored for this problem gives a fine choice for the design of network structure.

  View details for DOI 10.1155/2019/7546215

  View details for Web of Science ID 000488776000001

  View details for PubMedID 31641370

  View details for PubMedCentralID PMC6766681

• Second Glance framework (secG): Enhanced Ulcer Detection with Deep Learning on a Large Wireless Capsule Endoscopy Dataset Wang, S., Xing, Y., Zhang, L., Gao, H., Zhang, H., Jiang, Chen, Z., Chen, G. SPIE-INT SOC OPTICAL ENGINEERING. 2019

  View details for DOI 10.1117/12.2540456

  View details for Web of Science ID 000502121300030

• Systematic implementation of spectral CT with a photon counting detector for liquid security inspection NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT Xu, X., Xing, Y., Wang, S., Zhang, L. 2018; 893: 99–108

  View details for DOI 10.1016/j.nima.2018.03.011

  View details for Web of Science ID 000430167800013

• Enhanced material separation with a quasi-monochromatic CT imaging method using a photon counting detector NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT Wang, S., Xing, Y., Zhang, L., Xu, X. 2018; 881: 9–15

  View details for DOI 10.1016/j.nima.2017.10.066

  View details for Web of Science ID 000418525100002

• Quasi-monochromatic imaging in x-ray CT via spectral deconvolution using photon-counting detectors PHYSICS IN MEDICINE AND BIOLOGY Wang, S., Gao, H., Zhang, L., Wu, D., Xu, X. 2017; 62 (6): 2208–23

  Abstract

  Photon-counting detectors can obtain the spectral information from an incident x-ray spectrum, although the detected counts may differ from the incident counts due to the detector response. If uncorrected or uncompensated, the response will lead to distortion in CT reconstruction. With the intention of reducing the distortion and exploring the potential of photon-counting detectors, a novel reconstruction strategy with spectral deconvolution, which attempts to set itself apart from traditional material decomposition frameworks, is proposed in this paper. It applies deconvolution to the energy window counts using a calibrated detector response and then uses the post-deconvolution photon counts to reconstruct images in multi-energy windows. The output has a quantitative meaning as a quasi-monochromatic attenuation coefficient, because a relatively narrow energy window width is selected. The deconvolution settings and results are carefully discussed in the numerical simulation. An experimental study is then carried out to verify the effectiveness and robustness. The results show that the reconstructed attenuation coefficients after deconvolution fit the standard reference data very well in most of the energy windows, which implies the feasibility of this quasi-monochromatic imaging method.

  View details for DOI 10.1088/1361-6560/aa5a47

  View details for Web of Science ID 000395801000004

  View details for PubMedID 28099156

• More Accurate and Less Noisy Spectral Deconvolution Strategy using Photon Counting Detectors Wang, S., Zhang, L., Xu, X., IEEE IEEE. 2017

  View details for Web of Science ID 000455836200293

• A hybrid Monte Carlo model for the energy response functions of X-ray photon counting detectors NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT Wu, D., Xu, X., Zhang, L., Wang, S. 2016; 830: 397-406

  View details for DOI 10.1016/j.nima.2016.05.097

  View details for Web of Science ID 000381530300051

• A weighted polynomial based material decomposition method for spectral x-ray CT imaging. Physics in medicine and biology Wu, D., Zhang, L., Zhu, X., Xu, X., Wang, S. 2016; 61 (10): 3749-83

  Abstract

  Currently in photon counting based spectral x-ray computed tomography (CT) imaging, pre-reconstruction basis materials decomposition is an effective way to reconstruct densities of various materials. The iterative maximum-likelihood method requires precise spectrum information and is time-costly. In this paper, a novel non-iterative decomposition method based on polynomials is proposed for spectral CT, whose aim was to optimize the noise performance when there is more energy bins than the number of basis materials. Several subsets were taken from all the energy bins and conventional polynomials were established for each of them. The decomposition results from each polynomial were summed with pre-calculated weighting factors, which were designed to minimize the overall noises. Numerical studies showed that the decomposition noise of the proposed method was close to the Cramer-Rao lower bound under Poisson noises. Furthermore, experiments were carried out with an XCounter Filte X1 photon counting detector for two-material decomposition and three-material decomposition for validation.

  View details for DOI 10.1088/0031-9155/61/10/3749

  View details for PubMedID 27082291

• Preliminary Study of Quantitative X-ray Spectral Imaging with Spectral Deconvolution Wang, S., Zhang, L., Xu, X., Wu, D., IEEE IEEE. 2016

  View details for Web of Science ID 000432419500116