JSJunming (Seraphina) Shi

Junming (Seraphina) Shi

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

Junming (Seraphina) Shi has a total of 13 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Highly adaptive LASSO: Machine learning that provides valid nonparametric inference in realistic models

    by Zachary Butzin-Dozier, Sky Qiu, Alan E. Hubbard, Junming (Seraphina) Shi, Mark J. van der Laan

    A bstract: Understanding treatment effects on health-related outcomes using real-world data requires defining a causal parameter and imposing relevant identification assumptions to translate it into a statistical estimand. Semiparametric methods, like the targeted maximum likelihood estimator (TMLE), have been developed to construct asymptotically linear estimators of these parameters. To further establish the asymptotic efficiency of these estimators, two conditions must be met: 1) the relevant components of the data likelihood must fall within a Donsker class, and 2) the estimates of nuisance parameters must converge to their true values at a rate faster than n −1/4 . The Highly Adaptive LASSO (HAL) satisfies these criteria by acting as an empirical risk minimizer within a class of càdlàg functions with a bounded sectional variation norm, which is known to be Donsker. HAL achieves the desired rate of convergence, thereby guaranteeing the estimators’ asymptotic efficiency. The function class over which HAL minimizes its risk is flexible enough to capture realistic functions while maintaining the conditions for establishing efficiency. Additionally, HAL enables robust inference for non-pathwise differentiable parameters, such as the conditional average treatment effect (CATE) and causal dose-response curve, which are important in precision health. While these parameters are often considered in machine learning literature, these applications typically lack proper statistical inference. HAL addresses this gap by providing reliable statistical uncertainty quantification that is essential for informed decision-making in health research.


  • Open Access

    Predicting Long COVID in the National COVID Cohort Collaborative Using Super Learner: Cohort Study

    by Zachary Butzin-Dozier, Yunwen Ji, Haodong Li, Jeremy Coyle, Junming Shi, Rachael V Phillips, Andrew N Mertens, Romain Pirracchio, Mark J van der Laan, Rena C Patel, et al.

    Background: Postacute sequelae of COVID-19 (PASC), also known as long COVID, is a broad grouping of a range of long-term symptoms following acute COVID-19. These symptoms can occur across a range of biological systems, leading to challenges in determining risk factors for PASC and the causal etiology of this disorder. An understanding of characteristics that are predictive of future PASC is valuable, as this can inform the identification of high-risk individuals and future preventative efforts. However, current knowledge regarding PASC risk factors is limited. Objective: Using a sample of 55,257 patients (at a ratio of 1 patient with PASC to 4 matched controls) from the National COVID Cohort Collaborative, as part of the National Institutes of Health Long COVID Computational Challenge, we sought to predict individual risk of PASC diagnosis from a curated set of clinically informed covariates. The National COVID Cohort Collaborative includes electronic health records for more than 22 million patients from 84 sites across the United States. Methods: We predicted individual PASC status, given covariate information, using Super Learner (an ensemble machine learning algorithm also known as stacking) to learn the optimal combination of gradient boosting and random forest algorithms to maximize the area under the receiver operator curve. We evaluated variable importance (Shapley values) based on 3 levels: individual features, temporal windows, and clinical domains. We externally validated these findings using a holdout set of randomly selected study sites. Results: We were able to predict individual PASC diagnoses accurately (area under the curve 0.874). The individual features of the length of observation period, number of health care interactions during acute COVID-19, and viral lower respiratory infection were the most predictive of subsequent PASC diagnosis. Temporally, we found that baseline characteristics were the most predictive of future PASC diagnosis, compared with characteristics immediately before, during, or after acute COVID-19. We found that the clinical domains of health care use, demographics or anthropometry, and respiratory factors were the most predictive of PASC diagnosis. Conclusions: The methods outlined here provide an open-source, applied example of using Super Learner to predict PASC status using electronic health record data, which can be replicated across a variety of settings. Across individual predictors and clinical domains, we consistently found that factors related to health care use were the strongest predictors of PASC diagnosis. This indicates that any observational studies using PASC diagnosis as a primary outcome must rigorously account for heterogeneous health care use. Our temporal findings support the hypothesis that clinicians may be able to accurately assess the risk of PASC in patients before acute COVID-19 diagnosis, which could improve early interventions and preventive care. Our findings also highlight the importance of respiratory characteristics in PASC risk assessment. International Registered Report Identifier (IRRID): RR2-10.1101/2023.07.27.23293272


