About

Deborah Stevenson is the Associate Vice Provost and Dean of Academic Advising at Stanford University, overseeing Academic Advising Operations. She previously served as the founding Director of the Center for Student Academic Success at Gonzaga University, where she created an integrated center providing academic advising, supplemental learning support, and disability services. Deborah has experience teaching first-year experience and academic recovery courses, consulting with faculty on course design and classroom management, and serving as a regional accreditation evaluator. As a first-generation college student who faced significant challenges during her undergraduate education, she is motivated by a commitment to student success and creating equitable, adaptive learning environments that empower students to become active and independent learners. Deborah holds an undergraduate degree in political science and a master’s degree in organizational leadership, both from Gonzaga University.

Research topics

• Medicine
• Internal medicine
• Pediatrics
• Psychiatry
• Biology
• Bioinformatics
• Radiology
• Surgery
• Physical therapy
• Emergency medicine
• Obstetrics
• Intensive care medicine
• Anesthesia
• Physiology
• Genetics
• Computational biology

Selected publications

• Evidence That Cerebral Visual Impairment May Evolve after Initial Brain Injury

  Ophthalmology · 2026-01-07

  articleOpen access
  PublisherOA PDFDOI

• AI Guided Parenteral Nutrition Therapy After Hematopoietic Stem Cell Transplantation

  npj Digital Medicine · 2026-05-06

  articleOpen access

  Adults undergoing hematopoietic cell transplantation often develop serious complications that cause rapid nutritional decline. We developed and evaluated an AI approach to standardize intravenous nutrition (total parenteral nutrition, TPN) during this vulnerable period. Using real-world records from Stanford Health Care (6402 transplants, 2008-2025), we analyzed 1473 adults who received TPN, totaling 27,447 patient-days, linking each day's clinical state to the next day's prescription. We created a library of 30 standardized TPN regimens and trained a model to recommend next-day dose adjustments based on laboratory data and the existing prescription. (Pearson r ≈ 0.71). We then assessed an AI policy learned from past care and found that the Reinforcement learning agent selected dose adjustments with a higher composite score than the existing clinical policy. These results show that AI-guided TPN is feasible and may enhance bedside decision-making for adult transplant care, warranting prospective evaluation.

  PublisherOA PDFDOI

• Leveraging health records to identify diagnoses associated with recurrent pregnancy loss across two medical centers

  iScience · 2026-01-07

  articleOpen access

  Recurrent pregnancy loss (RPL), defined as 2 or more pregnancy losses, affects 5-6% of ever-pregnant individuals. Approximately half of these cases have no identifiable explanation. In this study, we aim to identify diagnoses associated with RPL and generate hypotheses about RPL etiology utilizing electronic health record (EHR) data. We implemented a case-control study comparing the history of over 1,600 diagnoses between RPL and live birth patients, leveraging the University of California San Francisco (UCSF) and Stanford University EHR databases. In total, our study includes 8,496 RPL (UCSF: 3,840, Stanford: 4,656) and 53,278 control (UCSF: 17,259, Stanford: 36,019) patients. Menstrual abnormalities and infertility-associated diagnoses are significantly positively associated with RPL in both medical centers. Age-stratified analysis revealed that the majority of RPL-associated diagnoses have higher odds ratios for patients <35 years compared with 35+ years patients. While Stanford results are sensitive to control for healthcare utilization, UCSF results are stable across analyses with and without utilization.

