Wenyuan Qi

Canon (United States)

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Evren Asma

Canon (United States)

Yongyi Yang

Zhejiang University

Michael A. King

University of Massachusetts Chan Medical School

Xiaofeng Niu

Taiyuan University of Technology

Miles N. Wernick

Illinois Institute of Technology

Chung Chan

Institut de la Vision

P. Hendrik Pretorius

University of Massachusetts Chan Medical School

Jinyi Qi

University of California, Davis

Tiantian Li

University of California, Davis

Jeff Kolthammer

Canon (United States)

Cooperative Institutions

Canon (United States)

Toshiba (United States)

University of California, Davis

Chinese Academy of Medical Sciences & Peking Union Medical College

Zhejiang University

University of Massachusetts Chan Medical School

Shanghai Jiao Tong University

Sun Yat-sen University

Beijing Hospital

Capital Medical University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Assessment of Spatial Resolution in PET Images Denoised with Deep Convolutional Neural Networks

2021 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC) (2020)

Li Yang Wenyuan Qi Chung Chan Evren Asma

Deep convolutional neural networks (DCNN) have been successfully used for denoising natural images as well as medical images. In previous work we had used DCNNs for positron emission tomography (PET) image denoising and demonstrated improved performance on region of interest (ROI) quantitation. Another important metric for characterizing PET images is spatial resolution and it is unclear how DCNN denoising impacts spatial resolution in PET images. In this work, we investigate the spatial resolution properties of DCNN-denoised images with realistic patient backgrounds. We inserted 10 mm spheres at different contrast levels into lungs and livers of real patient datasets through GATE simulations. We then measured the full-width-at-half-maximum (FWHM) values along the principal directions of these inserted lesions on the DCNN denoised images and compared these results with those from Gaussian post-filtered images. We also evaluated the relationship between the measured spatial resolution levels and lesion contrasts in different background regions. The results show that DCNN denoising can simultaneously achieve improved spatial resolution and improved ROI quantification performance compared to standard, Gaussian post-filtering.

10.1109/nss/mic42677.2020.9508022

Cite

Citations (1)

Reconstruction with angular compensation in respiratory-gated cardiac SPECT

2022 IEEE International Conference on Image Processing (ICIP) (2014)

Wenyuan Qi Yongyi Yang Miles N. Wernick P. Hendrik Pretorius Michael A. King

In respiratory-gated cardiac SPECT with amplitude binning, the acquisition time can vary greatly both among respiratory gates and among acquisition angles within each gate. If not properly accounted for, this uneven distribution in acquired data statistics will lead to limited-angle artifacts in reconstruction, which in turn can impact on the accuracy of respiratory motion correction. We investigate a compensation scheme for this uneven distribution by directly taking into account in the imaging model the actual acquisition time at different projection angles. In the experiment, we demonstrated this approach with simulated NCAT imaging data, for which quantitative results were obtained on both the reconstructed myocardium and estimated motion; we also tested the proposed approach on a set of clinical acquisition. The results show that the proposed approach could effectively suppress the limited-angle artifacts and improve the reconstruction in terms of both image accuracy and lesion detectability.

Spect imaging

Data set

10.1109/icip.2014.7025738

Cite

Citations (3)

Modeling and simulation of aero-thermal radiation of dome based on inverse ray tracing

Seventh Asia Pacific Conference on Optics Manufacture and 2021 International Forum of Young Scientists on Advanced Optical Manufacturing (APCOM and YSAOM 2021) (2022)

Lei Gao Wenyuan Qi Huan Li Shuang Wang Mingna Liu

When infrared imaging guided aircraft is flying at high speed, the aerodynamic heated dome generates strong infrared radiation, which reduces the detection performance of system. In order to research the influence of high-temperature dome on imaging quality and improve simulation efficiency, firstly, a fluid thermal solid coupling calculation model of dome under trajectory is established to obtain the non-uniform temperature field with time, secondly, the inverse ray tracing method is proposed according to reversibility of optical path, trace light emitted from detector pixel to the inner surface of dome, then irradiance of the pixel is obtained through radiation emission and transmission characteristics of the dome, finally, traverse all pixels to obtain the thermal noise image. The results show that when working time is 7s and the aircraft speed and height gradually decrease, the dome temperature reaches maximum value of 417.6K at the end, irradiance reaches maximum value of 0.1297W/m², image signal-to-noise ratio(SNR) decreases by 32.73% and contrast decreases by 80.59%. The calculation method of ray transmission based on inverse ray tracing and the evaluation method for degraded image quality are proposed in this paper, which lays a foundation for optimization of infrared detection system band and suppression of aerodynamic heat in the follow-up work.

