Yi Lian
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Yi Lian has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Qi Long
Department of Biostatistics, Epidemiology and Informatics, University of Pennsylvania, Philadelphia, USA
4 shared publications Xiaoqian Jiang
University of Texas Health Science Center, Houston, TX, USA
3 shared publications- Robert William Platt
Department of Epidemiology, Biostatistics, and Occupational Health McGill University Montreal Quebec Canada
2 shared publications - Almut Gertrud Winterstein
Department of Pharmaceutical Outcomes and Policy, University of Florida, Gainesville, USA
1 shared publication - Andrew J. Foy
Division of Cardiology, Penn State Milton S. Hershey Medical Center, Hershey, Pennsylvania
1 shared publication Archer Yi Yang
Department of Mathematics and Statistics, McGill University, Montréal, Québec, Canada
1 shared publication
Latest Publications
- Open Access
- 8 October 2026
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 26 (1) , 396
D3MI: an efficient and powerful federated imputation method for bias reduction in the analysis of distributed incomplete data by accounting for within-site correlation and between-site heterogeneity
Background: Electronic health records (EHRs) collected from diverse healthcare institutions offer a rich and representative data source for clinical research. Federated learning enables analysis of these distributed data without sharing sensitive patient-level information, preserving privacy. However, missing data remain a major challenge and can introduce substantial bias if not properly addressed. Very few distributed imputation methods currently exist, and they fail to account for two critical aspects of EHR data: correlation within sites and variability across sites. We aim to fill this important methodological gap. Methods: We propose Distributed Mixed Model-based Multiple Imputation (D3MI), a novel federated imputation method designed to reduce bias in distributed EHRs. D3MI integrates the strengths from federated learning techniques, statistical learning methods for correlated data, and multilevel imputation algorithms to explicitly account for both within-site correlation and between-site heterogeneity using site-specific random effects. It operates under the Missing at Random (MAR) assumption. It preserves privacy by avoiding sharing raw data and features communication and computational efficiency. Results: Through extensive simulation studies, we demonstrate that D3MI outperforms SOTA distributed imputation methods in both accuracy and consistency. We further demonstrate the use of D3MI in a real-world EHR case study involving incomplete and clustered data from participating hospitals in the Georgia Coverdell Acute Stroke Registry. Conclusion: By explicitly modeling the complex structure of distributed EHR data, D3MI addresses key limitations of existing approaches. It provides a powerful and efficient solution for handling missing data in distributed and privacy-sensitive settings and enhances the rigor and reproducibility of collaborative clinical research.
- 8 October 2026
- Published by Springer Nature in BMC Medical Informatics and Decision Making
- Vol. 26 (1) , 396
Open AccessD3MI: an efficient and powerful federated imputation method for bias reduction in the analysis of distributed incomplete data by accounting for within-site correlation and between-site heterogeneity
Background: Electronic health records (EHRs) collected from diverse healthcare institutions offer a rich and representative data source for clinical research. Federated learning enables analysis of these distributed data without sharing sensitive patient-level information, preserving privacy. However, missing data remain a major challenge and can introduce substantial bias if not properly addressed. Very few distributed imputation methods currently exist, and they fail to account for two critical aspects of EHR data: correlation within sites and variability across sites. We aim to fill this important methodological gap. Methods: We propose Distributed Mixed Model-based Multiple Imputation (D3MI), a novel federated imputation method designed to reduce bias in distributed EHRs. D3MI integrates the strengths from federated learning techniques, statistical learning methods for correlated data, and multilevel imputation algorithms to explicitly account for both within-site correlation and between-site heterogeneity using site-specific random effects. It operates under the Missing at Random (MAR) assumption. It preserves privacy by avoiding sharing raw data and features communication and computational efficiency. Results: Through extensive simulation studies, we demonstrate that D3MI outperforms SOTA distributed imputation methods in both accuracy and consistency. We further demonstrate the use of D3MI in a real-world EHR case study involving incomplete and clustered data from participating hospitals in the Georgia Coverdell Acute Stroke Registry. Conclusion: By explicitly modeling the complex structure of distributed EHR data, D3MI addresses key limitations of existing approaches. It provides a powerful and efficient solution for handling missing data in distributed and privacy-sensitive settings and enhances the rigor and reproducibility of collaborative clinical research.
- 22 May 2025
- Vol. 2024, 703-712
Federated Multiple Imputation for Variables that Are Missing Not At Random in Distributed Electronic Health Records.
