Mr. Shaoyan Pan Profile

Shaoyan Pan

at Emory Univ

SPIE Involvement:

Author

Publications (17)

Proceedings Article | 3 April 2024 Poster + Paper

MRI-based prostate cancer detection using cross-shaped windows transformer

Yuheng Li, Jacob Wynne, Jing Wang, Richard L. Qiu, Justin Roper, Shaoyan Pan, Ashesh Jani, Tian Liu, Pretesh Patel, Hui Mao, Xiaofeng Yang

Proceedings Volume 12927, 129271T (2024) https://doi.org/10.1117/12.3006812

KEYWORDS: Transformers, Cancer detection, Windows, Prostate cancer, Prostate, Magnetic resonance imaging, Education and training, Principal component analysis, Object detection, Deep learning

Read Abstract +

Proceedings Article | 2 April 2024 Poster + Paper

Synthetic CT generation from MRI using 3D diffusion model

Shaoyan Pan, Elham Abouei, Jacob Wynne, Tonghe Wang, Richard L. Qiu, Yuheng Li, Chih-Wei Chang, Junbo Peng, Shihan Qiu, Justin Roper, Pretesh Patel, David Yu, Hui Mao, Xiaofeng Yang

Proceedings Volume 12926, 129262L (2024) https://doi.org/10.1117/12.3006578

KEYWORDS: Computed tomography, Magnetic resonance imaging, Diffusion, 3D modeling, Denoising, Brain, Prostate, Education and training, Radiotherapy, Lithium

Read Abstract +

Proceedings Article | 2 April 2024 Presentation + Paper

Patient-specific 3D volumetric CBCT image reconstruction with single x-ray projection using denoising diffusion probabilistic model

Shaoyan Pan, Shao-Yuan Lo, Chih-Wei Chang, Ellahheh Salari, Tonghe Wang, Justin Roper, Aparna Kesarwala, Xiaofeng Yang

Proceedings Volume 12931, 129310P (2024) https://doi.org/10.1117/12.3006561

KEYWORDS: Cone beam computed tomography, Diffusion, 3D modeling, X-rays, 3D projection, Denoising, Image processing, Education and training, CT reconstruction, Image restoration

Read Abstract +

Proceedings Article | 2 April 2024 Poster + Paper

Dose-weighted proton linear energy transfer map generation using a deep learning framework

Yuan Gao, Chih-Wei Chang, Shaoyan Pan, Junbo Peng, Chaoqiong Ma, Pretesh Patel, Justin Roper, Jun Zhou, Xiaofeng Yang

Proceedings Volume 12930, 1293023 (2024) https://doi.org/10.1117/12.3006962

KEYWORDS: Proton therapy, Deep learning, Education and training, Monte Carlo methods, Gallium nitride, Tissues, Energy transfer, Radiotherapy, Prostate cancer, Performance modeling

Read Abstract +

Proceedings Article | 2 April 2024 Poster + Paper

Cardiac MRI segmentation using block-partitioned transformer with global-local information integration

Elham Abouei, Shaoyan Pan, Mingzhe Hu, Aparna Kesarwala, Jun Zhou, Justin Roper, Tian Liu, Xiaofeng Yang

Proceedings Volume 12930, 1293021 (2024) https://doi.org/10.1117/12.3006929

KEYWORDS: Image segmentation, Cardiovascular magnetic resonance imaging, Windows, Transformers, Magnetic resonance imaging, Deep learning, Information fusion, Visualization, Network architectures

Read Abstract +

Proceedings Article | 2 April 2024 Poster + Paper

Using diffusion model to generate high-resolution MRI

Chih-Wei Chang, Shaoyan Pan, Junbo Peng, Elahheh Salari, Justin Roper, Richard Qiu, Yuan Gao, Tian Liu, Hui-Kuo Shu, Hui Mao, Xiaofeng Yang

Proceedings Volume 12930, 129301S (2024) https://doi.org/10.1117/12.3006586

KEYWORDS: Magnetic resonance imaging, Diffusion, Lawrencium, Image processing, Deep learning, Brain, Image quality, Denoising, Data modeling, Neuroimaging

