VDVien Ngoc Dang

Vien Ngoc Dang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Vien Ngoc Dang has a total of 27 co-authors. Here are the top co-authors by number of shared publications:

  • Karim Lekadir

    Artificial Intelligence in Medicine Lab, Facultat de Matemátiques i Informática Universitat de Barcelona Spain

    4 shared publications
  • Karim Lekadir

    Artificial Intelligence in Medicine Lab (BCN-AIM), Department of Mathematics and Computer Science, University of Barcelona, Barcelona, Spain

    4 shared publications
  • Charlotte Cecil

    Affiliations Department of Child and Adolescent Psychiatry/Psychology, Erasmus MC, University Medical Center Rotterdam, Rotterdam, The Netherlands, Department of Epidemiology, Erasmus MC, Rotterdam, University Medical Center Rotterdam, Rotterdam, The Netherlands, Molecular Epidemiology, Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, The Netherlands

    3 shared publications
  • Jerónimo Hernández-González

    Affiliation Departament d’Informàtica, Matemàtica Aplicada i Estadística, Universitat de Girona, Girona, Spain

    3 shared publications
  • Adrià Casamitjana

    Computer Vision and Robotics Institute University of Girona Girona Spain

    2 shared publications
  • Anna Cascarano

    Departament de Matemàtiques i Informàtica, Facultat de Matemàtiques i Informàtica, Universitat de Barcelona, Barcelona, Spain

    2 shared publications

Latest Publications

  • Open Access

    Artificial Intelligence for Medical Image Analysis in Low-Resource Settings: A Systematic Review

    by Tanya Akumu, Victor M. Campello, Lidia Garrucho, Vien Ngoc Dang, Thomas Keller, Xue Zhao, Emmah Ng'Ang'A, María J. Ledesma-Carbayo, Marius G. Linguraru, Karim Lekadir

    Artificial intelligence (AI) for medical image analysis is increasingly being developed for low-resource settings, where specialist expertise, imaging equipment, labelled data, and computing resources may be limited. We systematically reviewed 189 studies published between January 2021 and January 2026 to examine how these constraints relate to imaging modalities, clinical applications, and AI methods in low-resource settings. Each study was also assessed against 15 criteria adapted from the FUTURE-AI principles for trustworthy AI. Shortages of clinical expertise were the most common constraint, and portable or lower-cost imaging was widely used. Transfer learning was the most frequent technical strategy (59%), followed by data augmentation (54%) and efficient model design (36%). Medical-image pretraining and self-supervised learning were each used in 4% of studies, and model compression in 3%. Applications were concentrated in retinal disease, pulmonary disease, antenatal care, parasitic diseases, and cervical screening. Beyond these technical and clinical patterns, the trustworthy AI assessment showed a clear gap between documenting how systems were developed and demonstrating how they performed under relevant local conditions. Subgroup performance was evaluated in 24% of studies, intended users or realistic working conditions in 25%, and model explanations with intended users in 4%. Consistent with this gap, among 15 studies that reported the same model and metric during development and in the local setting, 12 performed worse locally. Future research should prioritise representative local evaluation, intended-user assessment, demonstrated resource suitability, and reproducible research practices.

  • Open Access

    Bias and Fairness in Radiomics: A Comparative Analysis of Machine Learning Models on Four Oncology Datasets

    by Esmeralda Ruiz Pujadas, Jerónimo Hernández‐González, Vien Ngoc Dang, Marina Camacho, Karim Lekadir, Martijn P. A. Starmans

    Radiomics, the practice of mining quantitative features from medical imaging, has gained increasing popularity in diagnosing and treating cancer. However, clinical deployment and adoption of radiomics tools in real‐world practice remain limited. To increase trust in radiomics models, one of the main remaining challenges is ensuring fairness by addressing algorithmic bias and promoting equitable outcomes. This study aims to identify key fairness challenges in radiomics and demonstrate how state‐of‐the‐art mitigation techniques can be applied to address them. Moreover, we aim to provide new guidelines for both the selection of mitigation techniques and procedural recommendations for future directions in fairness. This study evaluates the fairness of radiomics models using six machine learning algorithms applied to four publicly available oncology datasets: Liver, Gastrointestinal, Desmoid‐type fibromatosis, and Liposarcoma. Fairness was assessed using Equal Opportunity Difference (EOD) with respect to two protected attributes: age and sex. We employed three bias mitigation techniques—Disparate Impact Remover (DIR), Reweighting, and Reject Option Classification (ROC)—to examine their effect on fairness metrics and model performance. DIR, Reweighting, and ROC each showed dataset‐ and attribute‐specific improvements in fairness. For example, DIR notably reduced sex‐related bias in the Liposarcoma dataset, while ROC offered the best fairness–performance trade‐off in the Liver dataset. Reweighting was effective in mitigating age‐related bias in Liposarcoma. In the Desmoid dataset, applying ROC to Extra Trees marginally improved fairness; however, age‐related bias persisted. Overall, the applied techniques reduced disparities in true positive rates between groups, enhancing fairness. Our findings show that while radiomics models can be accurate, they often exhibit bias with respect to demographic groups. Carefully addressing this trade‐off is crucial for developing radiomics models that are accurate but also trustworthy for clinical deployment.