Large electronic health records (EHR) have been widely implemented and are available for research activities. The magnitude of such databases often requires storage and computing infrastructure that are distributed at different sites. Restrictions on data-sharing due to privacy concerns have been another driving force behind the development of a large class of distributed and/or federated machine learning methods. While missing data problem is also present in distributed EHRs, albeit potentially more complex, distributed multiple imputation (MI) methods have not received as much attention. An important advantage of distributed MI, as well as distributed analysis, is that it allows researchers to borrow information across data sites, mitigating potential fairness issues for minority groups that do not have enough volume at certain sites. In this paper, we propose a communication-efficient and privacy-preserving distributed MI algorithms for variables that are missing not at random.
- 22 May 2025
- Vol. 2024, 703-712
Federated Multiple Imputation for Variables that Are Missing Not At Random in Distributed Electronic Health Records.
Large electronic health records (EHR) have been widely implemented and are available for research activities. The magnitude of such databases often requires storage and computing infrastructure that are distributed at different sites. Restrictions on data-sharing due to privacy concerns have been another driving force behind the development of a large class of distributed and/or federated machine learning methods. While missing data problem is also present in distributed EHRs, albeit potentially more complex, distributed multiple imputation (MI) methods have not received as much attention. An important advantage of distributed MI, as well as distributed analysis, is that it allows researchers to borrow information across data sites, mitigating potential fairness issues for minority groups that do not have enough volume at certain sites. In this paper, we propose a communication-efficient and privacy-preserving distributed MI algorithms for variables that are missing not at random.
- Open Access
D3MI: an efficient and powerful federated imputation method for bias reduction in the analysis of distributed incomplete data by accounting for within-site correlation and between-site heterogeneity
Objective: Electronic health records (EHRs) collected from diverse healthcare institutions offer a rich and representative data source for clinical research. Federated learning enables analysis of these distributed data without sharing sensitive patient-level information, preserving privacy. However, missing data remain a major challenge and can introduce substantial bias if not properly addressed. Very few distributed imputation methods currently exist, and they fail to account for two critical aspects of EHR data: correlation within sites and variability across sites. We aim to fill this important methodological gap. Methods: We propose Distributed Mixed Model-based Multiple Imputation (D3MI), a novel federated imputation method designed to reduce bias in distributed EHRs. D3MI integrates the strengths from federated learning techniques, statistical learning methods for correlated data, and multilevel imputation algorithms to explicitly account for both and within-site correlation and between-site heterogeneity using site-specific random effects. It preserves privacy by avoiding sharing raw data and features communication and computational efficiency. Results: Through extensive simulation studies, we demonstrate that D3MI outperforms SOTA distributed imputation methods in both accuracy and consistency. We further demonstrate the use of D3MI in a real-world EHR case study involving incomplete and clustered data from participating hospitals in the Georgia Coverdell Acute Stroke Registry. Conclusion: By explicitly modeling the complex structure of distributed EHR data, D3MI addresses key limitations of existing approaches. It provides a powerful and efficient solution for handling missing data in distributed and privacy-sensitive settings and enhances the rigor and reproducibility of collaborative clinical research.
Open AccessD3MI: an efficient and powerful federated imputation method for bias reduction in the analysis of distributed incomplete data by accounting for within-site correlation and between-site heterogeneity
Objective: Electronic health records (EHRs) collected from diverse healthcare institutions offer a rich and representative data source for clinical research. Federated learning enables analysis of these distributed data without sharing sensitive patient-level information, preserving privacy. However, missing data remain a major challenge and can introduce substantial bias if not properly addressed. Very few distributed imputation methods currently exist, and they fail to account for two critical aspects of EHR data: correlation within sites and variability across sites. We aim to fill this important methodological gap. Methods: We propose Distributed Mixed Model-based Multiple Imputation (D3MI), a novel federated imputation method designed to reduce bias in distributed EHRs. D3MI integrates the strengths from federated learning techniques, statistical learning methods for correlated data, and multilevel imputation algorithms to explicitly account for both and within-site correlation and between-site heterogeneity using site-specific random effects. It preserves privacy by avoiding sharing raw data and features communication and computational efficiency. Results: Through extensive simulation studies, we demonstrate that D3MI outperforms SOTA distributed imputation methods in both accuracy and consistency. We further demonstrate the use of D3MI in a real-world EHR case study involving incomplete and clustered data from participating hospitals in the Georgia Coverdell Acute Stroke Registry. Conclusion: By explicitly modeling the complex structure of distributed EHR data, D3MI addresses key limitations of existing approaches. It provides a powerful and efficient solution for handling missing data in distributed and privacy-sensitive settings and enhances the rigor and reproducibility of collaborative clinical research.
