ABOUT ME

I am a doctoral candidate in Operations Research at the Massachusetts Institute of Technology, advised by Prof. Dimitris Bertsimas. My primary research interests lie at the intersection of Optimization and Machine Learning, with applications to healthcare and insurance. My work explores practical solutions to address real-world industry needs drawn from conversations and requests from clinicians and health organizations, but solved in new, creative ways that leverage state-of-the-art analytics techniques. In this pursuit, I have developed novel machine learning algorithms for healthcare data as well as personalized predictive and prescriptive models based on structured and text-based information. Moreover, I am particularly interested in studying the implications of these models on automated decision making. For this reason, I aim to create the foundations, both in terms of theory and applications, in the area of algorithmic insurance, a topic that I initiated during my doctoral studies.

Prior to joining MIT, I worked as a Business Analyst in McKinsey&Company at Athens, Greece. I earned my BS in Management Science and Technology from the Athens University of Economics and Business where my focus was on Operations Research and Business Analytics.

Contact

Agni Orfanoudaki
MIT - ORC
Email: agniorf@mit.edu
GoogleScholar
LinkedIn
Curriculum Vitae


RESEARCH INTERESTS

  • Predictive and Prescriptive Analytics and their Applications
  • Health Care
  • Algorithmic Insurance
  • Personalized Medicine

    • EDUCATION

    Ph.D. Candidate in Operations Research, MIT, (expected 2021)
    B.Sc. Management Science & Technology, Athens University of Economics & Business, 2016
    Visiting Student. TUM School of Management, Technical University of Munich, 2015

    AWARDS AND HONORS

    INFORMS William Pierskalla Best Paper Award 2020, INFORMS, 2020
    Bayer Women in Operations Research Scholarship 2020, INFORMS, 2020
    Theodore Vassilakis Fellowship 2019, MIT, 2019
    First Prize at Assistive Technology Hack 2019 MIT, March 2019 (link)
    Winner of the Machine Learning Across Disciplines Challenge, MIT, February 2019 (link)
    First Prize at EarthHack 2019, MIT, January 2019 (link)
    Best Academic Performance Award, Athens University of Economics & Business, 2012-2016
    Winner of McKinsey Challenge Competition, Athens, July 2015

    TALKS AND PRESENTATIONS

    INFORMS Annual Meeting, 2020. Pricing the Cost of Algorithmic Risk.
    INFORMS Annual Meeting, 2020. Personalized Predictions and Prescriptions for COVID-19 Patients: A ML Approach.
    INFORMS Annual Meeting, 2020. Personalized Stroke Risk Estimation: The Impact of Missing Data Imputation.
    Cambridge University, COVID-19 Seminar, 2020. From Predictions to Prescriptions: A Data-Driven Response to COVID-19.
    Kellogg-Wharton, OM Workshop, 2020. Personalized Stroke Risk Estimation: The Impact of Missing Data Imputation.
    International Stroke Conference, 2020. The Non-Linear Framingham Stroke Risk Score.
    Society of Thoracic Surgeons Annual Meeting, 2020. Non-Linear Mortality and Morbidity Risk Score for Cardiac Surgery.
    Boston INFORMS Chapter Meeting, January 2020. Personalized Treatment for CAD Patients: A Machine Learning Approach.
    INFORMS Annual Meeting, 2019. Natural Language Processing from Radiology Reports for Ischemic Stroke Patients.
    INFORMS Annual Meeting, 2019. Interpretable Clustering: An Optimization Approach.
    INFORMS Healthcare, 2019. The Non-Linear Framingham Stroke Risk Score.
    NeurIPS, 2018. Imputation of Clinical Covariates in Time.
    American Heart Assocation Annual Meeting, 2018. Personalized treatment for CAD patients: Impact of Machine Learning.
    INFORMS Annual Meeting, 2018. Non-linear Stroke Risk prediction using data from the Framingham Heart Study.
    INFORMS Annual Meeting, 2018. Interpretable Clustering: An Optimal Trees Approach.
    Athens University of Economics and Business, 2017. Optimal Survival Trees.
    INFORMS Annual Meeting, 2017. Personalized treatment for CAD patients: Impact of Machine Learning.
    INFORMS Healthcare, 2017. Personalized predictive and prescriptive models for CAD patients: An Optimal Trees Approach.

    SERVICE AND OUTREACH

    Reviewer for INFORMS Journal on Optimization, Health Care Management Science, NeurIPS, PloS one, Machine Learning for Healthcare, Annals of Thoracic Surgery
    Vice President of the MIT Hellenic Student Association Board, 2019 - Present
    Operations Research Seminar Co-Organizer, Fall, 2019
    INFORMS Officer at the Operations Research Center, 2017