Jie Yin Yee

Jie Yin Yee

Auto-generated profile, based on publicly available information

Jie Yin Yee

Institute of Mental Health, Singapore

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
14 February 2025 in Translational Psychiatry
Verified
Journal Article
Open access
PDF file
JYJie Yin Yee
Jie Yin Yee
SPSer-Xian Phua
Ser-Xian Phua
YSYuen Mei See
Yuen Mei See
AAAnand Kumar Andiappan
Anand Kumar Andiappan
WGWilson Wen Bin Goh
Wilson Wen Bin Goh
JLJimmy Lee
Jimmy Lee
We apply machine learning techniques to navigate the multifaceted landscape of schizophrenia. Our method entails the development of predictive models, emphasizing peripheral inflammatory biomarkers, which are classified into treatment response subgroups: antipsychotic-responsive, clozapine-responsive, and clozapine-resistant. The cohort comprises 146 schizophrenia patients (49 antipsychotics-responsive, 68 clozapine-responsive, 29 clozapine-resistant) and 49 healthy controls. Protein levels of immune biomarkers were quantified using the Olink Target 96 Inflammation Panel (Olink®, Uppsala, Sweden). To predict labels, a support vector machine (SVM) classifier is trained on the Olink®data matrix and evaluated via leave-one-out cross-validation. Associated protein biomarkers are identified via recursive feature elimination. We constructed three separate predictive models for binary classification: one to discern healthy controls from individuals with schizophrenia (AUC = 0.74), another to differentiate individuals who were responsive to antipsychotics (AUC = 0.88), and a third to distinguish treatment-resistant individuals (AUC = 0.78). Employing machine learning techniques, we identified features capable of distinguishing between treatment response subgroups. In this study, SVM demonstrates the power of machine learning to uncover subtle signals often overlooked by traditional statistics. Unlike t-tests, it handles multiple features simultaneously, capturing complex data relationships. Chosen for simplicity, robustness, and reliance on strong feature sets, its integration with explainable AI techniques like SHapely Additive exPlanations enhances model interpretability, especially for biomarker screening. This study highlights the potential of integrating machine learning techniques in clinical practice. Not only does it deepen our understanding of schizophrenia’s heterogeneity, but it also holds promise for enhancing predictive accuracy, thereby facilitating more targeted and effective interventions in the treatment of this complex mental health disorder.

1 September 2016 in Psychiatry Research
Verified
Journal Article
Open access
JYJie Yin Yee
Jie Yin Yee
YSYuen Mei See
Yuen Mei See
NRNur Amirah Abdul Rashid
Nur Amirah Abdul Rashid
SNSasi Neelamekam
Sasi Neelamekam
JLJimmy Lee
Jimmy Lee
Total vitamin D levels had been commonly reported to be lowered in patients with chronic psychotic illnesses in countries from the higher latitudes. However, studies on patients with first episode psychosis (FEP) are limited. In this study we investigated serum concentrations of total and bioavailable vitamin D levels in FEP patients compared to healthy controls and the association between symptom severity and vitamin D components. A total of 31 FEP patients and 31 healthy controls were recruited from Institute of Mental Health, Singapore. FEP patients were identified using Structured Clinical Interview for DSM-IV Axis I disorders (SCID-1) and severity symptoms were assessed using the positive and negative syndrome scale (PANSS). Sera from participants were analyzed for total vitamin D, vitamin D-binding protein (DBP) and bioavailable vitamin D. Linear regressions were performed to examine the associations between serum total and bioavailable vitamin D and the PANSS subscales. Current study noted a significantly lower bioavailable vitamin D was in the FEP group and an association between bioavailable vitamin D and negative symptoms in FEP patients in a population with a consistent supply of sun exposure throughout the year.

1 October 2017 in Brain, Behavior, and Immunity
Verified
Journal Article
Open access
JYJie Yin Yee
Jie Yin Yee
MNMilawaty Nurjono
Milawaty Nurjono
WNWai Yee Ng
Wai Yee Ng
STStephanie Ruth Teo
Stephanie Ruth Teo
TLTih-Shih Lee
Tih-Shih Lee
JLJimmy Lee
Jimmy Lee
There is a growing interest in the association between schizophrenia and the activation of inflammatory system with signs of acute phase (AP) response. Majority of such studies had focused on C-reactive protein (CRP). The aims of the present study were (i) to examine the gene expression profiles of other acute phase proteins (APP), namely haptoglobin (HP), alpha-1 antitrypsin (A1T), and alpha-2 macroglobulin (A2M) in patients with first episode psychosis (FEP) over a period of three months and (ii) to explore the association between APP levels and severity of symptoms. In this study, HP, A1T and A2M gene expression levels from whole blood were measured at recruitment, 1- and 3-month follow-up visits using quantitative PCR (qPCR) in 43 patients with FEP and in 57 healthy controls. Diagnoses was ascertained on the Structured Clinical Interview for DSM-IV-TR. Severity of symptoms in patients was assessed on the Positive and Negative Syndrome Scale (PANSS) and a previously validated 5-factor PANSS structure was applied in the subsequent analyses. The FEP sample comprised of 28 (65.1%) individuals with schizophrenia, 12 (27.9%) with schizophreniform disorder and 3 (7%) with schizoaffective disorder. HP gene expression level was noted to be significantly higher in patients than controls at all three time points: recruitment (P=0.049), 1-month follow up (P=0.002) and 3-month follow up (P=0.005). PANSS positive, depression, and excitement symptom factors showed significant associations with HP (P=0.002), A1T (P=0.016) and A2M (P=0.034), respectively. These findings remained significant after controlling for age, gender, smoking status and accumulated chlorpromazine dosage. The current study provides information on HP, A1T and A2M gene expression profiles in FEP patients and their associations with psychopathology. This provides support for the hypothesis that inflammation is related to schizophrenia and further encourages studies on immune-inflammatory markers to understand the relationship between inflammation and schizophrenia.