- 1 April 2025
- Published by Wolters Kluwer Health in Circulation: Cardiovascular Quality and Outcomes
- Vol. 18 (4) , e011188
Association Between Use of WATCHMAN Device and 1-Year Mortality Using High-Dimensional Propensity Scores to Reduce Confounding
BACKGROUND: Previous observational studies showed left atrial appendage occlusions with the WATCHMAN device reduced 1-year mortality, which conflicted with evidence generated from randomized controlled trials. We proposed to use the high-dimensional propensity score (hdPS) to assist in nonactive comparator selection (prevalent user of medication) and compared 1-year mortality between patients with atrial fibrillation who received the WATCHMAN device (percutaneous left atrial appendage occlusion device [pLAAO]) and direct oral anticoagulants in 2 matched cohorts based on (1) traditional propensity score (PS) and (2) integrating traditional PS with information learned from hdPS. METHODS: Patients entered the cohort once diagnosed with atrial fibrillation in the 15% of Medicare fee-for-service claims database from 2011 to 2018. Patients could enter the study cohort upon receiving WATCHMAN or at an outpatient visit with an atrial fibrillation diagnosis, respectively. We used PS matching with a 1:3 ratio for patients in pLAAO and direct oral anticoagulant groups. In cohort 2, we implemented a multistep approach with information learned from hdPS. The Cox proportional hazards model was used to estimate hazard ratios of outcomes with 95% CIs. RESULTS: In cohort 1, we identified 1159 and 3477 patients in the pLAAO and direct oral anticoagulant groups with a mean age of 78.1 versus 77.5 years, 44.9% versus 40.8% of women, and a 1-year mortality rate of 8.02 versus 8.97/100 person-years (hazard ratio, 0.87 [95% CI, 0.69–1.09]). With the support of hdPS, in cohort 2, we excluded patients with malignant cancer and added frailty score in the PS model. We identified 953 and 2859 patients in the pLAAO and direct oral anticoagulant groups with a mean age of 78.1 versus 77.9 years, 47.2% versus 46.1% of women, and a 1-year mortality rate of 7.45 and 7.69/100 person-years (hazard ratio, 0.95 [95% CI, 0.73–1.24]). CONCLUSIONS: No association was found between pLAAO and 1-year mortality, which is consistent with existing evidence from randomized controlled trials. The hdPS approach provides an opportunity to improve nonactive comparator selection in traditional PS analysis.
- 1 April 2025
- Published by Wolters Kluwer Health in Circulation: Cardiovascular Quality and Outcomes
- Vol. 18 (4) , e011188
Association Between Use of WATCHMAN Device and 1-Year Mortality Using High-Dimensional Propensity Scores to Reduce Confounding
BACKGROUND: Previous observational studies showed left atrial appendage occlusions with the WATCHMAN device reduced 1-year mortality, which conflicted with evidence generated from randomized controlled trials. We proposed to use the high-dimensional propensity score (hdPS) to assist in nonactive comparator selection (prevalent user of medication) and compared 1-year mortality between patients with atrial fibrillation who received the WATCHMAN device (percutaneous left atrial appendage occlusion device [pLAAO]) and direct oral anticoagulants in 2 matched cohorts based on (1) traditional propensity score (PS) and (2) integrating traditional PS with information learned from hdPS. METHODS: Patients entered the cohort once diagnosed with atrial fibrillation in the 15% of Medicare fee-for-service claims database from 2011 to 2018. Patients could enter the study cohort upon receiving WATCHMAN or at an outpatient visit with an atrial fibrillation diagnosis, respectively. We used PS matching with a 1:3 ratio for patients in pLAAO and direct oral anticoagulant groups. In cohort 2, we implemented a multistep approach with information learned from hdPS. The Cox proportional hazards model was used to estimate hazard ratios of outcomes with 95% CIs. RESULTS: In cohort 1, we identified 1159 and 3477 patients in the pLAAO and direct oral anticoagulant groups with a mean age of 78.1 versus 77.5 years, 44.9% versus 40.8% of women, and a 1-year mortality rate of 8.02 versus 8.97/100 person-years (hazard ratio, 0.87 [95% CI, 0.69–1.09]). With the support of hdPS, in cohort 2, we excluded patients with malignant cancer and added frailty score in the PS model. We identified 953 and 2859 patients in the pLAAO and direct oral anticoagulant groups with a mean age of 78.1 versus 77.9 years, 47.2% versus 46.1% of women, and a 1-year mortality rate of 7.45 and 7.69/100 person-years (hazard ratio, 0.95 [95% CI, 0.73–1.24]). CONCLUSIONS: No association was found between pLAAO and 1-year mortality, which is consistent with existing evidence from randomized controlled trials. The hdPS approach provides an opportunity to improve nonactive comparator selection in traditional PS analysis.
