Sze-chuan Suen · University of Southern California · PhdFit
Resume-aware faculty matching
Find professors who actually fit you
Upload your resume. PhdFit's six research agents compare your background with faculty profiles, recent publications, lab focus, and outreach opportunities, then rank professors with evidence you can review.
Resume-aware matchingEvidence from recent researchOutreach drafts included
Advisor Match Brief Resume uploaded
Resume signals5 found
NLPHCIFairnessGraph MLPyTorch
Six-agent analysisRunning
Axel · Candidate AnalystResume parsed
Nova · Professor ResearcherRecent papers
Echo · Match ExplainerEvidence linked
Atlas · Strategy AdvisorFit ranked
Lyra · Outreach WriterDraft queued
Juno · Meeting Prep CompilerNext step ready
SC
Dr. Sarah Chen
Stanford · NLP & interpretability
91
fit
Fairness + NLP projects overlap with her recent work.
Strong methods fit: model evaluation and interpretability.
Clear outreach angle for current trustworthy AI research.
Outreach angle
Ask how her lab is extending interpretability methods into fairness audits for real-world AI systems.
Sze-chuan Suen
· Associate Professor Industrial and Systems EngineeringVerified
University of Southern California · Daniel J. Epstein Department of Industrial and Systems Engineering
Sze-chuan Suen is an Assistant Professor at the Daniel J. Epstein Department of Industrial and Systems Engineering and is affiliated with the Leonard D. Schaeffer Center for Health Policy and Economics at the University of Southern California. His research interests focus on simulation, dynamic systems modeling, Markov decision processes, optimization, and cost-effectiveness analysis. He applies these methodologies primarily in health policy modeling, disease control, and medical decision making. Professor Suen is actively engaged in mentoring and is currently accepting PhD students interested in healthcare questions that can be addressed using the aforementioned methods.
In 2022, mpox clade IIb disseminated around the world, causing outbreaks in more than 117 countries. Despite the decay of the 2022 epidemic and the increased immunity within sexual networks, mpox continues to persist in North America without extinction, raising concerns of future outbreaks. We combined phylodyamic inference and microsimulation modeling to understand the heterogeneous dynamics governing local mpox persistence in Los Angeles County (LAC) from 2023 to 2024. Our Bayesian phylodynamic analysis revealed a time-varying pattern of viral importations into the county, which seeded mpox outbreak clusters that display "stuttering chains" dynamics. Our phylodynamics-informed microsimulation model demonstrated that the mpox cases in LAC can be explained by a combination of waves of viral introductions, a median effective reproductive rate below one, and a return to near-baseline sexual behaviors after the 2022 epidemic. Our counterfactual scenario modeling showed that frequent public health interventions that either promote increased isolation of infectious individuals or enact behavior-modifying campaigns during the periods with the highest viral importation intensity are actionable and effective at curbing mpox cases. Our work highlights the factors that maintain present-day mpox dynamics in a large, urban US county and describes how to leverage these results into community-centered public health interventions.
In 2022, mpox clade IIb disseminated around the world, causing outbreaks in more than 117 countries. Despite the decay of the 2022 epidemic and the accumulation of immunity within queer sexual networks, mpox continues to persist at low incidence in North America without extinction, raising concerns of future outbreaks. We combined phylodynamic inference and microsimulation modeling to understand the heterogeneous dynamics governing local mpox persistence in Los Angeles County (LAC) from 2023–2024. Our Bayesian phylodynamic analysis revealed a time–varying pattern of viral importations into the county that seeded a skewed distribution of mpox outbreak clusters that display a ″stuttering chains″ dynamic. Our phylodynamics–informed microsimulation model demonstrated that the persistent number of mpox cases in LAC can be explained by a combination of waves of viral introductions, a median Rt significantly below one, and a return to near–baseline sexual behaviors that were altered during the 2022 epidemic. Finally, our counterfactual scenario modeling showed that public health interventions that either promote increased isolation of symptomatic, infectious individuals or enact behavior–modifying campaigns during the periods with the highest viral importation intensity are both actionable and effective at curbing mpox cases. Our work highlights the heterogeneous factors that maintain present–day mpox dynamics in a large, urban US county and describes how to leverage these results to design timely and community–centered public health interventions.