  • Open Access

    Characterizing the role of early life factors in machine learning-based multimorbidity risk prediction

    by Vien Ngoc Dang, Charlotte Cecil, Carmine M. Pariante, Jerónimo Hernández-González, Karim Lekadir

    Recent evidence suggests that psycho-cardio-metabolic (PCM) multimorbidity finds its origins in exposure to early-life factors (ELFs), making the exploration of this association crucial for understanding and effective management of these complex health issues. Moreover, risk prediction models for cardiovascular diseases (CVD) and diabetes, as recommended by current clinical guidelines, typically demonstrate sub-optimal performance in clinically relevant sub-populations where these ELFs may play a substantial role. Our methodological approach investigates the contribution of ELFs to machine-learning-based risk prediction models for comorbid populations, incorporating a wide set of early-life and proximal variables, with a special focus on prenatal and postnatal ELFs. To address the complexity of integrating diverse early-life and proximal factors, we leverage models capable of handling high-dimensional, heterogeneous data sources to enhance prediction accuracy in complex clinical populations. The long-term predictive ability of ELFs, along with their influence on model decisions, is assessed with the learned models, and global and local model-agnostic interpretative techniques allow us to elucidate some interactions leading to multimorbidity. The data for this study is derived from the UK Biobank, showcasing both the strengths and limitations inherent in utilizing a single, large-scale database for such research. Our results show enhanced predictive performance for CVD (AUC-ROC: +7.9%, Acc: +14.7%, Cohen’s d: 1.5) among individuals with concurrent mental health issues (depression or anxiety) and diabetes. Similarly, we demonstrate improved diabetes risk prediction (AUC-ROC: +12.3%, Acc: +13.5%, Cohen’s d: 2.5) in those with concurrent mental health conditions and CVD. The inspection of these models, which integrate a large set of ELFs and other predictors (including the 7-core Framingham and UKDiabetes variables), provides key information that could lead to a more profound understanding of psycho-cardio-metabolic multimorbidity. Our findings highlight the utility of incorporating life-course factors into risk models. Integrating a diverse range of physiological, psychological, and ELFs becomes particularly pertinent in the context of multimorbidity.


  • Open Access

    Empirical Comparison of Post-processing Debiasing Methods for Machine Learning Classifiers in Healthcare

    by Vien Ngoc Dang, Víctor M. Campello, Jerónimo Hernández-González, Karim Lekadir

    Machine learning classifiers in healthcare tend to reproduce or exacerbate existing health disparities due to inherent biases in training data. This relevant issue has brought the attention of researchers in both healthcare and other domains, proposing techniques that deal with it in different stages of the machine learning process. Post-processing methods adjust model predictions to ensure fairness without interfering in the learning process nor requiring access to the original training data, preserving privacy and enabling the application to any trained model. This study rigorously compares state-of-the-art debiasing methods within the family of post-processing techniques across a wide range of synthetic and real-world (healthcare) datasets, by means of different performance and fairness metrics. Our experiments reveal the strengths and weaknesses of each method, examining the trade-offs between group fairness and predictive performance, as well as among different notions of group fairness. Additionally, we analyze the impact on untreated attributes to ensure overall bias mitigation. Our comprehensive evaluation provides insights into how these debiasing methods can be optimally implemented in healthcare settings to balance accuracy and fairness.


  • Mitigating Overdiagnosis Bias in CNN-Based Alzheimer’s Disease Diagnosis for the Elderly

    by Vien Ngoc Dang, Adrià Casamitjana, Jerónimo Hernández-González, Karim Lekadir, For The Alzheimer’s Disease Neuroimaging Initiative


  • Open Access

    Fairness and bias correction in machine learning for depression prediction across four study populations

    by Vien Ngoc Dang, Anna Cascarano, Rosa H. Mulder, Charlotte Cecil, Maria A. Zuluaga, Jerónimo Hernández-González, Karim Lekadir

    A significant level of stigma and inequality exists in mental healthcare, especially in under-served populations. Inequalities are reflected in the data collected for scientific purposes. When not properly accounted for, machine learning (ML) models learned from data can reinforce these structural inequalities or biases. Here, we present a systematic study of bias in ML models designed to predict depression in four different case studies covering different countries and populations. We find that standard ML approaches regularly present biased behaviors. We also show that mitigation techniques, both standard and our own post-hoc method, can be effective in reducing the level of unfair bias. There is no one best ML model for depression prediction that provides equality of outcomes. This emphasizes the importance of analyzing fairness during model selection and transparent reporting about the impact of debiasing interventions. Finally, we also identify positive habits and open challenges that practitioners could follow to enhance fairness in their models.


  • Open Access

    Fairness and bias correction in machine learning for depression prediction: results from four study populations

    by Vien Ngoc Dang, Anna Cascarano, Rosa H. Mulder, Charlotte Cecil, Maria A. Zuluaga, Jerónimo Hernández-González, Karim Lekadir

    A significant level of stigma and inequality exists in mental healthcare, especially in under-served populations. Inequalities are reflected in the data collected for scientific purposes. When not properly accounted for, machine learning (ML) models leart from data can reinforce these structural inequalities or biases. Here, we present a systematic study of bias in ML models designed to predict depression in four different case studies covering different countries and populations. We find that standard ML approaches show regularly biased behaviors. We also show that mitigation techniques, both standard and our own post-hoc method, can be effective in reducing the level of unfair bias. No single best ML model for depression prediction provides equality of outcomes. This emphasizes the importance of analyzing fairness during model selection and transparent reporting about the impact of debiasing interventions. Finally, we provide practical recommendations to develop bias-aware ML models for depression risk prediction.

  • Auditing Unfair Biases in CNN-Based Diagnosis of Alzheimer’s Disease

    by Vien Ngoc Dang, Adrià Casamitjana, Martijn P. A. Starmans, Carlos Martín-Isla, Jerónimo Hernández-González, Karim Lekadir, For The Alzheimer’S Disease Neuroimaging Initiative