- Open Access
Federated Multiple Imputation for Variables that Are Missing Not At Random in Distributed Electronic Health Records
Large electronic health records (EHR) have been widely implemented and are available for research activities. The magnitude of such databases often requires storage and computing infrastructure that are distributed at different sites. Restrictions on data-sharing due to privacy concerns have been another driving force behind the development of a large class of distributed and/or federated machine learning methods. While missing data problem is also present in distributed EHRs, albeit potentially more complex, distributed multiple imputation (MI) methods have not received as much attention. An important advantage of distributed MI, as well as distributed analysis, is that it allows researchers to borrow information across data sites, mitigating potential fairness issues for minority groups that do not have enough volume at certain sites. In this paper, we propose a communication-efficient and privacy-preserving distributed MI algorithms for variables that are missing not at random.
Open AccessFederated Multiple Imputation for Variables that Are Missing Not At Random in Distributed Electronic Health Records
Large electronic health records (EHR) have been widely implemented and are available for research activities. The magnitude of such databases often requires storage and computing infrastructure that are distributed at different sites. Restrictions on data-sharing due to privacy concerns have been another driving force behind the development of a large class of distributed and/or federated machine learning methods. While missing data problem is also present in distributed EHRs, albeit potentially more complex, distributed multiple imputation (MI) methods have not received as much attention. An important advantage of distributed MI, as well as distributed analysis, is that it allows researchers to borrow information across data sites, mitigating potential fairness issues for minority groups that do not have enough volume at certain sites. In this paper, we propose a communication-efficient and privacy-preserving distributed MI algorithms for variables that are missing not at random.
- 3 January 2023
- Published by Taylor & Francis in Technometrics
- Vol. 65 (2) , 281-295
A Tweedie Compound Poisson Model in Reproducing Kernel Hilbert Space
Tweedie models can be used to analyze non-negative continuous data with a probability mass at zero. There have been wide applications in natural science, healthcare research, actuarial science and other fields. The performance of existing Tweedie models can be limited on today’s complex data problems with challenging characteristics such as nonlinear effects, high-order interactions, high-dimensionality and sparsity. In this paper, we propose a kernel Tweedie model Ktweedie and its sparse variant SKtweedie that can simultaneously address the above challenges. Specifically, nonlinear effects and high-order interactions can be flexibly represented through a wide range of kernel functions, which is fully learned from the data; In addition, while the Ktweedie can handle high-dimensional data, the SKtweedie with integrated variable selection can further improve the interpretability. We perform extensive simulation studies to justify the prediction and variable selection accuracy of our method, and demonstrate the applications in ratemaking and loss-reserving in general insurance. Overall, the Ktweedie and SKtweedie outperform existing Tweedie models when there exist nonlinear effects and high-order interactions, particularly when the dimensionality is high relative to the sample size. The model is implemented in an efficient and user-friendly R package ktweedie (https://cran.r-project.org/package=ktweedie). Supplementary material is available online. Keywords: Insurance claims data; Loss-reserving; Ratemaking; Sparse kernel methods; Zero inflated data.
- 3 January 2023
- Published by Taylor & Francis in Technometrics
- Vol. 65 (2) , 281-295
A Tweedie Compound Poisson Model in Reproducing Kernel Hilbert Space
Tweedie models can be used to analyze non-negative continuous data with a probability mass at zero. There have been wide applications in natural science, healthcare research, actuarial science and other fields. The performance of existing Tweedie models can be limited on today’s complex data problems with challenging characteristics such as nonlinear effects, high-order interactions, high-dimensionality and sparsity. In this paper, we propose a kernel Tweedie model Ktweedie and its sparse variant SKtweedie that can simultaneously address the above challenges. Specifically, nonlinear effects and high-order interactions can be flexibly represented through a wide range of kernel functions, which is fully learned from the data; In addition, while the Ktweedie can handle high-dimensional data, the SKtweedie with integrated variable selection can further improve the interpretability. We perform extensive simulation studies to justify the prediction and variable selection accuracy of our method, and demonstrate the applications in ratemaking and loss-reserving in general insurance. Overall, the Ktweedie and SKtweedie outperform existing Tweedie models when there exist nonlinear effects and high-order interactions, particularly when the dimensionality is high relative to the sample size. The model is implemented in an efficient and user-friendly R package ktweedie (https://cran.r-project.org/package=ktweedie). Supplementary material is available online. Keywords: Insurance claims data; Loss-reserving; Ratemaking; Sparse kernel methods; Zero inflated data.