A5039055018- Artificial Intelligence in Healthcare
- Machine Learning in Healthcare
- Dementia and Cognitive Impairment Research
- Blood Pressure and Hypertension Studies
- Diabetes, Cardiovascular Risks, and Lipoproteins
- Diabetes Management and Research
- Diabetes Treatment and Management
- Biomedical Text Mining and Ontologies
- Ocular and Laser Science Research
- Cerebrovascular and Carotid Artery Diseases
- Non-Invasive Vital Sign Monitoring
- Medication Adherence and Compliance
- Alzheimer's disease research and treatments
- Diabetes Management and Education
- Cardiac Health and Mental Health
- Cardiovascular Health and Disease Prevention
- Infrared Thermography in Medicine
- Heart Rate Variability and Autonomic Control
- Neurological Disease Mechanisms and Treatments
- Cardiovascular Function and Risk Factors
- Memory and Neural Mechanisms
- Chronic Disease Management Strategies
National University of Singapore
2021-2025
National University Health System
2022-2025
Abstract This study looked at novel data sources for cardiovascular risk prediction including detailed lifestyle questionnaire and continuous blood pressure monitoring, using ensemble machine learning algorithms (MLAs). The reference conventional score compared against was the Framingham Risk Score (FRS). outcome variables were low or high based on calcium 0 100 above. Ensemble MLAs built naive bayes, random forest support vector classifier generalized linear regression, regressor stochastic...
Background: The major mechanisms of dementia and cognitive impairment are vascular neurodegenerative processes. Early diagnosis can facilitate timely interventions to mitigate progression. Objective: This study aims develop a reliable machine learning (ML) model using socio-demographics, risk factors, structural neuroimaging markers for early in multi-ethnic Asian population. Methods: consisted 911 participants from the Epidemiology Dementia Singapore (aged 60– 88 years, 49.6% male). Three...
Abstract Background Clinical trials have demonstrated that initiating oral anti-diabetic drugs (OADs) significantly reduce glycated hemoglobin (HbA1c) levels. However, variability in lifestyle modifications and OAD adherence impact on their actual effect glycemic control. Furthermore, evidence dose adjustments discontinuation of HbA1c is lacking. This study aims to use real-world data determine the initiation, up-titration, down-titration, levels, among Asian patients managed primary care....
Background The hippocampus plays a central role in cognition and hippocampal atrophy is key hallmark of Alzheimer's disease. Evidence has suggested associations between subfield volumes specific cognitive domains dementia risk. However, to our knowledge, no study examined the decline across different over time. Objective We investigated changes together with incident memory clinic cohort. Methods Associations three years ( n = 443) were analyzed using generalized estimating equations, 283)...
Abstract Background Clinical risk prediction models (CRPMs) use patient characteristics to estimate the probability of having or developing a particular disease and/or outcome. While CRPMs are gaining in popularity, they have yet be widely adopted clinical practice. The lack explainability and interpretability has limited their utility. Explainability is extent which model’s process can described. Interpretability degree user understand predictions made by model. Methods study aimed...
Patient similarity analytics has emerged as an essential tool to identify cohorts of patients who have similar clinical characteristics some specific patient interest. In this study, we propose a measure called D3K that incorporates domain knowledge and data-driven insights. Using the electronic health records (EHRs) 169,434 with either diabetes, hypertension or dyslipidaemia (DHL), construct feature vectors containing demographics, vital signs, laboratory test results, prescribed...
Type-2 diabetes mellitus (T2DM) is a medical condition in which oral medications avail to patients curb their hyperglycaemia after failed dietary therapy. However, individual responses the prescribed pharmacotherapy may differ due clinical profiles, comorbidities, lifestyles and adherence. One approach identify similar within same community predict likely response medications. This study aims present an evidence-based medication recommendation system (DMRS) underpinned by patient similarity...
Abstract Background: A Clinical Risk Prediction Model (CRPM) uses patient characteristics to estimate the probability about having or developing a particular disease and/or outcome. While CRPMs are gaining in popularity, they have yet be adopted routinely clinical practice. The lack of explainability and interpretability has limited its utility. Explainability is extent which model’s prediction process can described. Interpretability degree user understand predictions made by model.Methods:...
Abstract Evidence on the influence of patient characteristics HbA 1c treatment response for add-on medications in patients with type 2 diabetes (T2D) is unclear. This study aims to investigate predictors three (sulfonylureas (SU), dipeptidyl peptidase-4 (DPP-4) and sodium–glucose cotransporter-2 (SGLT-2) inhibitor) metformin monotherapy treated T2D. retrospective cohort was conducted using electronic health record data from six primary care clinics Singapore. A total 9748 adult T2D receiving...
Introduction: Cardiovascular disease was the top cause of deaths and disability in Singapore 2018, contributing extensively to local healthcare burden. Primary prevention identifies at-risk individuals for swift implementation preventive measures. This has been traditionally done using Singapore-adapted Framingham Risk Score (SG FRS). However, its most recent recalibration more than a decade ago. Recent changes patient demographics risk factors have undermined accuracy SG FRS, rising...