|
A New Fast-Response VA-FFS LC Device Using Three-Electrode Design with Improved Transmission for VR Applications
Professor Wing-Kit Choi
Graduate Institute of Photonics and Optoelectronics, National Taiwan University
臺灣大學光電所 蔡永傑教授
We have proposed a new three-electrode VA-FFS liquid crystal device structure that improves optical transmission while maintaining the fast response required for VR/AR display applications.
Liquid crystal devices with very fast response time are important for applications such as virtual reality (VR) and augmented reality (AR) displays, where high-speed operation is required. In this work, we propose a new design to enhance the optical transmission of fast-response VA-FFS (Vertically Aligned Fringe-Field Switching) liquid crystal devices while maintaining their fast switching characteristics.
In conventional 3D VA-FFS structures, disclinations or “virtual walls” often appear near the center of the pixel, which can reduce the brightness of the switched state. To address this issue, we introduce a three-electrode design that generates additional fringing electric fields to help switch LC molecules in the central region of the pixel.
Simulation results show that the proposed design can effectively reduce the dark disclination region and improve the transmission of the device while maintaining the fast response characteristics of VA-FFS LC modes. These results suggest that the proposed structure provides a promising approach for high-speed liquid crystal devices used in VR/AR display systems.
|

|
|
Fig. 1. Proposed three-electrode VA-FFS structure with a modified common electrode containing a central opening and an additional bottom pixel electrode with opposite polarity, enabling stronger electric fields in the central region of the pixel.
|
|

|
|
Fig. 2. Comparison of liquid crystal switching and transmission profiles between the conventional and proposed structures. The three-electrode design activates the central pixel region and reduces the dark area, resulting in improved optical transmission.
|
Reference:
[1] W.K. Choi, B.K. Tseng, and C.H. Tung, “A New Fast-Response VA-FFS LC Device Using Three-Electrode Design with Improved Transmission for VR Applications,” SID Symposium Digest of Technical Papers, pp. 2111–2113, 2025.
Deep Learning for Retinal Lesion Detection in Multimodal Ophthalmic Imaging
Professor Hsiang-Chieh Lee
Graduate Institute of Photonics and Optoelectronics, National Taiwan University
臺灣大學光電所 李翔傑教授
This study introduces a high-performance deep learning framework for the automated detection and classification of retinal lesions, with a specific focus on myopic choroidal neovascularization (mCNV). Distinguishing between active and inactive mCNV is vital to preventing irreversible vision loss, yet traditional diagnostic methods such as fluorescein angiography are invasive and time-consuming. The proposed methodology leverages a multimodal approach by integrating Optical Coherence Tomography (OCT) and OCT Angiography (OCTA) images, providing both structural and functional retinal insights. To address the limitations of conventional Convolutional Neural Networks (CNNs) in capturing long-range dependencies, the framework adopts a Swin Transformer architecture featuring a hierarchical shifted-window design. Preliminary results using a VGG16 architecture on single-modality OCT data demonstrate a testing accuracy of 92%, a precision of 89%, and an AUC of 93%. Grad-CAM visualizations confirm the model's ability to accurately localize lesion-related regions. The research concludes that transitioning to a Swin Transformer-based multimodal system will significantly enhance diagnostic robustness and clinical utility. This work highlights the potential of integrated multimodal learning to support large-scale ophthalmic screening and clinical decision-making.
|

|
|
Fig. 1. Multimodal images (a) OCT_active/OCTA_active (b) OCT_active/OCTA_inactive (c) OCT_inactive/OCTA_active (d) OCT_inactive/OCTA_inactive
|
|

|
|
Fig. 2. Grad-CAM visualization of OCT images
|
|