  PublisherDOI

• Author Correction: Machine learning models predict long COVID outcomes based on baseline clinical and immunologic factors

  Communications Medicine · 2026-02-23

  articleOpen access
  PublisherOA PDFDOI

• Machine learning models predict long COVID outcomes based on baseline clinical and immunologic factors

  Communications Medicine · 2026-01-03 · 1 citations

  articleOpen access

  The post-acute sequelae of SARS-CoV-2 (PASC), also known as long COVID, remain a significant health issue that is incompletely understood. Predicting which acutely infected individuals will develop long COVID is challenging due to the absence of established biomarkers, clear disease mechanisms, or well-defined sub-phenotypes. Machine learning (ML) models may address this gap by leveraging clinical data to enhance diagnostic precision. Clinical data, including antibody titers and viral load measurements collected at the time of hospital admission, are used to predict the likelihood of acute COVID-19 progressing to long COVID. Machine learning models are trained and evaluated for predictive performance. Feature importance analysis is performed to identify the most influential predictors. The machine learning models achieve median AUROC values ranging from 0.64 to 0.66 and AUPRC values between 0.51 and 0.54, demonstrating predictive capabilities. Low antibody titers and high viral loads at hospital admission emerge as the strongest predictors of long COVID outcomes. Comorbidities—such as chronic respiratory, cardiac, and neurologic diseases—and female sex are also identified as significant risk factors. Machine learning models identify patients at risk for developing long COVID based on baseline clinical characteristics. These models guide early interventions, improve patient outcomes, and mitigate the long-term public health impacts of SARS-CoV-2. Long COVID, or post-acute sequelae of SARS-CoV-2, is a prolonged health condition that can occur after acute COVID-19 infection. However, the ability to predict who will develop long COVID remains limited due to the absence of clear tests or biomarkers. We looked at patients’ medical information, including the amount of virus in their body at hospital admission, and how strong their immune response was. Using computer programs that can find hidden patterns in large sets of data, we discovered that people with a weaker immune response, higher amounts of virus, certain long term health problems and women are more likely to develop long COVID. This study highlights that computer-based tools could help doctors identify high-risk patients early and provide care that may prevent long-term complications. Jayavelu, Samaha et al., apply machine learning models on hospital admission data, including antibody titers and viral load, to identify patients at high risk for Long COVID. Low antibody levels, high viral loads, chronic diseases, and female sex are key predictors, supporting early, targeted interventions.

  PublisherOA PDFDOI

• Empiric azithromycin alters the upper respiratory microbiome and resistome without anti-inflammatory benefit in COVID-19

  Nature Microbiology · 2026-03-16 · 1 citations

  articleOpen access

  Azithromycin is a widely used antibiotic and was frequently used to treat hospitalized patients during the COVID-19 pandemic. The impact of empiric azithromycin use on the respiratory microbiome in patients with viral respiratory infections is unclear. Here we used longitudinal metatranscriptomics on nasal swabs from a prospective multicentre cohort of 1,164 patients hospitalized for COVID-19. We compared the upper respiratory microbiome, resistome and systemic immune response in patients treated with azithromycin (n = 366) with those who received no antibiotics (n = 474) or other antibiotics (n = 324). We found that azithromycin altered microbiome composition and increased the expression and relative proportion of macrolide/lincosamide/streptogramin (MLS) resistance genes. These changes occurred after 1 day of exposure and persisted for over a week. MLS resistance gene expression was associated with commensals and potential pathogens, while there were no differences in host inflammatory gene expression in blood and airways. This demonstrates that empiric azithromycin treatment impacts the upper respiratory microbiome and resistome without apparent anti-inflammatory benefit.

  PublisherOA PDFDOI

• Single-cell-level digital twins for preterm birth prevention strategies

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-10-01 · 1 citations

  preprintOpen access

  Abstract Digital twin models can accelerate therapeutic development by enabling low-risk testing of candidate interventions. In preterm labor (PTL), a major pregnancy complication where clinical trials face unique ethical and financial barriers, digital twins are especially valuable for evaluating new therapies targeting immune dysfunctions driving PTL. Yet, current models lack single-cell resolution, limiting detection of cell-type-specific mechanisms, off-target effects, and the design of personalized interventions. We present Simulated Immunome Modeling of Clinical Outcomes (SIMCO), a single-cell-level digital twin framework that models immunomodulatory treatment effects on the timing of labor using immunome-wide, single-cell simulations. SIMCO’s digital twins are trained and validated on a newly generated mass cytometry atlas of the pregnant immunome exposed to nine candidate drugs preselected for PTL prevention. Applying SIMCO to an independent cohort of pregnant individuals, we simulate treatment effects on gestational length, screening for candidate drugs that delay labor timing and providing system-level mechanistic insight for each drug candidate. Tetrahydrofolate, maprotiline, and the combination of aspirin and lansoprazole emerged as top candidates for PTL prevention, delaying labor onset primarily through enhanced mTOR signaling in innate immune cells and attenuated JAK/STAT signaling in naïve CD4⁺ T cells. The codebase is available at https://github.com/ofondeur/SIMCO/ .