Read Abstract +

Proceedings Article | 2 April 2024 Poster + Paper

Full-dose PET synthesis from low-dose PET using 2D high efficiency denoising diffusion probabilistic model

Shaoyan Pan, Elham Abouei, Junbo Peng, Joshua Qian, Jacob Wynne, Tonghe Wang, Chih-Wei Chang, Justin Roper, Jonathon Nye, Hui Mao, Xiaofeng Yang

Proceedings Volume 12930, 129301Q (2024) https://doi.org/10.1117/12.3006565

KEYWORDS: Positron emission tomography, Diffusion, Denoising, Computed tomography, Medical imaging, 3D modeling, Image segmentation, Deep learning, Magnetic resonance imaging, Machine learning

Read Abstract +

Proceedings Article | 29 March 2024 Presentation + Paper

GhostMorph: a computationally efficient model for deformable inter-subject registration

Mingzhe Hu, Shaoyan Pan, Xiaofeng Yang

Proceedings Volume 12928, 129280W (2024) https://doi.org/10.1117/12.3006851

KEYWORDS: Image registration, Deformation, Medical imaging, Convolution, Performance modeling, Image processing, Anatomy, 3D modeling, Liver, Tumors

Read Abstract +

Proceedings Article | 29 March 2024 Poster + Paper

Surface-based volumetric image reconstruction for image-guided procedures using a data-driven framework

Chih-Wei Chang, Shaoyan Pan, Zhen Tian, Tonghe Wang, Marian Axente, Joseph Shelton, Tian Liu, Justin Roper, Xiaofeng Yang

Proceedings Volume 12928, 1292822 (2024) https://doi.org/10.1117/12.3006510

KEYWORDS: Radiotherapy, Computed tomography, 3D image processing, Image restoration, Anatomy, 3D modeling, 3D image reconstruction, Deep learning, Image quality, Machine learning

Read Abstract +

Proceedings Article | 10 April 2023 Poster + Paper

Ultrasound-based dominant intraprostatic lesion classification with swin transformer

Yuheng Li, Boran Zhou, Jing Wang, Shaoyan Pan, Ashesh Jani, Tian Liu, Pretsh Patel, Xiaofeng Yang

Proceedings Volume 12470, 124700T (2023) https://doi.org/10.1117/12.2653634

KEYWORDS: Transformers, Image classification, Prostate, Deep learning, Ultrasonography, Modeling, Medical imaging

Read Abstract +

Proceedings Article | 10 April 2023 Presentation + Paper

Multi-structure segmentation on cardiac MRI using multilayer perceptron mixer network

Shaoyan Pan, Chaoqiong Ma, Chih-Wei Chang, Jacob Wynne, Justin Roper, Tian Liu, Xiaofeng Yang

Proceedings Volume 12468, 124680D (2023) https://doi.org/10.1117/12.2653944

KEYWORDS: Image segmentation, Magnetic resonance imaging, Education and training, Transformers, Medical imaging, Cardiovascular magnetic resonance imaging, Image processing, Semantics, Visualization, Myocardium

