Isabelle Xin Yu Lee

Isabelle Xin Yu Lee

Auto-generated profile, based on publicly available information

Isabelle Xin Yu Lee

Department of Tissue Engineering & Cell Therapy, Singapore Eye Research Institute , Singapore 169856

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
29 January 2025 in British Journal of Ophthalmology
Verified
Journal Article
Open access
CTCalesta Hui Yi Teo
Calesta Hui Yi Teo
CLChang Liu
Chang Liu
ILIsabelle Xin Yu Lee
Isabelle Xin Yu Lee
MLMolly Tzu-Yu Lin
Molly Tzu-Yu Lin
FLFengyi Liu
Fengyi Liu
YLYu Chi Liu
Yu Chi Liu
Background/aims: To identify the risk factors for neuropathic corneal pain (NCP) following corneal refractive surgery and to report its clinical manifestations, imaging and proteomic characteristics.Methods: This 1 year prospective cohort study included 100 eyes that underwent small incision lenticule extraction (SMILE) or laser-assisted in situ keratomileusis (LASIK). Ocular surface assessments, in-vivo confocal microscopy scans, tear neuromediators and proteomics analyses were performed. NCP was assessed using the ocular pain assessment survey. Univariate and multivariate analyses were conducted to identify the risk factors associated with postoperative NCP.Results: The incidence of NCP was 13.3% and 10.5% after SMILE and LASIK, respectively (p=0.70). In SMILE, preoperative manifest refractive spherical equivalent (MRSE) and spherical power (both p=0.02) were significantly higher in the NCP compared with the non-NCP group. In LASIK, NCP eyes had a significantly lower corneal nerve fibre length (CNFL) (p=0.02), lower nerve fractal dimension (p=0.003), higher nerve fibre width (p=0.04) and larger neuroma area (p=0.04) than non-NCP eyes. In SMILE, higher preoperative MRSE was a significant risk factor for postoperative NCP (95% CI: 0.48—1.96, p=0.04). An MRSE greater than −8.0 diopter was 9.57 times more likely to develop postoperative NCP (OR=9.57, p=0.002). In LASIK, lower preoperative corneal nerve fibre density (95% CI:0.13—1.11, p=0.05) and CNFL (95% CI:0.09—1.25, p=0.05) were significant risk factors for postoperative NCP. Significant increases in tear nerve growth factor, calcitonin gene-related peptide, Frizzled class receptor 7 and nucleoside-diphosphate kinase three were observed in postoperative NCP.Conclusions: The reported characteristics and risk factors would identify patients susceptible to NCP after corneal refractive surgery.

20 November 2025 in Transplantation and Cellular Therapy
Verified
Journal Article
Closed access
CLChang Liu
Chang Liu
XWXinyue Wang
Xinyue Wang
SYSharon Wan Jie Yeo
Sharon Wan Jie Yeo
ILIsabelle Xin Yu Lee
Isabelle Xin Yu Lee
YLYeh Ching Linn
Yeh Ching Linn
LLLi Lim
Li Lim

28 January 2026 in The Journal of Clinical Endocrinology & Metabolism
Verified
Journal Article
Closed access
SCSuresh Rama Chandran
Suresh Rama Chandran
JSJasmine Koh Shimin
Jasmine Koh Shimin
YLYu Chi Liu
Yu Chi Liu
ILIsabelle Xin Yu Lee
Isabelle Xin Yu Lee
MOMay Zin Oo
May Zin Oo
YLYu Chi Liu
Yu Chi Liu
Context: The relationship between corneal neuropathy measured by in vivo confocal microscopy (IVCM) and continuous glucose monitoring (CGM) remains unclear. Objective: We describe the associations between corneal neuropathy and CGM metrics. Methods: A single-center cross-sectional study recruited people with type 2 diabetes. All participants wore a 14-day blinded Freestyle Libre Pro CGM and underwent corneal IVCM. Multiple generalized estimating equation models were created to study the association of independent variables (HbA1c and CGM metrics) on dependent variables (corneal nerve fiber and cellular metrics) adjusted for age, sex, ethnicity, hypertension, hyperlipidemia, body mass index (BMI), and diabetes duration. Benjamini–Hochberg procedure controlled for multiplicity. Results: Of 307 recruited people, 277 were included in the analysis (median age 58 years, 49.5% female, 60.6% Chinese, BMI 26.6 kg/m², diabetes duration 12 years, HbA1c 8.1%). HbA1c associated negatively (β −0.609, pseudoR² = 0.068, P = .013) with corneal nerve fiber density and positively (β .003, pseudoR² = 0.066, P = .027) with corneal nerve fiber width (CNFW). Time-in-range (TIR) and time-in-tight-range (TITR) negatively (both β −.006; pseudoR² = 0.075 and 0.066, respectively; both P < .03), and time-above-range (TAR) (β .006, pseudoR² = 0.073, P = .021) positively, associated with CNFW. Among corneal cellular metrics, TAR (β .088, pseudoR² = 0.055, P = .004), mean glucose (β .049, pseudoR² = 0.055, P = .004), and glucose variability (SD, mean amplitude of glycemic excursions, and mean of daily differences: β .13, 0.046, and .164; pseudoR²= 0.053, 0.044, and 0.060, respectively, all P < .03) associated positively with average epithelial size, while TIR and TITR had a negative association (β −.084 and −.104; pseudoR² = 0.052 and 0.057, respectively, both P < .01). Conclusion: CGM measures, including mean glycemia, TIR, TITR, and glucose variability, were significantly associated with corneal neuropathy, supporting the role of CGM metrics in optimizing glycemia.

4 November 2026 in British Journal of Ophthalmology
Verified
Journal Article
Open access
PDF file
MYMingyi Yu
Mingyi Yu
VFValencia Foo
Valencia Foo
NSNicole Sie
Nicole Sie
PKPierre Deman Kibleur
Pierre Deman Kibleur
ADAurélien Ducournau
Aurélien Ducournau
YLYu Chi Liu
Yu Chi Liu
Aims To describe and validate a deep learning (DL) model for fully automatic assessment of corneal endothelium cell (CEC) parameters from specular microscopy images. Methods A total of 896 images were collected retrospectively from 440 subjects, including healthy and diseased corneas with varying image quality. Five DL models with different architectures were trained to extract eight key CEC characteristics, including cell density (CD), coefficient of variation (CV), pachymetry, average cell area (Ave), cell area standard deviation (SD), hexagonality coefficient (HEX), maximum (Max) and minimum (Min) cell area. Ground truth (GT) was derived from built-in software of the specular microscopy (Konan NSP-9900) with manual cell-centre annotation. An independent dataset comprising 100 images from 100 subjects was used for external validation. Results Among all tested models, the ResNet V.1 101 model achieved the best performance with the lowest mean squared error (MSE) of 0.24. The model demonstrated high accuracy in quantifying CEC parameters, particularly for CV, Ave, SD, Max and Min, with deviations ranging from 0.93% to 1.93%. The mean deviation was 3.10%±2.74% across all parameters. DL prediction and GT showed strong and significant correlations for CD, Ave, SD, Max and Min (r=0.89–0.99; all p<0.001), with excellent agreement for CD (intraclass correlation coefficient 0.964 (95% CI 0.944 to 0.976), p<0.001). External validation showed comparable predictive performance with a mean MSE of 0.26 and a mean deviation of 4.39%±2.60%. Conclusions The proposed DL model showed robust performance in quantifying eight key CEC parameters across specular microscopy images with varying corneal health conditions and image quality.