Skip to content

Details

We are anticipating another great Meetup. Many thanks to Brock Napierkowski and Booz Allen Hamilton for sponsoring the event and providing FREE pizza and beer! We will start at 6:30 with pizza and networking. Following at 7:00, we will move into the meeting room for two great talks on using machine learning (to include a Rubin Causal modeling approach) to address needs within healthcare. We are excited to have Mike Thurber, Lead Data Scientist at Elder Research, and Monica Rajendiran, Data Scientist at CCRi, as our speakers. Please find the title and abstract for their talks below:

Speaker: Monica Rajendiran

Title:
Predicting patient revisits at the University of Virginia Health System Emergency Department

Abstract:
This study focuses on the predictive identification of patients frequently revisiting the University of Virginia Health System Emergency Department. Identifying these patients can help the Emergency Department formulate strategies that improve patient care and decrease excess Emergency Department utilization. The Health System in particular faces a number of unique challenges in its ongoing mission to reduce extraneous patient revisits. In addition to its status as an academic hospital, it serves a broad geographic region as one of five level-I trauma centers in the Commonwealth of Virginia. In this study we utilized 5 years of data from the University of Virginia Health System data warehouse. These data contain information on 91,297 patients and 196,902 unique encounters, including details on patient demographics, diagnoses and hospital departments visited. From these raw data we engineered features, trained gradient boosted decision trees, and experimented with unsupervised clustering techniques to best approximate 30-day Emergency Department revisit risk at the conclusion of each patient encounter. Our best model for revisit risk resulted in a Receiver Operator Characteristic Area Under the Curve of 0.75. Furthermore, we exhibit the real-time performance of our model as a tool to rank which at-risk patients should receive priority for Emergency Department resources. This test demonstrated a significant improvement over the current allocation of Emergency Department social worker resources with a daily Mean Average Precision of 0.83. The methodologies proposed in this paper exhibit an end-to-end framework to transform raw administrative claims and limited clinical data into predictive models that help the Emergency Department better manage resources and target interventions.

Speaker: Mike Thurber

Title: Causation Matters in Data Science

Abstract:

As technical and scientific professionals, we all know very well that correlation does not necessarily imply causation. Yet we continue to see published reports in popular media and even reputable professional journals where correlation is falsely portrayed as causation. The effect of this error is staggering: The public’s confidence in “scientific” reports continues to decline, bad decisions are repeatedly made on well-known popular but poorly executed studies. Data mining and data science get bad reviews because deployed models do not work as expected. Health and lives are compromised based on popularized information, information "documented" to be backed up by good science. Who is to be believed? And what can data scientists do, typically without prospective controlled experimental data do to identify and measure causal relationships with statistical confidence? This talk will address the topics of the real-world consequences of confusion about causation, and emerging sound data science methods for quantifying causal effects, including an example measuring some effects of c-section deliveries on newborn infants, based on publicly available CDC data of US births in 2012.

Looking forward to seeing everyone at the last Meetup of the year!

https://secure.meetupstatic.com/photos/event/2/6/f/4/600_466209972.jpeg

Related topics

Sponsors

NumFOCUS

NumFOCUS

Promoting open code for better science

You may also like