My Research
My interests include the application and invention of machine learning models to answer complex, real-world health problems. In my doctoral position at Columbia University Medical Center, I worked with Dr. Adler Perotte. Together, we developed novel deep-learning and probabilistic methods that support causal claims generated from observational electronic health record (EHR) data. (Dissertation Title, Machine Learning Methods for Causal Inference with Observational Biomedical Data. Watch it here.) Prior to my doctoral candidacy, I completed a master’s degree in Biostatistics and Epidemiology at Columbia University Mailman School of Public Health, where I studied under Dr. Stephen Morse. Currently, I work on the Clinical Informatics team at the Regeneron Genetics Center.
Publications
Averitt AJ, Vanitchanant N, Ranganath R, Perotte A. The Counterfactual Chi-GAN. J Biomed Inform. Sep 2020. *Highlighted in OHDSI Symposium 2020 Year in Review. link.
Averitt AJ, Weng C, Ryan P, Perotte A. Translating evidence into practice: eligibility criteria fail to eliminate clinically significant differences between real-world and study populations. npj Digital Medicine. May 2020. *Highlighted in OHDSI Symposium 2020 Year in Review. link.
Averitt AJ, Slovis BH, Tariq A, Vawdrey D, Perotte A. Characterizing the Opioid Epidemic Using Electronic Health Records. JAMIA Open. Nov 2018.*Honorable Mention in AMIA 2020 Year in Review. link.
Averitt AJ, Natarajan K. Going Deep: The Role of Neural Networks for Renal Survival and Beyond. KI Reports. 2017. link.
Slovis BH, Averitt AJ, Glassberg J, Lowry T Kuperman G, Shapiro J. 175 Hospital Crossover of Patients With Sickle Cell Disease in a Health Information Exchange. Annals of Emergency Medicine, Volume 68, Issue 4, S69. Oct 2016. link.
Buono J, Mathur K, Averitt AJ, Andrae DA. Economic Burden of Irritable Bowel Syndrome with Diarrhea: Retrospective Analysis of a U.S. Commercially Insured Population. JMCP. Nov 2016. link.
Shust G, Jao J, Rodriguez-Caprio G, Posada R, Chen K, Averitt AJ, Sperling R. Salvage Regimens Containing Darunavir, Etravirine, Raltegravir, or Enfuvirtide in Highly Treatment-Experienced Perinatally Infected Pregnant Women. Journal of Pediatric Infectious Disease Society. 2013. link.
Averitt AJ. FTC Hold’s Roundtable Discussion Regarding Motor Vehicle Industry. FTC: Watch [Alexandria] 18 Oct. 2011, No.793 ed. Print.
Averitt AJ. Gilead Science vies for purchase of Pharmasset. FTC: Watch [Alexandria] 1 Dec. 2011, No. 796 ed. Print.
Podium Talks
Averitt AJ, Vanitchanant N, Ranganath R. Perotte A. Adversarially-Learned Balancing Weights for Causal Inference. American Medical Informatics (AMIA) Symposium. Virtual. 2020.
Averitt AJ, Slovis BH, Tariq A, Vawdrey D, Perotte A. Characterizing the Urban Opioid Epidemic Using Electronic Health Records. American Medical Informatics (AMIA) Symposium. Washington DC. 2019. *Winner, Top 10 Video Abstract.
Averitt AJ, Perotte AJ. Noisy-Or Risk Allocation for Causal Inference. American Medical Informatics (AMIA) Symposium. San Francisco, CA. 2018.
Averitt AJ, Weng C, Perotte AJ. Clinical Trial Eligibility Criteria Fail to Meet Burden of Generalizability. American Medical Informatics (AMIA) Symposium. Washington DC. 2017.
Slovis BH, Averitt AJ, Kuperman G, Glassberg T, Lowry T. Hospital Crossover of Patients with Sickle Cell Disease in a Health Information Exchange. ACEP. Las Vegas, NV. 2016.
