Hsuan Ting Chang

Prof. Hsuan Ting Chang

orcid

Research Focus & Metrics

Academic Career Length

31Years (1996-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 March 2017 in Journal of the Optical Society of America B
Verified
Journal Article
Closed access
PDF file
CCChien-Yu Chen
Chien-Yu Chen
WLWu-Chun Li
Wu-Chun Li
HCHsuan Ting Chang
Hsuan Ting Chang
CCChih-Hao Chuang
Chih-Hao Chuang
TCTsung-Jan Chang
Tsung-Jan Chang
In this study, a computer-generated holography based on the three-dimensional (3-D) modified Gerchberg–Saxton algorithm (MGSA) for phase retrieval is proposed and demonstrated in a panoramic projection system. The point-based Fresnel transform is utilized in the MGSA to retrieve the panoramic computer-generated phase-only hologram (CGPOH) of a 3-D object considering the different viewing angles of the object. The CGPOH capable of panoramic 3-D projections can be determined using the rotation matrices and illuminating the light sources at the corresponding viewing angles. Computer simulations and optical experiments are conducted to verify that the determined CGPOH can successfully generate the panoramic projection images with various depths and at different viewing angles.

30 April 2024 in Diagnostics
Verified
Journal Article
Open access
PDF file
COChih-Ying Ou
Chih-Ying Ou
ICI-Yen Chen
I-Yen Chen
HCHsuan Ting Chang
Hsuan Ting Chang
CWChuan-Yi Wei
Chuan-Yi Wei
DLDian-Yu Li
Dian-Yu Li
Chuan-Yu ChangChuan-Yu Chang
Chuan-Yu Chang
We present a deep learning (DL) network-based approach for detecting and semantically segmenting two specific types of tuberculosis (TB) lesions in chest X-ray (CXR) images. In the proposed method, we use a basic U-Net model and its enhanced versions to detect, classify, and segment TB lesions in CXR images. The model architectures used in this study are U-Net, Attention U-Net, U-Net++, Attention U-Net++, and pyramid spatial pooling (PSP) Attention U-Net++, which are optimized and compared based on the test results of each model to find the best parameters. Finally, we use four ensemble approaches which combine the top five models to further improve lesion classification and segmentation results. In the training stage, we use data augmentation and preprocessing methods to increase the number and strength of lesion features in CXR images, respectively. Our dataset consists of 110 training, 14 validation, and 98 test images. The experimental results show that the proposed ensemble model achieves a maximum mean intersection-over-union (MIoU) of 0.70, a mean precision rate of 0.88, a mean recall rate of 0.75, a mean F1-score of 0.81, and an accuracy of 1.0, which are all better than those of only using a single-network model. The proposed method can be used by clinicians as a diagnostic tool assisting in the examination of TB lesions in CXR images.

18 December 2019 in Photonics
Verified
Journal Article
Open access
PDF file
HCHsuan Ting Chang
Hsuan Ting Chang
YWYao-Ting Wang
Yao-Ting Wang
CCChien-Yu Chen
Chien-Yu Chen
We propose an angle multiplexing method for optics-based image encryption using a phase-only computer-generated hologram (POCGH) in the tilted Fresnel transform (TFrT) domain. Modified Gerchberg-Saxton algorithms, based on the three types of rotation manipulation in both the hologram and reconstruction planes, are used with their corresponding TFrT parameters to extract the phase-only functions (POFs) of the target images. All the extracted POFs are then phase-modulated and summed to obtain the final POCGH, which is capable of multiplexing and avoiding overlap in the reconstructed images. The computer simulation results show that the images corresponding to the various rotation manipulations at the hologram and image reconstruction planes can be successfully restored with high correlation coefficients. Due to the encrypted nature of the multiplexed images, a higher system security level can be achieved, as the images can only be correctly displayed when all the required parameters in the TFrT are available. The angle sensitivity on the image quality for each manipulation is also investigated.

