Publications by authors named "Linchuan Xu"
Purpose: We constructed a multitask learning model (latent space linear regression and deep learning [LSLR-DL]) in which the 2 tasks of cross-sectional predictions (using OCT) of visual field (VF; central 10°) and longitudinal progression predictions of VF (30°) were performed jointly via sharing the deep learning (DL) component such that information from both tasks was used in an auxiliary manner (The Association for Computing Machinery's Special Interest Group on Knowledge Discovery and Data Mining [SIGKDD] 2021). The purpose of the current study was to investigate the prediction accuracy preparing an independent validation dataset.
Design: Cohort study.
Article Synopsis
- The text discusses a methodology for detecting changes in data streams, specifically in COVID-19 case numbers, to provide early warning signals for potential epidemics.
- A new information-theoretic concept called differential minimum description length change statistics (D-MDL) is introduced for measuring change signals in the data.
- The study shows that D-MDL can effectively identify significant increases or decreases in cases up to six days in advance, which could greatly enhance response times to epidemics and is tied to factors like the basic reproduction number (R0) and social distancing measures.
Purpose: To investigate whether OCT measurements can improve visual field (VF) trend analyses in glaucoma patients using the deeply regularized latent-space linear regression (DLLR) model.
Design: Retrospective cohort study.
Participants: Training and testing datasets included 7984 VF results from 998 eyes of 592 patients and 1184 VF results from 148 eyes of 84 patients with open-angle glaucoma, respectively.
Article Synopsis
- - The study aimed to predict the visual field (VF) of glaucoma patients using deep learning techniques combined with optical coherence tomography (OCT) measurements, focusing on the central 10° of vision.
- - Researchers conducted a cross-sectional study with 505 eyes from glaucoma patients and 86 eyes from normal subjects, developing two convolutional neural network models to predict VF sensitivity based on various retinal layer thicknesses.
- - The CNN-TR model showed the best prediction performance with a lower root mean squared error (6.32 dB), outperforming other methods like CNN-PR, support vector regression, and multiple linear regression in achieving more accurate visual field predictions.