Diego Martinez, PhD
Johns Hopkins' Center for Data Science in Emergency Medicine
I am Assistant Professor of Emergency Medicine and of Health Sciences Informatics at Johns Hopkins University and investigator at the Malone Center for Engineering in Healthcare and the Center for Data Science in Emergency Medicine. My research area is applied machine learning and operational research for complex healthcare delivery systems. Applications include predictive clinical decision support for triage and disease diagnostics and prescriptive capacity management for hospital operations, emphasizing reducing inequalities and improving healthcare. I teach the informatics seminar and grand rounds at Johns Hopkins School of Medicine and a graduate class on secondary uses of electronic health records data at the Johns Hopkins Bloomberg School of Public Health. Some of my current projects include:
​
​COVID-19 Response
  • Health Disparities (with Kathleen Page, Felipe Feijoo, Richard Rothman, Eili Klein, Jeremiah Hinson and Scott Levin). We are creating data and systems to monitor and minimize the impact of COVID-19 in socially marginalized groups through enhanced testing and vaccination campaigns. With the support of the NIH.
  • ​Capacity Management (with Jim Scheulen). We are using advanced modeling methods to better manage our health system’s capacity to confront surges of COVID-19. With the support of the Johns Hopkins Health System.
Medical Decision Making
  • Connected Emergency Care (with Scott Levin and Jeremiah Hinson). We are developing predictive EMR-integrated decision support tools to reduce health and financial harm for patients with respiratory tract infections seen in the ED. With the support of the AHRQ.
  • Kidney Injury Management (with Scott Levin and Jeremiah Hinson). We are developing algorithms to predict the onset of acute kidney injury in ED patients, and deploying EMR-integrated decision support for ED clinicians. With the support of the AHRQ.
Hospital Operations
  • Waiting List Management (with Diana Prieto, Jose Zayas, and Felipe Feijoo). We are investigating the effects of prolonged waiting in countries with universal healthcare, and developing data-driven waiting list management systems. With the support of the AHW.
  • Epidemics on Networks (with Eili Klein). We are investigating the epidemic processes by which infectious diseases spread over networks of contact between healthcare workers and patients, and developing decision support to minimize hospital-acquired infection risks. With the support of the CDC​.
I always have time for cool project ideas, collaborations or speaking -- write me at dmart101 at jhu dot edu

2020
Nov 2020 (Infection Control & Hospital Epidemiology)
Lin, Tseng, Martinez, Klein
We model the spread of carbapenem-resistant Enterobacteriaceae outbreaks throughout hospitals in Maryland and examined the impact of coordinated infection control efforts.
Sep 2020 (Journal of Hospital Infection)
Hinson, Rothman, Carroll, Mostafa, Ghobadi, Smith, Martinez, Shaw-Saliba, Klein, Levin
We measured the clinical and operational benefits of targeted rapid molecular testing for SARS-CoV-2 in a health system.
Aug 2020 (JAMIA)
​Lin, Siddiqui, Bernstein, Martinez, Gardner, Albright, Igusa
We characterized collaborations of over 4,000 pharmaceutical, academic, healthcare, nonprofit, and government organizations in the clinical trials context and identify key features of successful collaboration
Dec 2020 (Invited Talk)
AI in Emergency Medicine: Designing and implementing the effector arm
Universidad de Chile, Chile
Dec 2020 (Invited Talk)
Implementing Safe and Reliable Machine Learning: Key areas of development
Pontificia Universidad Católica de Valparaíso, Chile
Jun 2020 (JAMA)
Martinez, Hinson, Klein, Irvin, Saheed, Page, Levin
We analyzed temporal trends in positivity rates for SARS-CoV-2 in the Baltimore–District of Columbia region by race/ethnicity
Jul 2020 (Annals of Emergency Medicine)
Martinez, Levin, Klein, Parikh, Menez, Taylor, Hinson
We show machine-learning-based classifiers can spot symptoms of a common kidney condition (AKI) 72 hours earlier than current practice in the emergency department
Mar 2020 (JAMIA)
Martinez, Cai, Oke, Jarrell, Feijoo, Appelbaum, Klein, Barnes, Levin
We characterize unit-to-unit hospital equipment sharing and its impact on shortages, and to evaluate a system-control tool that balances inventory across all care areas, enabling increased availability of pumps.
Apr 2020 (medRxiv)
Lin, Strauss, Pinz, Martinez, Tseng, Schueller, Gatalo, Yang, Levin, Klein
We show that the exponential growth in COVID-19 cases can be explained by transmission of asymptomatic and mild cases that are typically unreported at the beginning of pandemic events
Dec 2020 (Invited Talk)
Prediction of Acute Kidney Injury in the Emergency Department
McGill University, Canada