29 March 2023 in Diagnostics
Verified
Journal Article
Open access
PDF file
TLTsung-Chun Lee
Tsung-Chun Lee
CAClara Lavita Angelina
Clara Lavita Angelina
PKPradermchai Kongkam
Pradermchai Kongkam
Hsiu-Po WangHsiu-Po Wang
Hsiu-Po Wang
RRRungsun Rerknimitr
Rungsun Rerknimitr
HCHsuan Ting Chang
Hsuan Ting Chang
Accurate classification of pancreatic cystic lesions (PCLs) is important to facilitate proper treatment and to improve patient outcomes. We utilized the convolutional neural network (CNN) of VGG19 to develop a computer-aided diagnosis (CAD) system in the classification of subtypes of PCLs in endoscopic ultrasound-guided needle-based confocal laser endomicroscopy (nCLE). From a retrospectively collected 22,424 nCLE video frames (50 videos) as the training/validation set and 11,047 nCLE video frames (18 videos) as the test set, we developed and compared the diagnostic performance of three CNNs with distinct methods of designating the region of interest. The diagnostic accuracy for subtypes of PCLs by CNNs with manual, maximal rectangular, and U-Net algorithm-designated ROIs was 100%, 38.9%, and 66.7% on a per-video basis and 88.99%, 73.94%, and 76.12% on a per-frame basis, respectively. Our per-frame analysis suggested differential levels of diagnostic accuracy among the five subtypes of PCLs, where non-mucinous PCLs (serous cystic neoplasm: 93.11%, cystic neuroendocrine tumor: 84.31%, and pseudocyst: 98%) had higher diagnostic accuracy than mucinous PCLs (intraductal papillary mucinous neoplasm: 84.43% and mucinous cystic neoplasm: 86.1%). Our CNN demonstrated superior specificity compared to the state-of-the-art for the classification of mucinous PCLs (IPMN and MCN), with high specificity (94.3% and 92.8%, respectively) but low sensitivity (46% and 45.2%, respectively). This suggests the complimentary role of CNN-enabled CAD systems, especially for clinically suspected mucinous PCLs.

12 January 2026 in Journal of Biomedical Science
Verified
Journal Article
Open access
PDF file
CAClara Lavita Angelina
Clara Lavita Angelina
FXFu-Ren Xiao
Fu-Ren Xiao
SVSunil Vyas
Sunil Vyas
PYPan-Chyr Yang
Pan-Chyr Yang
HCHsuan Ting Chang
Hsuan Ting Chang
YLYuan Luo
Yuan Luo
Background: Accurate classification and segmentation of brain tumors in MRI scans are essential for diagnosis and treatment planning. However, the heterogeneous morphology of brain tumors, including irregular shapes, sizes, and spatial variability, makes this task highly challenging. Traditional convolutional neural networks (CNNs) lack rotational and translational invariance, which limits their ability to generalize across different orientations. Methods: This study introduces a geometric deep learning framework called Modified Special Euclidean (Mod-SE(2)), which integrates geometric priors to enhance spatial consistency and reduce reliance on data augmentation. By incorporating symmetry-preserving group convolutions and spatial priors, Mod-SE(2) improves the robustness in tumor classification (namely Mod-Cls-SE(2)) and segmentation (mentioned as Mod-Seg-SE(2)). Unlike conventional CNNs, geometric deep learning encodes roto-translation symmetry directly into the architecture. This addresses the spatial variability and orientation sensitivity that are common in MRI-based diagnostics. Mod-SE(2) was evaluated on three MRI datasets and two other medical image datasets for classification and segmentation tasks. It incorporates lifting layers, group convolutions, and feature recalibration. It was benchmarked against U-Net, NN U-Net, VGG16, VGG19, and ResNet architectures. Results: Mod-Cls-SE(2) achieved an average classification accuracy of 0.914, outperforming ResNet101 with 0.682, VGG16 with 0.705, and their variants. In the binary classification of five tumor types (AVM, Meningioma, Pituitary, Metastases, and Schwannoma) from the private dataset, the model achieved an accuracy of 0.935 and a precision of 0.960 for pituitary tumors and a precision of 0.96. For segmentation tasks, Mod-Seg-SE(2) achieved a dice coefficient of 0.9503 and an IoU of 0.9616 on the BraTS2020 dataset. This result exceeds those of U-Net and NN U-Net with dice scores of 0.797 and 0.815, respectively. The model also reduced inference time and demonstrated strong computational performance. Conclusions: Mod-SE(2) uses geometric priors to improve the spatial consistency, efficiency, and interpretability in brain tumor analysis. Its symmetry-aware design enables better generalization across tumor shapes and outperforms traditional methods across all key metrics. The Mod-SE(2) CNN ensures accurate boundary delineation, supporting neurosurgical planning, intraoperative navigation, and downstream applications such as Monte Carlo-based radiotherapy simulations and PET-MRI co-registration. Future work will extend the model to 3D volumes and validate its clinical readiness.

26 January 2017 in Applied Optics
Verified
Journal Article
Closed access
PDF file
HCHsuan Ting Chang
Hsuan Ting Chang
JSJ.-W. Shui
J.-W. Shui
KLK.-P. Lin
K.-P. Lin
In this paper, a joint multiple-image encryption and multiplexing system, which utilizes both the nonnegative matrix factorization (NMF) scheme and digital holography, is proposed. A number of images are transformed into noise-like digital holograms, which are then decomposed into a defined number of basis images and a corresponding weighting matrix using the NMF scheme. The determined basis images are similar to the digital holograms and appear as noise-like patterns, which are then stored as encrypted data and serve as the lock in an encryption system. On the other hand, the column vectors in the weighting matrix serve as the keys for the corresponding plain images or the addresses of the multiplexed images. Both the increased uniformity of the column weighting factors and the parameters used in the digital holography enhance the security of the distributed keys. The experimental results show that the proposed method can successfully perform multiple-image encryption with high-level security.