Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xiuyi Fan has a total of 204 co-authors. Here are the top co-authors by number of shared publications:

  • Li Rong Wang

    Affiliations College of Computing and Data Science, Nanyang Technological University, Singapore, Singapore, A*STAR Centre for Frontier AI Research, Singapore, Singapore

    9 shared publications
  • Li Rong Wang

    College of Computing and Data Science, Nanyang Technological University, Singapore, Singapore

    8 shared publications
  • Hemalatha Shanmugam

    7 shared publications
  • Shu Ping Lim

    7 shared publications
  • Xinran Xu

    7 shared publications
  • Bingwen Eugene Fan

    Affiliations Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore, Singapore, Department of Haematology, Tan Tock Seng Hospital, Singapore, Singapore, Department of Laboratory Medicine, Khoo Teck Puat Hospital, Singapore, Singapore, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore

    6 shared publications

Latest Publications

  • From cell images to clinical insight: Clinically grounded explainable classification of peripheral blood and bone marrow cells with cross-modal learning

    by Timothy Xiao Jing Yeo, Li Rong Wang, Magdalene Yeok Yu Cheong, Xinran Xu, Hemalatha Shanmugam, Yu Yue Hew, Pik Wan Erica Chiang, Xiao Wei Ang, Rhesha Balwan, Siti Thuraiya Binte Abdul Latiff, et al.

  • Open Access

    Interpretable Forecasting of Irregular Vital Signs with a Missing-Aware Neuro-Fuzzy Inference Framework

    by Zhuoxuan Zhan, Li Rong Wang, Chang Cai, Yunkai Andrew Li, Yih Yng Ng, Xiuyi Fan

    Background and Objectives: Forecasting vital-sign trajectories from routinely collected clinical data can support patient monitoring by anticipating short-term physiological changes before predefined deterioration thresholds are crossed. In practice, however, vital-sign time series are often sparse and irregularly sampled, and missingness patterns may themselves reflect clinical processes rather than random noise. These characteristics challenge conventional forecasting models and complicate clinical interpretation. This study proposes an interpretable forecasting framework for irregular multivariate vital-sign time series.Methods: We developed a Missing-Aware Fuzzy Inference Network (MAFIN) for short-term forecasting of irregular vital signs. MAFIN explicitly distinguishes observed, missing, and padded entries during input encoding, allowing missingness patterns to contribute directly to prediction. The model combines an evolving rule-based global inference module with a lightweight local temporal estimator, and fuses the two streams using a confidence signal derived from rule support. We evaluated MAFIN on four real-world critical-care datasets, MIMIC-III, PhysioNet Challenge 2012, eICU, and INSPIRE, under both full-history and rolling forecasting protocols, and compared it with strong baselines for irregular time-series prediction.Results: Across the four datasets and two evaluation protocols, MAFIN achieved competitive and often superior forecasting performance relative to baseline methods. It obtained the best or second-best RMSE in nearly all settings, with particularly consistent performance on MIMIC-III and INSPIRE, while remaining competitive on PhysioNet 2012 and eICU. In addition to predictive performance, the model generated intrinsic clinician-readable IF–THEN rules for each prediction, supporting case-level interpretability without requiring post-hoc explanation methods.Conclusions: Explicit modeling of missingness and intrinsic rule-based reasoning can provide an effective computational approach for forecasting irregular vital-sign trajectories. The proposed framework combines competitive predictive performance with transparent case-level explanations, supporting retrospective auditability in biomedical monitoring settings and offering a useful basis for future interpretable clinical forecasting research.

  • Open Access

    Building a Global Research Network for Fair, Accountable, Interpretable, and Responsible AI in Emergency Care: Protocol for a FAIR-EC Study

    by Chuan Hong, Jonathan Chong Kai Liew, Jaeyong Yu, Tomás Barry, Audrey L Blewer, Daniel M Buckland, Tianrun Cai, Won Chul Cha, Bibhas Chakraborty, Wei Chen, et al.