Read Abstract +

In this work, we propose MLP-Vnet, a token-based U-shaped multilayer linear perceptron-mixer (MLP-Mixer) network, incorporating a convolutional neural network for multi-structure segmentation on cardiac magnetic resonance imaging (MRI). The proposed MLP-Vnet is composed of an encoder and decoder. Taking an MRI scan as input, the semantic features are extracted by the encoder with one early convolutional block followed by four consecutive MLP-Mixer blocks. Then, the extracted features are passed to the decoder with mirrored architecture of the encoder to form a N-classes segmentation map. We evaluated our proposed network on the Automated Cardiac Diagnosis Challenge (ACDC) dataset. The performance of the network was assessed in terms of the volume- and surface-based similarities between the predicted contours and the manually delineated ground-truth contours, and computational efficiency. The volume-based similarities were measured by the Dice score coefficient (DSC), sensitivity, and precision. The surface-based similarities were measured by Hausdorff distance (HD), mean surface distance (MSD), and residual mean square distance (RMSD). The performance of the MLP-Vnet was compared with four state-of-the-art networks. The proposed network demonstrated statistically superior DSC and superior sensitivity or precision on all the three structures to the competing networks (p-value < 0.05): average DSC of 0.904, sensitivity of 0.908 and precision of 0.902 among all structures. The best surfaceased similarities were also demonstrated by the MLP-Vnet: average HD = 3.266 mm, MSD = 0.684 mm, and RMSD = 1.487 mm. Compared to the competing networks, the MLP-Vnet showed the shortest training time (7.32 hours) inference time per patient (3.12 seconds). The proposed MLP-Vnet is capable of using reasonable number of trainable parameters to solve the segmentation task on the cardiac MRI scans more quickly and accurately than the state-ofthe- art networks. This novel network could be a promising tool for accurate and efficient cardiac MRI segmentation to assist cardiac diagnosis and treatment decision making.

Proceedings Article | 7 April 2023 Poster + Paper

Mass density estimation based on single-energy computed tomography via deep learning

Yuan Gao, Chih-Wei Chang, Shaoyan Pan, Yang Lei, Tonghe Wang, Jun Zhou, Jeffrey Bradley, Tian Liu, Xiaofeng Yang

Proceedings Volume 12463, 124632X (2023) https://doi.org/10.1117/12.2654016

KEYWORDS: Artificial neural networks, Tissues, Proton therapy, Computed tomography, Performance modeling, Education and training, Deep learning, Data modeling, Lung, Calibration

Read Abstract +

Proceedings Article | 3 April 2023 Poster + Paper

Multi-organ CT segmentation using shifted-window multilayer perceptron mixer

Shaoyan Pan, Jacob Wynne, Mingzhe Hu, Tonghe Wang, Justin Roper, Pretesh Patel, Tian Liu, Xiaofeng Yang

Proceedings Volume 12466, 124661U (2023) https://doi.org/10.1117/12.2653956

KEYWORDS: Image segmentation, Windows, Computed tomography, Kidney, Head, Radiotherapy, Abdomen, Rectum, Medical imaging, Gallbladder

Read Abstract +

Proceedings Article | 3 April 2023 Poster + Paper

Deep learning-based multi-organ CT segmentation with adversarial data augmentation

Shaoyan Pan, Shao-Yuan Lo, Min Huang, Chaoqiong Ma, Jacob Wynne, Tonghe Wang, Tian Liu, Xiaofeng Yang

Proceedings Volume 12464, 124642D (2023) https://doi.org/10.1117/12.2653970

KEYWORDS: Image segmentation, Kidney, Liver, Stomach, Gallbladder, Education and training, Deep learning, Aorta, Adversarial training, Computed tomography

Read Abstract +

In this work, we propose an adversarial attack-based data augmentation method to improve the deep-learning-based segmentation algorithm for the delineation of Organs-At-Risk (OAR) in abdominal Computed Tomography (CT) to facilitate radiation therapy. We introduce Adversarial Feature Attack for Medical Image (AFA-MI) augmentation, which forces the segmentation network to learn out-of-distribution statistics and improve generalization and robustness to noises. AFA-MI augmentation consists of three steps: 1) generate adversarial noises by Fast Gradient Sign Method (FGSM) on the intermediate features of the segmentation network’s encoder; 2) inject the generated adversarial noises into the network, intentionally compromising performance; 3) optimize the network with both clean and adversarial features. The effectiveness of the AFA-MI augmentation was validated on nnUnet. Experiments are conducted segmenting the heart, left and right kidney, liver, left and right lung, spinal cord, and stomach in an institutional dataset collected from 60 patients. We firstly evaluate the AFA-MI augmentation using nnUnet and Token-based Transformer Vnet (TT-Vnet) on the test data from a public abdominal dataset and an institutional dataset. In addition, we validate how AFA-MI affects the networks’ robustness to the noisy data by evaluating the networks with added Gaussian noises of varying magnitudes to the institutional dataset. Network performance is quantitatively evaluated using Dice Similarity Coefficient (DSC) for volume-based accuracy. Also, Hausdorff Distance (HD) is applied for surface-based accuracy. On the public dataset, nnUnet with AFA-MI achieves DSC = 0.85 and HD = 6.16 millimeters (mm); and TT-Vnet achieves DSC = 0.86 and HD = 5.62 mm. On the robustness experiment with the institutional data, AFA-MI is observed to improve the segmentation DSC score ranging from 0.055 to 0.010 across all organs relative to clean inputs. AFA-MI augmentation further improves all contour accuracies up to 0.527 as measured by the DSC score when tested on images with Gaussian noises. AFA-MI augmentation is therefore demonstrated to improve segmentation performance and robustness in CT multi-organ segmentation.