Dome (geology)

Thermal Radiation

Traverse

10.1117/12.2617578

Cite

Citations (0)

Pharmacokinetics Of Indacaterol/Glycopyrronium/Mometasone Furoate In Healthy Chinese Volunteers: Results From A Randomized, Open-Label, Parallel-Group Study

Journal of Allergy and Clinical Immunology (2021)

Wenyuan Qi Wei Xue Kexin Li Soniya Vaidya Jian Sun

Indacaterol

Mometasone furoate

Crossover study

10.1016/j.jaci.2020.12.236

Cite

Citations (1)

Anatomically aided PET image reconstruction using deep neural networks

Medical Physics (2021)

Zhaoheng Xie Tiantian Li Xuezhu Zhang Wenyuan Qi Evren Asma

The developments of PET/CT and PET/MR scanners provide opportunities for improving PET image quality by using anatomical information. In this paper, we propose a novel co-learning three-dimensional (3D) convolutional neural network (CNN) to extract modality-specific features from PET/CT image pairs and integrate complementary features into an iterative reconstruction framework to improve PET image reconstruction.

Kernel (algebra)

10.1002/mp.15051

Cite

Citations (23)

Unsupervised deep learning framework for data‐driven gating in positron emission tomography

Medical Physics (2023)

Tiantian Li Zhaoheng Xie Wenyuan Qi Evren Asma Jinyi Qi

Physiological motion, such as respiratory motion, has become a limiting factor in the spatial resolution of positron emission tomography (PET) imaging as the resolution of PET detectors continue to improve. Motion-induced misregistration between PET and CT images can also cause attenuation correction artifacts. Respiratory gating can be used to freeze the motion and to reduce motion induced artifacts.In this study, we propose a robust data-driven approach using an unsupervised deep clustering network that employs an autoencoder (AE) to extract latent features for respiratory gating.We first divide list-mode PET data into short-time frames. The short-time frame images are reconstructed without attenuation, scatter, or randoms correction to avoid attenuation mismatch artifacts and to reduce image reconstruction time. The deep AE is then trained using reconstructed short-time frame images to extract latent features for respiratory gating. No additional data are required for the AE training. K-means clustering is subsequently used to perform respiratory gating based on the latent features extracted by the deep AE. The effectiveness of our proposed Deep Clustering method was evaluated using physical phantom and real patient datasets. The performance was compared against phase gating based on an external signal (External) and image based principal component analysis (PCA) with K-means clustering (Image PCA).The proposed method produced gated images with higher contrast and sharper myocardium boundaries than those obtained using the External gating method and Image PCA. Quantitatively, the gated images generated by the proposed Deep Clustering method showed larger center of mass (COM) displacement and higher lesion contrast than those obtained using the other two methods.The effectiveness of our proposed method was validated using physical phantom and real patient data. The results showed our proposed framework could provide superior gating than the conventional External method and Image PCA.

10.1002/mp.16642

Cite

Citations (2)

Motion correction of respiratory-gated PET image using deep learning based image registration framework

Tiantian Li Mengxi Zhang Wenyuan Qi Evren Asma Jinyi Qi

Artifacts caused by patient breathing and movement during PET data acquisition affect image quality. Respiratory gating has been proposed to gate the list-mode PET data into multiple bins over a respiratory cycle. Non-rigid registration of respiratory-gated PET images can reduce the motion artifacts and preserve the count statistics, but it is time consuming. In this work, we propose an unsupervised non-rigid image registration framework using deep learning. We use a differentiable spatial transformer layer to warp the source image to the target image and use a stacked structure for deformation field refinement. Estimated deformation fields were incorporated into an iterative image reconstruction algorithm to perform motion compensated PET image reconstruction. We validated the proposed method using simulation and clinical data and showed its ability to reduce the motion artifact in PET images.

Image registration

Artifact (error)

10.1117/12.2534851

Cite

Citations (2)

A quantitative study of motion estimation methods on 4D cardiac gated SPECT reconstruction

Medical Physics (2012)

Wenyuan Qi Yongyi Yang Xiaofeng Niu Michael A. King

Motion-compensated temporal processing can have a major impact on improving the image quality in gated cardiac single photon emission computed tomography (SPECT). In this work, we investigate the effect of different optical flow estimation methods for motion-compensated temporal processing in gated SPECT. In particular, we explore whether better motion estimation can substantially improve reconstructed image quality, and how the estimated motion would compare to the ideal case of known motion in terms of reconstruction.