Invited Talks
Averitt AJ. Noisy-Or Risk Allocation: A Probabilistic Model for Attributable Risk Estimation. OHDSI Stakeholder Meeting. November 17, 2020.
Averitt AJ. Causal Inference in Health. Columbia University. Digital Health, Prof. Noémie Elhadad. February 24, 2020.
Averitt AJ, Slovis BH. Characterizing the Opioid Epidemic Using Electronic Health Records. Thomas Jefferson University, Public Health Journal Club. Philadelphia, PA. January 27, 2020.
Averitt AJ. Characterizing the Opioid Epidemic Using Electronic Health Records. Emerging Addiction Science Workshop, Columbia University, Department of Psychiatry. New York, NY. January 16, 2020.
Averitt AJ. The Counterfactual Chi-GAN. OHDSI Stakeholder Meeting. October 22, 2019.
Averitt AJ. Machine Learning for Causal Inference Health. New York University (NYU) Machine Learning, Prof. Rajesh Ranganath. New York, NY. April 16, 2019.
Posters
Averitt AJ, Perotte A. Noisy-Or Risk Allocation: A Probabilistic Model for Attributable Risk Estimation. Observational Health and Data Science (OHDSI) Symposium. Virtual. 2020. *Winner, Best Methodological Contribution. link.
Averitt AJ, Vanitchanant N, Ranganath R. Perotte A. The Counterfactual Chi-GAN. Observational Health and Data Science (OHDSI) Symposium. Bethesda, MD. 2019. *Winner, Best Methodological Contribution. link.
Averitt AJ, Vanitchanant N, Ranganath R. Perotte A. The Counterfactual Chi-GAN. Atlantic Causal Inference Conference (ACIC). Montreal, Québec. 2019.
Averitt AJ, Vanitchanant N, Perotte A. The Counterfactual Chi-GAN. New York Academy of Science (NYAS) Machine Learning Symposium. New York, NY. 2019.
Averitt AJ, Perotte AJ. Noisy-Or Risk Allocation for Causal Inference. National Library of Medicine (NLM) Training Conference. Nashville, TN. 2018. *Winner, Best Poster Presentation.
Averitt AJ, Weng C, Perotte AJ. Do RCT Eligibility Criteria Identify Applicable Patients for Evidence-Based Medicine? Observational Health and Data Science (OHDSI) Symposium. Bethesda, MD. 2018.
Averitt AJ, Perotte AJ. Standardization of FDA Adverse Event Reporting System to the OHDSI Common Data Model. American Medical Informatics (AMIA) Symposium. Chicago, IL. 2016.
Buono J, Mathur K, Averitt AJ, Andrae DA. Economic Burden of Treatment Failure Among US Commercially Insured Patients with Irritable Bowel Syndrome with Diarrhea. AMCP Nexus. Orlando, FL. 2015.
Buono J, Mathur K, Averitt AJ, Andrae DA. Economic Burden of Irritable Bowel Syndrome With Diarrhea: Retrospective Analyses of a US Commercially Insured Population. American College of Gastroenterology. Honolulu, Hawaii. 2015.
Awards
Best Methodological Contribution. Noisy-Or Risk Allocation: A Probabilistic Model for Attributable Risk Estimation. Observational Health and Data Science (OHDSI) Symposium. 2020. Oct 18-19, 2020. Virtual.
Top 10 Video Abstract. Characterizing the Urban Opioid Epidemic Using Electronic Health Records. American Medical Informatics (AMIA) Symposium. Nov 16-20, 2019. Washington, DC. link.
Best Methodological Contribution.The Counterfactual Chi-GAN. Observational Health and Data Science (OHDSI) Symposium. Sept 16, 2019. Bethesda, MD.
Best Poster Presentation. Noisy-Or Risk Allocation for Causal Inference. NLM Informatics Training Conference. June 4-5, 2018. Nashville, TN.
Contact
Feel free to send me an email me about collaborations, research and work opportunities, or to just say hello!
amelia.averitt@gmail.com
aja2149@caa.columbia.edu
Tweets by AJAveritt