Events Add an event Speakers Talks Collections
 
Duration 03:06
16+
Video

A11 Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes

Justin Lovelace
Graduate Research Assistant at Carnegie Mellon University
+ 3 speakers
  • Video
  • Table of contents
  • Video
Machine Learning for Healthcare
August 7, 2020, Online, Los Angeles, CA, USA
Machine Learning for Healthcare
Request Q&A
Machine Learning for Healthcare
From the conference
Machine Learning for Healthcare
Request Q&A
Video
A11 Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Add to favorites
101
I like 0
I dislike 0
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
  • Description
  • Transcript
  • Discussion

About the talk

Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes with Guided Multi-Headed Attention

About speakers

Justin Lovelace
Graduate Research Assistant at Carnegie Mellon University
Nathan Hurley
Student Staff at Texas A&M University
Adrian Haimovich
Resident Physician at Yale New Haven Hospital
Bobak Mortazavi
Assistant Professor at Texas A&M University

I am an undergraduate honors student majoring in Computer Science and minoring in Mathematics in Texas A&M University. I started working in the lab towards the end of my sophomore year and completed my honors thesis under the advisement of Dr. Bobak Mortazavi during my junior year. My research has focused on the application of natural language processing techniques to the clinical text stored in electronic health records.

View the profile

I am an MD/PhD student studying Computer Science in the lab of Professor Bobak Mortazavi. My research interests are in machine learning and its applications to electronic health records. I started medical school with the Texas A&M University Medical School class of 2018, and I completed my first three years of medical school with that class. I then began working on my PhD, and plan to complete the fourth year of medical school after finishing my PhD.

View the profile

Adrian Haimovich graduated magna cum laude from Columbia University with a degree in Applied Mathematics. He subsequently joined the MD/PhD program at Yale University, where he completed his dissertation in synthetic biology in the lab of Dr Farren Isaacs

View the profile

I am an Assistant Professor in Computer Science & Engineering at Texas A&M University, and a member of the Center for Remote Health Technologies & Systems. Prior to joining Texas A&M, I was a postdoctoral associate under the supervision of Prof. Harlan Krumholz at the Center for Outcomes Research and Evaluation (CORE) and Prof. Sahand Negahban of the Department of Statistics. I worked as a graduate student (and earned my Ph.D.) in Computer Science under the supervision of Prof. Majid Sarrafzadeh at the University of California Los Angeles (UCLA) Wireless Health Institute (2014). I earned a B.A. in Applied Mathematics and a B.S. in Electrical Engineering and Computer Science from the Univeristy of California Berkeley (2007).

View the profile
Share

Hello, everyone. I'm just going to let you know, I'll be presenting our work titled dynamically. Extracting Alcon, specific problem is from Clinical notes, with God and multi-headed attention. This work was conducted in collaboration. My co-authors, I can barely talk to Adrian. How to glitch, Dr. Vobach more coffee aisle. First begin by discussing your motivation for work promise or an important component of the electronic health record. The documentation illnesses and injuries is losartan, and provide clinicians with an overview of a patient's medical condition. However, I

can make them ineffective and patient outcomes order for conduct our work on the publicly available next three days in this work racetrack problems. From all the free text message with a patient for his or her discharge made out to you. Probation outcomes. We predict to, Lee started adverse outcomes. Mission. In mortality we conduct experiments for two. Definitions for each outcome. Sounds are green Mission Inn in hospital. Mortality after a patient has left the ICU but before they leave the hospital 30-day readmission in

immortality assembly consider all adverse outcomes that occur within 30 days of discharge. When I went to use the frame of the redeveloped to extract the nearest promise, we develop in, in the in framework, that first utilizes a convolutional model with her problem, attention mechanism to start problems from the clinical narrative. Remember to the ICD codes associated with each patient stated to find the problem labels for each patient. We didn't predict the final outcome as a linear combination of the extracted problems. For Simplicity of his. Final prediction allows us to identify

which ones are important for the prediction inductive out Dynamic problem. On Alpha Centauri experimental results. We find that our proposed framework performs competitively with bass lines that directly predict the outcome of Interest using multiple different ICD codes arrive, considerations for the intermediate problem. This is proof immortality outcomes of the Rosary admission outcomes explored in this work. You're at we share an example of an extracted promise for a patient at risk for bouncer Creed Mission green. Observe that the higher probability for fluid disorders and puncture of

vessel Drive their risk prediction. For this patient looking at an example for a low-risk patient, we observe that it correctly identified medical characteristics such as their softball disease in the associated procedure. But these conditions near zero problem with enough or not associate with negative outcomes. We also conducted at user study or for medical expert. Rated-R Frameworks extracted problem with and found and they offered a statistically significant improvement over both the Baseline, tension model and a neutral rating. Their number of directions that we intend to pursue

for future work would like to extend our framework to other outcomes of Interest. Such a steps of the onset of intubation. We'd also like to continue to explore alternative problem representations to see if that can potentially improve our framework. We also intend to improve upon our problem, extraction architecture in hopes of improving the Performance programmer. Thank you for listening.