Optical Flow

Cardiac PET

Cardiac Imaging

10.1118/1.4738377

Cite

Citations (19)

Short PET Scans for Outside-the-FOV Scatter Estimation

Wenyuan Qi Chung Chan Yang Li Evren Asma

Scatter correction is an important step in PET image reconstruction. When the object is longer than the axial extent of the scanned region, outside-the-axial-field-of-view (OFOV) scatter correction is necessary and not trivial. For most scatter estimation techniques, both attenuation and emission distributions outside the axial FOV are needed. One direct way for OFOV scatter estimation would be to perform an extra PET and CT scan at bed positions adjacent to the axial region of interest in order to obtain attenuation and emission information. This approach however would both prolong the PET scan time and also increase the CT radiation dose. An alternative approach is to extrapolate the known activity and attenuation maps from inside the FOV but this could be inaccurate if the OFOV activity distribution is very different than that inside the FOV. In order to reduce the radiation dose and scan time while maintaining OFOV scatter estimation accuracy, we propose performing short PET scans over bed positions adjacent to the region of interest. Evaluation on clinical images shows that OFOV scatter correction using these short PET scans can significantly improve image quality compared to not performing OFOV scatter correction and bring image quality close to the case of using full scans for adjacent bed positions.

Field of view

Correction for attenuation

10.1109/nssmic.2018.8824685

Cite

Citations (0)

Performance Characterization of an SiPM-based Time-of-Flight Canon PET/CT Scanner

Xiaoli Li Wenyuan Qi Masaki Miyahara Jeff Kolthammer

14 Introduction: The use of silicon photomultipliers (SiPMs) as photosensors has enabled novel PET detector designs with improved physical performance characteristics. We measured the performance of an SiPM-based, time-of-flight (TOF) PET/CT scanner developed by Canon Medical Systems Corporation (Otawara, Japan). The scanner design is based on lutetium-yttrium oxyorthosilicate (LYSO) scintillator material, pixelated and coupled individually to SiPM photosensors. This study evaluated its important performance characteristics over clinically-relevant count rates according to the NEMA NU 2-2018 standard. Methods: The new SiPM-based PET/CT scanner has a 780-mm bore diameter and 27-cm axial field of view (FOV). The PET detector consists of 48 circularly-arranged detector units, each consisting of an array of 12 x 60 pixels. Each pixel measures 4.2 mm x 4.2 mm x 20 mm and is coupled individually (i.e. “one-to-one”) to an SiPM photosensor, thus providing complete coverage of the outward side of the scintillator array. In this study, the important performance characteristics of the PET system were evaluated. The energy discrimination capability was measured by placing a 2.4 mCi 68Ge line source at the center of the scanner FOV and collecting coincidence events. A histogram of acquired events’ energies was constructed and corrected for random events; the peak determined by fitting a parabola to the data within approximately 10% of the peak value; and the FWHM calculated as between the histogram values at half of the peak. The energy resolution was recorded as a percentage of the peak energy. System sensitivity was measured using 18F-FDG line sources inside five metal sleeves as according to the NEMA NU 2-2018 standard. The timing resolution, the noise equivalent count rate (NECR) and the scatter fraction were measured versus activity concentration according to NEMA NU 2-2018 across clinically relevant count rates. For comparison, timing resolution was also measured using a conventional method, i.e. a centered 2.4 mCi 68Ge line source in air. Image quality was characterized by filling the NEMA image quality phantom with a sphere to background ratio of 4:1 and scan for 4 minutes. The images were generated using a list mode based TOF-OSEM reconstruction incorporating resolution recovery with 4 iterations, 12 subsets and 2-mm image voxels. No post-reconstruction filter was applied. Regions of interests were drawn according to NEMA NU 2-2018 standard. Results: The measured energy resolution was 11.4% FWHM for the 68Ge line source at scanner center. The measured timing resolution was 255 ps for the same 68Ge line source. The measured sensitivity was 13.5 kcps/MBq, both when the 18F line source was at the scanner iso-center and 10 cm off center. The timing resolution ranges from 257 ps at an average effective radioactivity concentration of 5.3 kBq/mL to 262 ps at average effective radioactivity concentration of 9.6 kBq/mL, i.e. a singles rate equivalent to a typical whole-body patient scan. At 9.6 kBq/mL average effective radioactivity concentration, the scatter fraction is 0.36 and the NECR is 144.7 kcps. The contrast recovery coefficient ranges from 67.2% for the 10 mm diameter sphere to 93.1% for the 37 mm diameter sphere. The corresponding background variability ranges from 7.3% to 3.3% in the 10 mm to 37 mm background ROIs, respectively. Conclusions: The new SiPM-based PET/CT scanner from Canon demonstrates state of the art performance with 13.5 kcps/MBq sensitivity. Excellent timing resolution was measured across clinically relevant count rates.

Silicon Photomultiplier

Lyso-

Field of view

Time-of-Flight

Cite

Citations (10)