23 March 2018 in International Journal of Neuropsychopharmacology
Verified
Journal Article
Open access
PDF file
JYJie Yin Yee
Jie Yin Yee
TLTih-Shih Lee
Tih-Shih Lee
JLJimmy Lee
Jimmy Lee
Identifying biomarkers to enrich prognostication and risk predictions in individuals at high risk of developing psychosis will enable stratified early intervention efforts. Brain-derived neurotrophic factor has been widely studied in schizophrenia and in first-episode psychosis with promising results. The aim of this study was to examine the levels of serum brain-derived neurotrophic factor between healthy controls and individuals with ultra-high risk of psychosis. A sample of 106 healthy controls and 105 ultra-high risk of psychosis individuals from the Longitudinal Youth at Risk Study was included in this study. Ultra-high risk of psychosis status was determined using the Comprehensive Assessment of At-Risk Mental State at recruitment. Calgary Depression Scale for Schizophrenia was used to assess the severity of depression. All participants were followed up for 2 years, and ultra-high risk of psychosis remitters were defined by ultra-high risk of psychosis individuals who no longer fulfilled Comprehensive Assessment of At-Risk Mental State criteria at the end of the study period. Levels of brain-derived neurotrophic factor were measured in the serum by enzyme-linked immunosorbent assay method. The ultra-high risk of psychosis group had significantly higher baseline levels of serum brain-derived neurotrophic factor compared with the control group (3.7 vs 3.3 ng/mL, P=.018). However, baseline levels of serum brain-derived neurotrophic factor did not predict the development of psychosis (OR=0.64, CI=0.40–1.02) or remission (OR=0.83, CI=0.60–1.15) from ultra-high risk of psychosis status. Findings from our study did not support a role for serum brain-derived neurotrophic factor in predicting outcomes in ultra-high risk of psychosis individuals. However, the finding of higher levels of serum brain-derived neurotrophic factor in ultra-high risk of psychosis individuals deserves further study.

3 November 2023 in Asian Journal of Psychiatry
Verified
Journal Article
Open access
KLKeane Lim
Keane Lim
JYJie Yin Yee
Jie Yin Yee
YSYuen Mei See
Yuen Mei See
BNBoon Tat Ng
Boon Tat Ng
SZShushan Zheng
Shushan Zheng
MLMax Lam
Max Lam
Treatment-resistant schizophrenia (TRS) affects a substantial proportion of patients who do not respond adequately to antipsychotic medications, yet the underlying biological mechanism remains poorly understood. This study investigates the link between the genetic predisposition to schizophrenia and TRS. 857 individuals diagnosed with schizophrenia were divided into TRS (n = 142) and non-TRS (n = 715) based on well-defined TRS criteria. Polygenic risk scores (PRS) were calculated using schizophrenia genome-wide association summary statistics from East-Asian and European ancestry populations. PRS was estimated using both P-value thresholding and Bayesian framework methods. Logistic regression analyses were performed to differentiate between TRS and non-TRS individuals. The schizophrenia PRS derived from the East-Asian training dataset effectively distinguished between TRS and non-TRS individuals (R2 = 0.029, p = 4.86 ×10-5, pT = 0.1, OR = 1.52, 95% CI = 1.242-1.861), with higher PRS values observed in the TRS group. Similar PRS analysis was conducted based on the European ancestry GWAS summary statistics, but we found superior prediction based on the East-Asian ancestry discovery data. This study reveals an association between common risk variants for schizophrenia and TRS status, suggesting that the genetic burden of schizophrenia may partly contribute to treatment resistance in individuals with schizophrenia. These findings propose the potential use of genetic risk factors for early TRS identification and timely access to clozapine. However, the ancestral background of the discovery sample is crucial for successfully implementing PRS in clinical settings.

15 July 2025 in Schizophrenia
Verified
Journal Article
Open access
PDF file
JKJordon Junyang Kho
Jordon Junyang Kho
SSShangzheng Song
Shangzheng Song
STSamuel Ming Xuan Tan
Samuel Ming Xuan Tan
NFNur Hikmah Fitriyah
Nur Hikmah Fitriyah
MLMatheus Calvin Lokadjaja
Matheus Calvin Lokadjaja
WGWilson Wen Bin Goh
Wilson Wen Bin Goh
Mental illnesses often manifest through behavioral changes, with speech serving as a key medium for expressing thoughts and emotions. The use of computational linguistics on speech data in mental illnesses is a promising approach to uncover objective biomarkers for the early detection of mental illnesses. This study analyzed speech transcripts from 80 youths at ultra-high risk of psychosis (UHR) and 329 healthy controls, examining text features such as sentiment variability, cohesion, lexical sophistication, morphology, syntactic sophistication, and lexical diversity. Factor analysis revealed five key linguistic themes: Sentiment Intensity and Variability, Linguistic Register Alignment, Phonographic Uniqueness and Recognizability, Morphological Complexity and Imageability, and Lexical Richness and Typicalness. Regression analysis indicated UHR speech is characterized by diminished sentiment variability (β = –0.07), deviation from linguistic registers (β = –0.16), fewer phonographic neighbors (β = –0.11), lower morphological complexity (β = –0.36), and more predictable lexical structures (β = 0.05). Optimized machine learning (ML) models trained on Boruta-selected features achieved a mean AUC of 0.70. Our findings highlight the potential of sentiment and linguistic analyses in speech for training ML models to aid in early detection and monitoring of mental health conditions.