    Background: The current landscape of emergency care (EC) is marked by high demand, leading to issues such as emergency department boarding, overcrowding, and subsequent delays that impact the quality and safety of patient care. Integrating data science into EC can enhance decision-making with predictive, preventative, personalized, and participatory approaches. However, gaps in adherence to fairness, accountability, interpretability, and responsibility are evident, particularly due to barriers to data-sharing, which often result in a lack of transparency and robust oversight in these applications. Objective: The FAIR-EC (Fair, Accountable, Interpretable, and Responsible-Emergency Care) collaboration adapts the existing Fair, Accountable, Interpretable, and Responsible principles to address emerging challenges as data science integrates with EC. This initiative aims to transform EC by establishing ethical artificial intelligence standards specifically tailored for this integration. By bridging the gap between EC professionals, data scientists, and other stakeholders, the collaboration promotes international cooperation that leverages advanced data science techniques to enhance EC outcomes across different care settings. Methods: We propose a federated research design to analyze extensive datasets from various global institutions without compromising patient privacy. This approach transforms epidemiological research with advanced data science techniques, emphasizing the harmonization of data for comprehensive analyses across different health care systems. Results: The FAIR-EC initiative has facilitated the identification and harmonization of datasets from diverse geographical regions, enabling the examination of regional variations in EC practices. As of paper submission, participating sites have identified retrospective EC datasets totaling >2 million records (eg, Duke Health >400,000 and Singapore General Hospital >1.7 million records). Initial projects have demonstrated feasibility and operational readiness, including implementation of federated workflows and ongoing development of a federated scoring system, cross-site evaluation, and adaptation of association studies and predictive models across various regions. Cross-site harmonization and pilot analyses are underway (with local ethics approvals in progress), and first multisite results are expected to be submitted in mid-late 2026, with additional project-level publications anticipated in 2027. These efforts highlight the feasibility of leveraging advanced data science techniques to address the complexities of EC while preserving patient privacy without centralizing individual-level data. This project was funded from September 1, 2022, to August 31, 2023. Conclusions: FAIR-EC integrates data science ethically and effectively into EC, addressing challenges such as fragmented data, real-time handoffs, and public health crises. Its federated design harmonizes diverse data streams while preserving privacy, and its emphasis on ethical artificial intelligence aligns with the dynamic nature of EC. Despite challenges in data variability and system complexity, FAIR-EC establishes a strong foundation for innovation in global EC.

  • Open Access

    Aligning deep learning and interpretable models for blood cell classification: A dual-model framework for explainability

    by Magdalene Yeok Yu Cheong, Xinran Xu, Li Rong Wang, Timothy Xiao Jing Yeo, Hemalatha Shanmugam, Shu Ping Lim, Bingwen Eugene Fan, Xiuyi Fan

    Blood smear examination involves classifying cells by morphology under a microscope, a labour-intensive process prone to subjective variation. Recent deep learning models achieve strong performance in blood cell classification but remain as “black boxes”, offering little clinical transparency. We propose a dual-model framework that pairs a deep YOLO (You Only Look Once) classifier with a shallow, interpretable explainer. YOLO performs classification and segmentation, while clinically informed features are extracted from segmented images to train the explainer on YOLO predictions. SHapley Additive exPlanations (SHAP) quantify feature importance, with discrepancies flagged as “other reasons” to enhance transparency. Evaluated on proprietary and public datasets (PBC, Raabin), YOLO achieved AUCs of 0.997, 0.994, and 1.000 respectively, with the explainer demonstrating strong alignment: Score-CAM activation maps agreed with SHAP-identified features. In a user study with 9 trained and qualified haematologists and laboratory medical technologists, predictions achieved 96.9% concurrence. This work shows that the coupling of deep learning with interpretable models improves the confidence of predictions while providing clinical meaningful insights.

  • Missing-Aware Evolving Fuzzy Learning for MASLD Risk Prediction from Irregular Time Series

    by Zhuoxuan Zhan, Hong Sheng Cheng, Ian Yang Liew, Hwee Pin Phua, Angela Li Ping Chow, Eng Sing Lee, Kuo Chao Yew, Xiuyi Fan

  • Top Cited Papers
    Open Access

    A Competency Framework for Medical AI Education: Mixed Methods Study

    by Chang Cai, Gaoxia Zhu, Shang-Ming Zhou, Olivia Ng, Jamie Duell, Weng Kin Ho, Daisy Minghui Chen, Bernett Lee, Fang Li, Siyuan Liu, et al.