Proceedings Article | 4 April 2022 Presentation + Paper

CVT-Vnet: a convolutional-transformer model for head and neck multi-organ segmentation

Shaoyan Pan, Zhen Tian, Yang Lei, Tonghe Wang, Jun Zhou, Mark McDonald, Jeffrey Bradley, Tian Liu, Xiaofeng Yang

Proceedings Volume 12033, 120333L (2022) https://doi.org/10.1117/12.2611540

KEYWORDS: Head, Image segmentation, Neck, Optic nerve, Transformers, Computer programming, Visual process modeling, Radiotherapy, Computed tomography, Cancer

Read Abstract +

Proceedings Article | 4 April 2022 Poster + Paper

Male pelvic multi-organ segmentation using V-transformer network

Shaoyan Pan, Yang Lei, Tonghe Wang, Jacob Wynne, Justin Roper, Ashesh Jani, Pretesh Patel, Jeffrey Bradley, Tian Liu, Xiaofeng Yang

Proceedings Volume 12036, 120362A (2022) https://doi.org/10.1117/12.2628064

KEYWORDS: Transformers, Computer programming, Head, Image segmentation, Prostate, Computed tomography, Rectum, Bladder, Radiotherapy, Data modeling

Read Abstract +

Proceedings Article | 15 February 2021 Presentation + Paper

Generative adversarial networks and radiomics supervision for lung lesion synthesis

Shaoyan Pan, Jessica Flores, Chen Ting Lin, J. Webster Stayman, Grace Gang

Proceedings Volume 11595, 115950O (2021) https://doi.org/10.1117/12.2582151

KEYWORDS: Lung, Solids, Databases, Optical inspection

Read Abstract +

Realistic lesion generation is a useful tool for system evaluation and optimization. Generated lesions can serve as realistic imaging tasks for task-base image quality assessment, as well as targets in virtual clinical trials. In this work, we investigate a data-driven approach for categorical lung lesion synthesis using public lung CT databases. We propose a generative adversarial network with a Wasserstein discriminator and gradient penalty to stabilize training. We further included conditional inputs such that the network can generate user-specified lesion categories. Novel to our network, we directly incorporated radiomic features in an intermediate supervision step to encourage similar textures between generated and real lesions. We calculated the network using lung lesions from the Lung Image Database Consortium (LIDC) database. Lesions are divided into two categories: solid vs. non-solid. We performed quantitative evaluation of network performance based on four criteria: 1) overfitting, in terms of structural and morphological similarity to the training data, 2) diversity of generated lesions, in terms of similarity to other generated data, 3) similarity to real lesions, in terms of distribution of example radiomics features, and 4) conditional consistency, in terms of classification accuracy using a classifier trained on the training lesions. We imposed a quantitative threshold for similarity based on visual inspection. The percentage of non-solid and solid lesions that satisfy low overfitting and high diversity is 87.1% and 70.2% of non-solid and solid lesions respectively. The distribution of example radiomics features are similar in the generated and real lesions indicated by low Kullback–Leibler divergence scores: 1.62 for non-solid lesions and 1.13 for solid lesions. Classification accuracy for the generated lesions are comparable with that for the real lesions. The proposed network presents a promising approach for generating realistic lesions with clinically relevant features crucial for the comprehensive assessment of novel medical imaging systems.

Showing 5 of 17 publications

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years