NKNicholas I-Hsien Kuo

Nicholas I-Hsien Kuo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Nicholas I-Hsien Kuo has a total of 59 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    PRIME-CVD: Paired Synthetic Cohort and Structured EMR-Style Data Assets for Education in Cardiovascular Risk Modelling

    by Nicholas I-Hsien Kuo, Marzia Hoque Tania, Blanca Gallego, Louisa R. Jorm

    In recent years, progress in medical informatics and machine learning has been accelerated by the availability of openly accessible benchmark datasets. However, patient-level electronic medical record (EMR) data are rarely available for teaching or methodological development due to privacy, governance, and re-identification risks. This has limited opportunities for students from healthcare and clinical backgrounds to gain hands-on experience in coding, data management, and cardiovascular risk modelling. Here we introduce PRIME-CVD, a parametrically rendered informatics medical environment designed explicitly for medical education. PRIME-CVD comprises two openly accessible synthetic data assets, each representing 50,000 adults undergoing primary prevention for cardiovascular disease and parameterised to reproduce the statistical characteristics of a derivation cohort from the Australian New South Wales Lumos linked-data resource. The datasets are generated entirely from a user-specified causal directed acyclic graph parameterised using characteristics of the reference cohort, publicly available Australian population statistics and published epidemiologic effect estimates, rather than from patient-level EMR data or trained generative models. Data Asset 1 provides a clean, analysis-ready cohort suitable for teaching exploratory analysis, stratification, statistical coding, and survival modelling, while Data Asset 2 systematically transforms the same cohort into a structured relational, EMR-style database with deliberately introduced data-quality challenges, including missingness, heterogeneous terminology, inconsistent measurement units, and cross-table linkage requirements. Together, these complementary assets enable healthcare and clinical learners to develop practical skills in coding, data cleaning, harmonisation, cohort reconstruction, and cardiovascular risk modelling without exposing sensitive information. Because all individuals and events are generated de novo, PRIME-CVD preserves realistic subgroup imbalance and risk gradients while ensuring negligible disclosure risk. Although the released PRIME-CVD cohort represents adults with no prior history of cardiovascular disease undergoing primary-prevention risk assessment, the complete data-assembly pipeline is provided through the dedicated GitHub repository. Educators may adapt the supplied code, parameters, and data-generating assumptions to construct alternative synthetic cohorts suited to their own teaching objectives. PRIME-CVD is released under a Creative Commons Attribution 4.0 licence to support reproducible and scalable health data science education.

  • Open Access

    Dynamic prediction of cardiovascular risk from linked primary care records in New South Wales, Australia: protocol for a retrospective prognostic modelling study

    by Nicholas I-Hsien Kuo, Heidi Welberry, Sanja Lujic, Anna Campain, Daniel Wilson, Katie Harris, Michael Kidd, Patricia Correll, David Peiris, Louisa Jorm

    Introduction: Cardiovascular disease (CVD) remains a leading cause of preventable morbidity and mortality. Most existing CVD prediction tools use a baseline survival modelling framework, in which each predictor is represented by a single value recorded at the start of follow-up. Although risk calculators can be re-run using updated values, these models do not explicitly use the longitudinal, irregularly recorded information in primary-care electronic medical records (EMRs). Dynamic modelling approaches have the potential to generate updated CVD risk estimates at each general practitioner (GP) encounter, supporting both risk assessment and monitoring of changes over time. This protocol outlines the development and validation of transformer-based sequence models for encounter-level prediction of 5-year CVD risk, compared with dynamic landmark survival models and benchmarked against static CVD prediction tools. Methods and analysis: The study will use the New South Wales (NSW) Lumos linked health data asset, a large, privacy-preserving linkage of GP EMRs with hospital admissions, mortality and other administrative datasets across NSW, Australia. Eligible GP encounters will be included from 2018 onwards. The primary outcome is first fatal or non-fatal CVD event within 5 years of each eligible encounter, identified using ICD-10-AM codes in linked hospital and mortality data. Predictor variables include routinely collected demographics, chronic conditions, clinical measurements and medications, as well as healthcare utilisation patterns available at or before each encounter. Sex-specific dynamic landmark survival models (such as Cox proportional hazards models) with time-updated covariates (full and least absolute shrinkage and selection operator (LASSO)-regularised) will be developed alongside sex-specific transformer-based sequence models, fine-tuned for 5-year risk prediction at each encounter. Internal validation will use repeated resampling for landmark survival models with patient-level and temporal separation for sequence models. Geographic transportability will be evaluated using internal–external validation across primary health networks (PHNs). Performance will be evaluated using discrimination, calibration, decision-curve analysis and subgroup analyses and will be compared with existing static predictive models. Missing-data will be imputed where necessary: a range of imputation approaches will be compared to optimise computational efficiency but also support predictive generalisability. Ethics and dissemination: This study is conducted under NSW Health governance arrangements with approval from the NSW Population and Health Services Research Ethics Committee (2019/ETH00660). Analyses will be performed within a secure data environment using de-identified linked data. Results will be reported in accordance with TRIPOD-AI guidance and disseminated through peer-reviewed publications, scientific conferences and policy and consumer forums.


