Browsing by Author "Liu, Catherine Jui-Ling"

Now showing 1 - 3 of 3

A temperature-induced and shear-reversible assembly of latanoprost-loaded amphiphilic chitosan colloids: Characterization and in vivo glaucoma treatment
(2014-07-01)
Hsiao, Meng-Hsuan
;
Chiou, Shih-Hwa
;
Larsson, Mikael
;
Hung, Kuo-Hsuan
;
Wang, Yi-Ling
Hydrogels composed of assembled colloids is a material class that is currently receiving much interest and shows great promise for use in biomedical applications. This emerging material class presents unique properties derived from the combination of nanosized domains in the form of colloidal particles with a continuous gel network and an interspersed liquid phase. Here we developed an amphiphilic chitosan based, thermogelling, shear-reversible colloidal gel system for improved glaucoma treatment and addressed how preparation procedures and loading with the anti-glaucoma drug latanoprost and commonly used preservative benzalkonium chloride influenced the mechanical properties of and drug release from the colloidal gels. The results highlight that incorporated substances and preparation procedures have effects both on mechanical properties and drug release, but that the release of drug loaded in the colloidal carriers is mainly limited by transport out of the carriers, rather than by diffusion within the gel. The developed colloidal chitosan based gels hold outstanding biomedical potential, as confirmed by the ease of preparation and administration, low cytotoxicity in MTT assay, excellent biocompatibility and lowering of intraocular pressure for 40 days in a rabbit glaucoma model. The findings clearly justify further investigations towards clinical use in the treatment of glaucoma. Furthermore, the use of this shear-reversible colloidal gel could easily be extended to localized treatment of a number of critical conditions, from chronic disorders to cancer, potentially resulting in a number of new therapeutics with improved clinical performance. (C) 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.
Automated glaucoma detection based on deep convolutional neural network
(2018-07-01)
Ko, Yu-Chieh
;
Wey, Shin-Yu
;
Lee, Chen-Yi
;
Liu, Catherine Jui-Ling
;
電機學院
Deep learning assisted detection of glaucomatous optic neuropathy and potential designs for a generalizable model
(2020-05-14)
Ko, Yu-Chieh
;
Wey, Shih-Yu
;
Chen, Wei-Ta
;
Chang, Yu-Fan
;
Chen, Mei-Ju
Purpose To evaluate ways to improve the generalizability of a deep learning algorithm for identifying glaucomatous optic neuropathy (GON) using a limited number of fundus photographs, as well as the key features being used for classification. Methods A total of 944 fundus images from Taipei Veterans General Hospital (TVGH) were retrospectively collected. Clinical and demographic characteristics, including structural and functional measurements of the images with GON, were recorded. Transfer learning based on VGGNet was used to construct a convolutional neural network (CNN) to identify GON. To avoid missing cases with advanced GON, an ensemble model was adopted in which a support vector machine classifier would make final classification based on cup-to-disc ratio if the CNN classifier had low-confidence score. The CNN classifier was first established using TVGH dataset, and then fine-tuned by combining the training images of TVGH and DrishtiGS datasets. Class activation map (CAM) was used to identify key features used for CNN classification. Performance of each classifier was determined through area under receiver operating characteristic curve (AUC) and compared with the ensemble model by diagnostic accuracy. Results In 187 TVGH test images, the accuracy, sensitivity, and specificity of the CNN classifier were 95.0%, 95.7%, and 94.2%, respectively, and the AUC was 0.992 compared to the 92.8% accuracy rate of the ensemble model. For the Drishti-GS test images, the accuracy of the CNN, the fine-tuned CNN and ensemble model was 33.3%, 80.3%, and 80.3%, respectively. The CNN classifier did not misclassify images with moderate to severe diseases. Class-discriminative regions revealed by CAM co-localized with known characteristics of GON. Conclusions The ensemble model or a fine-tuned CNN classifier may be potential designs to build a generalizable deep learning model for glaucoma detection when large image databases are not available.