  PublisherOA PDFDOI

• Prenatal wildfire smoke, first trimester proteome, and subsequent child behavioral outcomes: An exploratory study

  Environmental Research · 2025-12-17

  articleOpen access
  PublisherOA PDFDOI

• Multiomic approach to study the immunologic profile preceding preterm birth

  Placenta · 2025-11-01

  article
  PublisherDOI

• A Transcriptomics-Based Computational Drug Repurposing Pipeline Identifies Simvastatin And Primaquine As Therapeutics For Endometriosis

  bioRxiv (Cold Spring Harbor Laboratory) · 2025-05-30 · 1 citations

  preprintOpen access

  ABSTRACT Endometriosis has limited treatment options, prompting the search for novel therapeutics. We previously used a transcriptomics-based computational drug repositioning pipeline to analyze public bulk transcriptomic data of eutopic endometrium from cases and controls and identified several drug candidates. Fenoprofen, our top in silico candidate, was validated in a rat model of endometriosis-associated pain. Building on this, we evaluated herein two additional candidates, simvastatin (a cholesterol-lowering drug) and primaquine (an antimalarial), based on strong endometrial gene expression reversal scores and favorable safety profiles. Using the rat model, we conducted behavioral testing, bulk RNA sequencing, and differential expression analysis to assess their therapeutic potential. We also assessed endometriosis diagnosis among patients prescribed simvastatin in electronic medical records (EMR) across six University of California (UC) healthcare institutions. In vivo validation using a rat model of endometriosis demonstrated that both simvastatin and primaquine significantly reduced vaginal hyperalgesia, a surrogate marker of endometriosis-related pain. RNA-seq of uteri and lesions confirmed reversal of disease-associated gene expression signatures following treatment. Analysis of UC-wide EMR data found lower relative risk of endometriosis among those prescribed simvastatin compared to a matched control group. Overall, simvastatin and primaquine attenuated pain-associated behaviors and reversed endometriosis-related gene expression changes in an animal model. Moreover, simvastatin prescription was associated with a lower relative risk of endometriosis in our retrospective multi-center cohort study. These findings highlight their potential as repurposed therapeutics for endometriosis and support the effectiveness of computational drug repositioning in identifying new treatment strategies. One Sentence Summary Simvastatin and primaquine reduced endometriosis pain and reversed gene signatures, with simvastatin also linked to lower disease risk.

  PublisherOA PDFDOI

Recent grants

• NIH Grant R01HL058013

  NIH · $1.4M · 2007

• NIH Grant U10HD027880

  NIH · $8.5M · 2016

• NIH Grant R01HL068703

  NIH · $1.8M · 2011

• NIH Grant M01RR000070

  NIH · $27.1M · 2011

• NIH Grant R21HD050351

  NIH · $519k · 2010

Frequent coauthors

• Hendrik J. Vreman

  400 shared

• Ronald J. Wong

  PragmatIC (United Kingdom)

  343 shared

• Gary M. Shaw

  Stanford University

  296 shared

• Daniel S. Seidman

  Tel Aviv University

  287 shared

• Rena Gale

  Herzog Hospital

  249 shared

• Barbara J. Stoll

  Children's Nutrition Research Center at Baylor College of Medicine

  204 shared

• Arie Laor

  190 shared

• Jon E. Tyson

  The University of Texas Health Science Center at Houston

  187 shared