Cackle comments for the website

Buy this talk

Access to the talk “A11 Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free

Ticket

Get access to all videos “Machine Learning for Healthcare”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Ticket

Interested in topic “Artificial Intelligence and Machine Learning”?

You might be interested in videos from this event

February 4 - 5, 2021
Online
26
115
ai, application, bot, chatbot, conversation, data, design, healthcare, ml

Similar talks

Ruijun Chen
Assistant Professor in Translational Data Science and Informatics at Geisinger
+ 8 speakers
Victor Rodriguez
MD-PhD Student at Columbia University
+ 8 speakers
Lisa Liu
PhD Candidate at Columbia Department of Biomedical Informatics
+ 8 speakers
Elliot Mitchell
Graduate Research Assistant at Columbia University Irving Medical Center
+ 8 speakers
Amelia Averitt
Manager, Clinical Informatics at Regeneron Pharmaceuticals
+ 8 speakers
Oliver Bear Don't Walk IV
PHD Student at Columbia University Graduate School of Arts and Sciences
+ 8 speakers
Shreyas Bhave
Software Engineering Intern at Counsyl
+ 8 speakers
Tony Sun
PhD Student at Columbia University in the City of New York
+ 8 speakers
Phyllis Thangaraj
Resident Physician at Yale New Haven Hospital
+ 8 speakers
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
John Morrison
Assistant Professor of Pediatrics at The Johns Hopkins University School of Medicine
+ 6 speakers
Ali Jalali
Senior Data Scientist at Biofourmis
+ 6 speakers
Hannah Lonsdale
Clinical Research Associate at The Johns Hopkins University
+ 6 speakers
Paola Dees
Physician at Johns Hopkins All Children's Hospital
+ 6 speakers
Brittany Casey
Pediatric Hospital Medicine Fellow at Johns Hopkins All Children’s Hospital
+ 6 speakers
Mohamed Rehman
Eric Kobren Professor of Applied Health Informatics at The Johns Hopkins University School of Medicine
+ 6 speakers
Luis Ahumada
ДолжностьDirector Center for Pediatric Data Science and Analytic Methodology at Johns Hopkins All Children's Hospital
+ 6 speakers
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Stephanie Skove
Bachelor of Science at Duke University
+ 12 speakers
Harvey Shi
Associate in Research at Duke Institute for Health Innovation (DIHI)
+ 12 speakers
Ziyuan Shen
Software Engineer at ZipRecruiter
+ 12 speakers
Michael Gao
Data Science Lead at Duke Institute for Health Innovation
+ 12 speakers
Mengxuan Cui
Statistical Modeling Engineer at Yidu Tech Inc.
+ 12 speakers
Marshall Nichols
Senior IT & Analytics Manager at Duke Institute for Health Innovation
+ 12 speakers
Suresh Balu
Program Director at Duke School of Medicine
+ 12 speakers
Cara O’Brien
Assistant Professor at Medicine at Duke University
+ 12 speakers
Armando Bedoya
Associate Chief Medical Informatics Officer, Duke University Health System at Duke Health Technology Solutions
+ 12 speakers
Dustin Tart
Duke University at Program Manager, Patient Response
+ 12 speakers
Benjamin Goldstein
Associate Professor at Duke University
+ 12 speakers
Mark Sendak
Population Health & Data Science Lead at Duke Institute for Health Innovation
+ 12 speakers
William Ratliff
Innovation Program Manager at Duke University Health System
+ 12 speakers
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free

Buy this video

Video
Access to the talk “A11 Dynamically Extracting Outcome-Specific Problem Lists from Clinical Notes”
Available
In cart
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free
Free

Conference Cast

With ConferenceCast.tv, you get access to our library of the world's best conference talks.

Conference Cast
947 conferences
37617 speakers
14378 hours of content
Justin Lovelace
Nathan Hurley
Adrian Haimovich
Bobak Mortazavi