    Background: Although artificial intelligence (AI) is increasingly adopted in health care, clinicians face barriers, including insufficient understanding, limited trust, and challenges in interpreting AI outputs. Existing frameworks, such as the United Nations Educational, Scientific and Cultural Organization (UNESCO) AI competency framework, lack clinical specificity. Additionally, there remains limited evidence on framework-based training programs for medical professionals. Objective: This study aimed to (1) develop a medical AI competency framework and (2) apply the framework to design an AI training program and pilot 1 module. Methods: We conducted a mixed methods study comprising 2 phases. In Study 1, we developed a medical AI competency framework by integrating the UNESCO AI framework with the Miller pyramid model. The framework was refined through expert input from 24 stakeholders (6 hospital administrators, 8 medical professionals, and 10 university instructors). Expert responses were analyzed using deductive content analysis based on predefined codebooks. In study 2, we designed an AI training program based on the framework, and evaluated it using a 2-round Delphi process. Nine educators with expertise in instructional design, medical education, and AI participated in the Delphi study. Consensus in round 1 was defined as an IQR≤1, agreement score proportion >75%, and full-score frequency >49.23%; in round 2, consensus was defined as agreement score proportion >80%. A pilot workshop with 28 participants and 4 instructors assessed the feasibility of 1 module using self-reported measures of satisfaction, engagement, and confidence. Results: A 6D 4-level medical AI competency framework was developed. Among the 24 experts, 19 (79.17%) mentioned competencies related to AI foundations, and 23 (95.83%) mentioned competencies related to application skills. The framework was translated into a 5-module training program, covering patient-centered and ethical AI, privacy and security, medical AI applications, bias and health equity, and generative AI in health care. Each module included 5 elements: content, learning goals, teaching activities, learning resources, and assessment. The Delphi process achieved complete consensus across all 25 elements of the training program. The pilot workshop indicated high participant satisfaction (mean 4.00, SD 0.52), good engagement (mean 3.80, SD 0.71‐mean 4.05, SD 0.51), and moderate self-reported confidence (mean 3.63, SD 0.53). These findings suggest that the module was feasible, although outcomes should be interpreted cautiously given the self-reported and short-term nature of the evaluation. Conclusions: The framework provides a structured reference for AI training program design in medical education. The workshop findings provide preliminary support for the feasibility of the program’s technical module, while also highlighting the need for broader and longer-term evaluations. Future work should expand the framework and training program to new regions and delivery formats (eg, semester-long courses and continuing medical education) and evaluate their long-term impact.


  • Open Access

    Development and Clinical Validation of an Artificial Intelligence-Based Automated Visual Acuity Testing System

    by Kelvin Zhenghao Li, Hnin Hnin Oo, Kenneth Chee Wei Liang, Najah Ismail, Jasmine Ling Ling Chua, Jackson Jie Sheng Chng, Yang Wu, Daryl Wei Ren Wong, Sumaya Rani Khan, Boon Peng Yap, et al.

    Background: To develop and validate an automated visual acuity (VA) testing system integrating artificial intelligence (AI)–driven speech and image recognition technologies, enabling self-administered, clinic-based VA assessment; Methods: The system incorporated a fine-tuned Whisper speech-recognition model with Silero voice activity detection and pose estimation through facial landmark and ArUco marker detection. A state-driven interface guided users through sequential testing with and without a pinhole. Speech recognition was enhanced using a local Singaporean English dataset. Laboratory validation assessed speech and pose recognition performance, while clinical validation compared automated and manual VA testing at a tertiary eye clinic; Results: The fine-tuned model reduced word error rates from 17.83% to 9.81% for letters and 2.76% to 1.97% for numbers. Pose detection accurately identified valid occluder states. Among 72 participants (144 eyes), automated unaided VA showed good agreement with manual VA (ICC = 0.77, 95% CI 0.62–0.85), while pinhole VA demonstrated moderate agreement (ICC = 0.63, 95% CI 0.25–0.83). Automated testing took longer (132.1 ± 47.5 s vs. 97.1 ± 47.8 s; p < 0.001), but user experience remained positive (mean Likert scale score 4.3 ± 0.8); Conclusions: The AI-based automated VA system delivered accurate, reliable, and user-friendly performance, supporting its feasibility for clinical implementation.

