David Ledbetter, data scientist at the Children’s Hospital Los Angeles, shares how his team is using TITAN X GPUs and deep learning to help provide better recommendations of drug treatments for children in their pediatric intensive care unit.
To train their models, 13,000 patient snapshots were created from ten years of electronic health records at the hospital to understand the interactions between a patient’s vital state, heart rate, blood pressure and the treatments they were given. By understanding the most important relationships in the data, they are then able to generate the probability of survival predictions for the patients moving forward as well as physiology predictions in order to simulate augmented treatments.
David presented his research poster “Dr. TED: Deep Learning Recommendation of Treatment from Electronic Data” at the 2016 GPU Technology Conference.
Share your GPU-accelerated science with us at http://nvda.ly/Vpjxr and with the world on #ShareYourScience.
Watch more scientists and researchers share how accelerated computing is benefiting their work at http://nvda.ly/X7WpH
Share Your Science: Leveraging Deep Learning for Personalized Drug Treatment Recommendations
Apr 15, 2016
Discuss (0)
AI-Generated Summary
- David Ledbetter and his team at the Children's Hospital Los Angeles used deep learning and NVIDIA TITAN X GPUs to improve drug treatment recommendations for children in their pediatric intensive care unit.
- The team created 13,000 patient snapshots from ten years of electronic health records to understand the relationships between a patient's vital state, heart rate, blood pressure, and treatments given.
- By analyzing these relationships, they generated probability of survival predictions and physiology predictions to simulate augmented treatments.
AI-generated content may summarize information incompletely. Verify important information. Learn more
About the Authors