American Journal Of Pathology · 2025-07-08 · 3 citations
articleOpen access
CD5 antigen-like (CD5L) is a multifunctional glycoprotein characterized for its role in the lipid metabolism, particularly within macrophages. In the liver, CD5L strongly correlates with liver injury. This study explored the role of CD5L on liver lipid accumulation and inflammatory response. CD5L promoted lipid uptake in hepatocytes and stellate cells. In multiple models of liver injury, expression of Cd5l was associated with that of Clec4f, a marker for liver macrophages, consistent with its role as a macrophage survival factor. Transwell assay was used to demonstrate a novel function of CD5L on promoting migration of natural killer T cells. This effect was independent of CD36, the characterized receptor of CD5L. This effect was also observed with liver macrophages, which are the cellular source for CD5L, and blocking CD5L attenuated natural killer T cell migration induced by liver macrophages. Finally, plasma levels of CD5L correlated with poor patient response to immune check point therapy that is dependent on response of T-cell populations. In addition, plasma CD5L levels correlated with levels of steatosis and severity of steatotic liver injury. Given the association between liver steatosis and poor response to immune checkpoint therapy, these data suggest that plasma CD5L levels may serve as a predictive biomarker for patient response to immune checkpoint therapy.
Abstract Person-generated health data (PGHD) from smartphones/wearables are invaluable for precision health, a field promoting health equity through tailored disease prevention, detection, and intervention strategies. However, pervasive convenience sampling in extant PGHD research introduces selection biases that systematically underrepresent disadvantaged groups, limit model generalizability, and risk exacerbating health disparities. Benchmark PGHD (representative, validated, longitudinal, and frequently repeated) are urgently needed to support model equity. To address this fieldwide limitation, we established American Life in Realtime (ALiR), a longitudinal population health study involving PGHD collected from a probability-based, nationally representative cohort using study-provided Fitbits and (as needed) 4G tablets. As a result, ALiR's 1,038 participants are broadly representative across comprehensive sociodemographic, behavioral, and health-related US population norms, overcoming disparities in established convenience samples (e.g. NIH's All of Us; AoU). Only two sources of differential enrollment remained: older age (odds ratio [OR]: 1.27, 99% CI: 1.12–1.45) during consent, lower education (OR: 0.86, 99% CI: 0.79–0.94) during enrollment, though oversampling individuals without bachelor's degrees sufficiently counterbalanced the latter. An illustrative coronavirus disease 2019 classification model—chosen for global significance, known disparities in experience and outcomes, and methodological relevance—trained using ALiR performed equivalently when tested in sample (area under the curve [AUC] = 0.84, 95% CI: 0.79–0.89) and out of sample on AoU (AUC = 0.83, 95% CI: 0.78–0.89) overall, and in historically underserved subgroups (AUC = 0.82–1.0). Conversely, an identically trained classification model using AoU underperformed by 35% out of sample on ALiR (overall AUC = 0.68, 95% CI: 0.61–0.75 vs. AUC = 0.93, 95% CI: 0.91–0.96 in sample), with worse performance in older female and non-White subgroups (by 22–40%). Our results suggest that probability sampling and hardware provisioning enabled cohort inclusivity and generalizable model performance, supporting ALiR's benchmarking potential for equitable recruitment, PGHD collection, and precision health application.
Sequential decision-making problems in the context of uncertainty naturally arise in healthcare settings. In general, the frequency at which decisions can be made or changed is determined by physical limitations, such as the frequency of doctor’s visits or transplantation offers. Quantifying the benefits of increasing the frequency of decision-making allows us to quantify the value of changing these physical constraints and thus improve the quality of care. In this article, we study the value provided by having additional decision-making opportunities in each epoch. We model this problem using a Markov Decision Process (MDP) framework. We provide structural properties of the optimal policies and quantify the difference in optimal values between MDP problems of different decision-making frequencies. We analyze numerical examples using liver transplantation in high-risk patients and treatment initiation in chronic kidney disease to illustrate our findings.