  • Open Access

    Robust Uncertainty Estimation under Distribution Shift via Difference Reconstruction

    by Xinran Xu, Li Rong Wang, Xiuyi Fan

    Estimating uncertainty in deep learning models is critical for reliable decision-making in high-stakes applications such as medical imaging. Prior research has established that the difference between an input sample and its reconstructed version produced by an auxiliary model can serve as a useful proxy for uncertainty. However, directly comparing reconstructions with the original input is degraded by information loss and sensitivity to superficial details, which limits its effectiveness. In this work, we propose Difference Reconstruction Uncertainty Estimation (DRUE), a method that mitigates this limitation by reconstructing inputs from two intermediate layers and measuring the discrepancy between their outputs as the uncertainty score. To evaluate uncertainty estimation in practice, we follow the widely used out-of-distribution (OOD) detection paradigm, where in-distribution (ID) training data are compared against datasets with increasing domain shift. Using glaucoma detection as the ID task, we demonstrate that DRUE consistently achieves superior AUC and AUPR across multiple OOD datasets, highlighting its robustness and reliability under distribution shift. This work provides a principled and effective framework for enhancing model reliability in uncertain environments.

  • Open Access

    Explainable Melanoma Diagnosis with Contrastive Learning and LLM-based Report Generation

    by Junwen Zheng, Xinran Xu, Li Rong Wang, Chang Cai, Lucinda Siyun Tan, Dingyuan Wang, Hong Liang Tey, Xiuyi Fan

    Deep learning has demonstrated expert-level performance in melanoma classification, positioning it as a powerful tool in clinical dermatology. However, model opacity and the lack of interpretability remain critical barriers to clinical adoption, as clinicians often struggle to trust the decision-making processes of black-box models. To address this gap, we present a Cross-modal Explainable Framework for Melanoma (CEFM) that leverages contrastive learning as the core mechanism for achieving interpretability. Specifically, CEFM maps clinical criteria for melanoma diagnosis-namely Asymmetry, Border, and Color (ABC)-into the Vision Transformer embedding space using dual projection heads, thereby aligning clinical semantics with visual features. The aligned representations are subsequently translated into structured textual explanations via natural language generation, creating a transparent link between raw image data and clinical interpretation. Experiments on public datasets demonstrate 92.79% accuracy and an AUC of 0.961, along with significant improvements across multiple interpretability metrics. Qualitative analyses further show that the spatial arrangement of the learned embeddings aligns with clinicians' application of the ABC rule, effectively bridging the gap between high-performance classification and clinical trust.

  • Open Access

    Evaluating a Multi-Modal Large Language Model for Ophthalmology Triage

    by Caius Goh, Jabez Ng, Wei Yung Au, Clarence See, Alva Lim, Jun Wen Zheng, Xiuyi Fan, Kelvin Li

    Background/Purpose: Ophthalmic triage is challenging for non-specialists due to limited training and rising global eye disease burden. This study evaluates a multimodal framework integrating clinical text and ophthalmic imaging with large language models (LLMs). Textual consistency filtering and chain-of-thought (CoT) reasoning were incorporated to improve diagnostic accuracy. Methods: A dataset of 56 ophthalmology cases from a Singapore restructured hospital was pre-processed with acronym expansion, sentence reconstruction, and textual consistency filtering. To address dataset size limitations, 100 synthetic cases were generated via one-shot GPT-4 prompting, validated by semantic checks and ophthalmologist review. Three diagnostic approaches were tested: Text-Only, Image-Assisted, and Image with CoT. Diagnostic performance was quantified using a novel SNOMED-CT-based dissimilarity score, defined as the shortest path distance between predicted and reference diagnoses in the ontology, which was used to quantify semantic alignment. Results: The synthetic dataset included anterior segment (n = 40), posterior segment (n = 35), and extraocular (n = 25) cases. The text-only approach yielded a mean dissimilarity of 6.353 (95% CI: 4.668, 8.038). Incorporation of image assistance reduced this to 5.234 (95% CI: 3.930, 6.540), while CoT prompting provided further gains when imaging cues were ambiguous. Conclusions: The multimodal pipeline showed potential in improving diagnostic alignment in ophthalmology triage. Image inputs enhanced accuracy, and CoT reasoning reduced errors from ambiguous features, supporting its feasibility as a pilot framework for ophthalmology triage.