  • Open Access

    Implicit bias in Critical Care Data: Factors affecting sampling frequencies and missingness patterns of clinical and biological variables in ICU Patients

    by Junming (Seraphina) Shi, Alan E. Hubbard, Nicholas Fong, Romain Pirracchio

    The presence of missing values in Electronic Health Records (EHRs) is a widespread and inescapable issue. Publicly available data sets mirror the incompleteness found in EHRs. Although the existing literature largely approaches missing data as a random phenomenon, the mechanisms behind these missing values are often not random with respect to important characteristics of the patients. Similarly, the sampling frequency of clinical or biological parameters is likely informative. The possible informative nature of patterns in missing data is often overlooked. For both missingness and sampling frequency, we hypothesize that the underlying mechanism may be at least consistent with implicit bias. To investigate this important issue, we introduce a novel analytical framework designed to rigorously examine missing data and sampling frequency in EHRs. We utilize the MIMIC-III dataset as a case study, given its frequent use in training machine learning models for healthcare applications. Our approach incorporates Targeted Machine Learning (TML) to study the impact of a series of demographic variables, including protected attributes such as age, sex, race, and ethnicity on the rate of missing data and sampling frequency for key clinical and biological variables in critical care settings. Our results expose underlying differences in the sampling frequency and missing data patterns of vital sign measurements and laboratory tests between different demographic groups. In addition, we find that these measurement patterns can provide significant predictive insights into patient outcomes. Consequently, we urge a reevaluation of the conventional understanding of missing data and sampling frequencies in EHRs. Acknowledging and addressing these biases is essential for advancing equitable and accurate healthcare through machine learning applications.


  • Open Access

    Predicting Long COVID in the National COVID Cohort Collaborative Using Super Learner: Cohort Study (Preprint)

    by Zachary Butzin-Dozier, Yunwen Ji, Haodong Li, Jeremy Coyle, Junming Shi, Rachael V Phillips, Andrew N Mertens, Romain Pirracchio, Mark J van der Laan, Rena C Patel, et al.

    BACKGROUND Postacute sequelae of COVID-19 (PASC), also known as long COVID, is a broad grouping of a range of long-term symptoms following acute COVID-19. These symptoms can occur across a range of biological systems, leading to challenges in determining risk factors for PASC and the causal etiology of this disorder. An understanding of characteristics that are predictive of future PASC is valuable, as this can inform the identification of high-risk individuals and future preventative efforts. However, current knowledge regarding PASC risk factors is limited. OBJECTIVE Using a sample of 55,257 patients (at a ratio of 1 patient with PASC to 4 matched controls) from the National COVID Cohort Collaborative, as part of the National Institutes of Health Long COVID Computational Challenge, we sought to predict individual risk of PASC diagnosis from a curated set of clinically informed covariates. The National COVID Cohort Collaborative includes electronic health records for more than 22 million patients from 84 sites across the United States. METHODS We predicted individual PASC status, given covariate information, using Super Learner (an ensemble machine learning algorithm also known as stacking) to learn the optimal combination of gradient boosting and random forest algorithms to maximize the area under the receiver operator curve. We evaluated variable importance (Shapley values) based on 3 levels: individual features, temporal windows, and clinical domains. We externally validated these findings using a holdout set of randomly selected study sites. RESULTS We were able to predict individual PASC diagnoses accurately (area under the curve 0.874). The individual features of the length of observation period, number of health care interactions during acute COVID-19, and viral lower respiratory infection were the most predictive of subsequent PASC diagnosis. Temporally, we found that baseline characteristics were the most predictive of future PASC diagnosis, compared with characteristics immediately before, during, or after acute COVID-19. We found that the clinical domains of health care use, demographics or anthropometry, and respiratory factors were the most predictive of PASC diagnosis. CONCLUSIONS The methods outlined here provide an open-source, applied example of using Super Learner to predict PASC status using electronic health record data, which can be replicated across a variety of settings. Across individual predictors and clinical domains, we consistently found that factors related to health care use were the strongest predictors of PASC diagnosis. This indicates that any observational studies using PASC diagnosis as a primary outcome must rigorously account for heterogeneous health care use. Our temporal findings support the hypothesis that clinicians may be able to accurately assess the risk of PASC in patients before acute COVID-19 diagnosis, which could improve early interventions and preventive care. Our findings also highlight the importance of respiratory characteristics in PASC risk assessment. INTERNATIONAL REGISTERED REPORT RR2-10.1101/2023.07.27.23293272