  • Open Access

    Limits of Generative Pre-Training in Structured EMR Trajectories with Irregular Sampling

    by Nicholas I-Hsien Kuo, Blanca Gallego, Louisa R. Jorm

    Foundation models are commonly trained using autoregressive pretraining and are increasingly applied to structured electronic medical records (EMRs) for clinical prediction. However, autoregressive pre-training was originally developed for sequence generation in text, and its suitability for capturing longitudinal clinical context in structured, irregularly sampled EMRs remains insufficiently evaluated. In this work, we return to first principles by examining the patient trajectories generated by autoregressive models. We train two models – a sequence-to-sequence LSTM and a reduced ETHOS-style Transformer – that share core architectural and training principles with state-of-the-art EHR foundation models. Using longitudinal antiretroviral therapy (ART) for HIV and acute hypotension co-horts, we evaluate the goodness-of-fit of generated trajectories via distributional and correlational fidelity, focusing on whether clinically meaningful relationships between variables are preserved over time. We find that cross-feature coherence degrades under irregular sampling, indicating that autoregressive models may not reliably capture dynamic patient context. These findings highlight the need for domain-specific, generation-based evaluation before relying on autoregressive pre-training as a foundation for transfer learning in healthcare.

  • Open Access

    PRIME-CVD: A Parametrically Rendered Informatics Medical Environment for Education in Cardiovascular Risk Modelling

    by Nicholas I-Hsien Kuo, Marzia Hoque Tania, Blanca Gallego, Louisa R Jorm

    In recent years, progress in medical informatics and machine learning has been accelerated by the availability of openly accessible benchmark datasets. However, patient-level electronic medical record (EMR) data are rarely available for teaching or methodological development due to privacy, governance, and re-identification risks. This has limited reproducibility, transparency, and hands-on training in cardiovascular risk modelling. Here we introduce PRIME-CVD, a parametrically rendered informatics medical environment designed explicitly for medical education. PRIME-CVD comprises two openly accessible synthetic data assets representing a cohort of 50,000 adults undergoing primary prevention for cardiovascular disease. The datasets are generated entirely from a user-specified causal directed acyclic graph parameterised using publicly available Australian population statistics and published epidemiologic effect estimates, rather than from patient-level EMR data or trained generative models. Data Asset 1 provides a clean, analysis-ready cohort suitable for exploratory analysis, stratification, and survival modelling, while Data Asset 2 restructures the same cohort into a relational, EMR-style database with realistic structural and lexical heterogeneity. Together, these assets enable instruction in data cleaning, harmonisation, causal reasoning, and policy-relevant risk modelling without exposing sensitive information. Because all individuals and events are generated de novo, PRIME-CVD preserves realistic subgroup imbalance and risk gradients while ensuring negligible disclosure risk. PRIME-CVD is released under a Creative Commons Attribution 4.0 licence to support reproducible research and scalable medical education.

  • Open Access

    Initiators of Semaglutide in General Practice in New South Wales, 2020–2023: A Retrospective Cohort Study

    by Nicholas I-Hsien Kuo, Flynn Hill, Clare Arnott, Kirsty Douglas, Sebastiano Barbieri, Shahana Ferdousi, Ziba Gandomkar, Blanca Gallego-Luxan, Mark Woodward, Louisa Jorm


  • Open Access

    Estimating 5-year absolute risk of cardiovascular disease using routinely collected electronic medical records from Australian general practices

    by Nicholas I-Hsien Kuo, Sebastiano Barbieri, Clare Arnott, Blanca Gallego, Ziba Gandomkar, Shahana Ferdousi, Kirsty Douglas, Mark Woodward, Louisa Jorm

