Yindalon Aphinyanaphongs

About

Yindalon Aphinyanaphongs, MD, PhD, is the Director of Translational Clinical Informatics for DataCore at NYU Grossman School of Medicine. Her role involves leading efforts in clinical informatics with a focus on translational applications. Her background includes medical training and doctoral research, which contribute to her expertise in data-driven clinical informatics. She is part of the Health Tech Hub team, working on innovative health technology initiatives that integrate clinical data and informatics to improve healthcare delivery and research.

Research topics

• Medicine
• Internal medicine
• Artificial Intelligence
• Machine Learning
• Computer Science
• Software engineering
• Psychology
• Gastroenterology
• Immunology
• Emergency medicine

Selected publications

• Thrombosis in Hospitalized Patients With COVID-19 in a New York City Health System

  JAMA · 2020 · 843 citations

  This study examines the incidence of and risk factors for venous and arterial thrombosis in patients hospitalized with COVID-19 in 4 New York City hospitals.

  DOI

• Health system-scale language models are all-purpose prediction engines

  Nature · 2023 · 432 citations

  to train a large language model for medical language (NYUTron) and subsequently fine-tune it across a wide range of clinical and operational predictive tasks. We evaluated our approach within our health system for five such tasks: 30-day all-cause readmission prediction, in-hospital mortality prediction, comorbidity index prediction, length of stay prediction, and insurance denial prediction. We show that NYUTron has an area under the curve (AUC) of 78.7-94.9%, with an improvement of 5.36-14.7%…

  DOI

• The TRIPOD-LLM reporting guideline for studies using large language models

  Nature Medicine · 2025-01-01 · 312 citations

  reviewOpen access
  PublisherOA PDFDOI

• Platelets contribute to disease severity in COVID‐19

  Journal of Thrombosis and Haemostasis · 2021 · 201 citations

  DOI

• Medical large language models are vulnerable to data-poisoning attacks

  Nature Medicine · 2025-01-08 · 146 citations

  articleOpen access

  The adoption of large language models (LLMs) in healthcare demands a careful analysis of their potential to spread false medical knowledge. Because LLMs ingest massive volumes of data from the open Internet during training, they are potentially exposed to unverified medical knowledge that may include deliberately planted misinformation. Here, we perform a threat assessment that simulates a data-poisoning attack against The Pile, a popular dataset used for LLM development. We find that replacemen…

  PublisherOA PDFDOI

Frequent coauthors

• Vincent J. Major

  NYU Langone Health

  38 shared

• Constantin F. Aliferis

  University of Minnesota

  28 shared

• Neil Jethani

  26 shared

• Jonathan Austrian

  New York University

  20 shared

• Rajesh Ranganath

  Courant Institute of Mathematical Sciences

  18 shared

• Narges Razavian

  NYU Langone Health

  15 shared

• Alisa Surkis

  13 shared

• Lawrence D. Fu

  NYU Langone Health

  12 shared

Similar researchers at New York University

• Xi Chenh-121
• Kyunghyun Choh-90
• Gérard Ben Aroush-53