2019
Sep 2019 (JMIR)
Jiang, Siddiqui, Barnes, Barouch, Korley, Martinez, Toerper, Cabral, Hamrock, Levin
We developed a dynamic readmission risk prediction model that yields daily predictions for patients hospitalized with heart failure
Sep 2019 (IIE Healthcare Systems Engineering)
Martinez, Jalalpour, Efron, Levin
We explained how to assess the impact of process improvement interventions with routinely collected longitudinal hospital data
Jan 2019 (BMC Public Health)
Martinez, Zhang, Bastias, Feijoo, Hinson, Martinez, Dunstan, Levin, Prieto
We measured a possible link between lengthy waits to access health care with an increased risk of mortality in waiting lists not prioritized by universal healthcare (Plan AUGE) in Chile
Jan 2019 (Emergency Nurses Association)
Whalen, Gardner, Martinez, Henry, McKenzie, Hinson, Levin
We quantify and qualify our evolution from traditional ESI triage system to machine-learning-based system as an exemplar of transition from research to practice and nursing integration
Jan 2019 (Society for Healthcare Epidemiology of America)
Martinez, Cai, Goodman, Paul, Lessler, Levin, Toerper, Milstone, Klein
We measured the extent at which contact isolation is an effective intervention to control the spread of antibiotic resistant bacteria within hospital ICUs

2018
Nov 2018 (Joint Commission Journal on Quality and Patient Safety)
Kane, Scheulen, Puttgen, Martinez, Levin, Bush, Huffman, Jacobs, Rupani, Efron
We designed and implemented a command center to manage patient flow at a large urban academic hospital using systems engineering principles
Nov 2018 (AHRQ Research Grant)
Levin (PI), Hinson (PI), Martinez (Co-Inv)
To improve decision-making by emergency clinicians that relates to diagnosis, treatment, and disposition of patients with respiratory tract infections in the emergency department
Nov 2018 (Invited Talk)
Network science to understand the spread of antibiotic-resistant bacteria in networks of contacts between humans
Valparaiso, Chile
Nov 2018 (Annals of Emergency Medicine)
Hinson, Martinez, Cabral, George, Whalen, Hansoti, Levin
We synthesized what is known about the performance of emergency department triage systems
Jun 2018 (Journal of Medical Systems)
Martinez, Kane, Jalapour, Scheulen, Rupani, Toteja, Barbara, Bush, Levin
We created and implemented an electronic dashboard to monitor hospital patient flow
May 2018 (Applied Physics Lab)
Johns Hopkins Applied Physics Lab Network Science Seminar
Diego Martinez
Invited talk. Laurel, MA
Jan 2018 (Int J Emergency Medicine)
Hinson, Martinez, Schmitz, Toerper, Radu, Scheulen, Levin, Stewart de Ramirez
We measured the extent to which patient characteristics contribute to inaccurate emergency department triage
Jan 2018 (Geisinger Health System)
Geisinger Health Operations Research Seminar
Diego Martinez
Invited talk. Danville, PA
Jan 2018 (JHU Research Grant)
Prieto (PI), Martinez (Co-PI)
To reduce waiting lists in a health delivery network for low income patients in Chile
Jan 2018 (Annals of Emergency Medicine)
Hinson, Martinez, Grams, Levin
We use electronic health records and machine learning to develop a predictive model for acute kidney injury (AKI) with capacity for identifying patients at high risk of developing AKI within 7 days of their emergency department visit

2017
Dec 2017 (Annals of Emergency Medicine)
Mistry, Steward de Ramirez, Kelen, Schmitz, Levin, Martinez, Psoter, Anton, Hinson
We performed an accuracy and reliability evaluation of the emergency severity index for emergency department triage
Aug 2017 (CDC Research Grant)
Klein (PI), Martinez (Co-Inv)
To improve understanding of how biological factors associated with the transmission of multidrug-resistant organisms impact interventions to reduce hospital-associated infections
May 2017 (IIE Healthcare Systems Engineering)
Wan, Zhang, Witz, Musselman, Yi, Mullen, Benneyan, Zayas-Castro, Rico, Cure, Martinez
We synthesized what is known about preventable hospital readmissions in the US
Feb 2017 (JAMA Pediatrics)
Woods-Hill, Fackler, McMillan, Ascenzi, Voskertchian, Colantuoni, Martinez, Toerper, Levin, Milstone
We created a new clinical practice guideline to reduce unnecessary testing in a tertiary pediatric ICU, which reduced blood sample collections without increasing missed sepsis cases

2016
Feb 2016 (Trials)
Martinez, Tsalatsanis, Yalcin, Zayas-Castro, Djulvegovic
We model and optimize the opening of clinical trials at an academic medical center
Sep 2016 (Health Care Management Science)
Martinez, Feijoo, Zayas-Castro, Das, Levin
We mathematically model the tension between market incentives and federal regulations in hospital markets
May 2016 (Journal for Healthcare Quality)
Rico, Liu, Martinez, Huang, Zayas-Castro, Fabri
We measured the extent to which patient and hospital characteristics increase the ​risk of preventable hospital readmissions

2015
October 2015 (BMC Medical Informatics and Decision Making)
Strauss, Martinez, Garcia, Taylor, Mateja, Fabri, Zayas-Castro
We identified design features for interoperable electronic medical records from a physician perspective
Nov 2015 (Applied Clinical Informatics)
Martinez, Mora, Gemmani, Zayas-Castro
​We uncover information needs of physicians from outside hospitals while managing patients in the hospitalist service
Disclaimer: Any opinions, findings, conclusions, or recommendations expressed herein are those of the author and do not necessarily reflect the views of Johns Hopkins University or any other organization supporting our research.