    Background Recent cardiovascular risk equations from the USA and United Kingdom use routinely collected electronic medical records (EMRs), while current equations used in Australia (AusCVDRisk) have not been validated locally. We assessed the feasibility and performance of using routinely collected EMRs from Australian primary care software systems to predict absolute risk of cardiovascular disease (CVD). Methods We used primary care EMR data from the New South Wales Health Lumos programme, covering 680 general practices, linked with hospital and death records. Individuals aged 30–74 years on 1 January 2017 with no prior CVD history and at least one record for an anthropometric measurement or pathology test were included. Sex-specific Cox proportional hazards models were used to estimate 5-year risk of a fatal or non-fatal CVD event. Predictors included demographics, smoking, chronic conditions, clinical variables and medications. Modelling used a 5×2 cross-validation approach. Discrimination, calibration and reclassification performance were assessed. Results Over a mean follow-up of 4.91 years, 33 578 CVD events were recorded in 850 216 patients. Full models with 28 predictors had Harrell’s C of 0.803 (95% CI 0.801 to 0.804) for females and 0.772 (95% CI 0.770 to 0.773) for males. Least absolute shrinkage and selection operator models with 12–15 predictors performed similarly. Models were well calibrated across age, socioeconomic and smoking strata. Geographic (internal–external) validation across 10 Primary Health Networks confirmed consistent discrimination (C-index range 0.747–0.788 for males; 0.772–0.827 for females). Percentile-based net reclassification improvement showed enhanced event detection compared with a model based on AusCVDRisk variables (event-Net Reclassification Improvement up to 0.185). Conclusions CVD risk equations based on routine Australian primary care data performed strongly and generalised well across diverse settings. These models offer potential for automated integration into general practice software to support real-time CVD risk assessment.


  • Open Access

    Limits of Generative Pre-Training in Structured EMR Trajectories with Irregular Sampling

    by Nicholas I-Hsien Kuo, Blanca Gallego, Louisa Jorm

    Foundation models refer to architectures trained on vast datasets using autoregressive pre-training from natural language processing to capture intricate patterns and motifs. They were originally developed to transfer such learned knowledge to downstream predictive tasks. Recently, however, some studies repurpose these learned representations for phenotype discovery without rigorous validation, risking superficially realistic but clinically incoherent embeddings. To test this mismatch, we trained two autoregressive models -- a sequence-to-sequence LSTM and a reduced Transformer -- on longitudinal ART for HIV and Acute Hypotension datasets. Controlled irregularity was added during training via random inter-visit gaps, while test sequences stayed complete. Patient-trajectory synthesis evaluated distributional and correlational fidelity. Both reproduced feature distributions but failed to preserve cross-feature structure -- showing that generative pre-training yields local realism but limited clinical coherence. These results highlight the need for domain-specific evaluation and support trajectory synthesis as a practical probe before fine-tuning or deployment.

  • Predicting cardiovascular events from routine mammograms using machine learning

    by Jennifer Yvonne Barraclough, Jennifer Yvonne Barraclough, Ziba Gandomkar, Ziba Gandomkar, Robert A Fletcher, Robert A Fletcher, Sebastiano Barbieri, Sebastiano Barbieri, Nicholas I-Hsien Kuo, Nicholas I-Hsien Kuo, et al.

    Background Cardiovascular risk is underassessed in women. Many women undergo screening mammography in midlife when the risk of cardiovascular disease rises. Mammographic features such as breast arterial calcification and tissue density are associated with cardiovascular risk. We developed and tested a deep learning algorithm for cardiovascular risk prediction based on routine mammography images. Methods Lifepool is a cohort of women with at least one screening mammogram linked to hospitalisation and death databases. A deep learning model based on DeepSurv architecture was developed to predict major cardiovascular events from mammography images. Model performance was compared against standard risk prediction models using the concordance index, comparative to the Harrells C-statistic. Results There were 49 196 women included, with a median follow-up of 8.8 years (IQR 7.7–10.6), among whom 3392 experienced a first major cardiovascular event. The DeepSurv model using mammography features and participant age had a concordance index of 0.72 (95% CI 0.71 to 0.73), with similar performance to modern models containing age and clinical variables including the New Zealand ‘PREDICT’ tool and the American Heart Association ‘PREVENT’ equations. Conclusions A deep learning algorithm based on only mammographic features and age predicted cardiovascular risk with performance comparable to traditional cardiovascular risk equations. Risk assessments based on mammography may be a novel opportunity for improving cardiovascular risk screening in women.


  • Open Access

    Masked Clinical Modelling: A Framework for Synthetic and Augmented Survival Data Generation

    by Nicholas I-Hsien Kuo, Blanca Gallego, Louisa R. Jorm

    Access to real clinical data is often restricted due to privacy obligations, creating significant barriers for healthcare research. Synthetic datasets provide a promising solution, enabling secure data sharing and model development. However, most existing approaches focus on data realism rather than utility – ensuring that models trained on synthetic data yield clinically meaningful insights comparable to those trained on real data. In this paper, we present Masked Clinical Modelling (MCM), a framework inspired by masked language modelling, designed for both data synthesis and conditional data augmentation. We evaluate this prototype on the WHAS500 dataset using Cox Proportional Hazards models, focusing on the preservation of hazard ratios as key clinical metrics. Our results show that data generated using the MCM framework improves both discrimination and calibration in survival analysis, outperforming existing methods. MCM demonstrates strong potential to support survival data analysis and broader healthcare applications.