Program Information

Big Data 1: Current Big Data Resources and Technology in Radiation Oncology

S Benedict¹, T McNutt², L Xing³, E Roelofs⁴, (1) UC Davis Cancer Center, Sacramento, CA, (2) Johns Hopkins University, Severna Park, MD, (3) Stanford University School of Medicine, Stanford, CA, (4) Maastricht Radiation Oncology, Maastricht, Zuid-Limburg, NL

Presentations

7:30 AM : Introduction to Big data in Radiation Oncology - S Benedict, Presenting Author
7:45 AM : Experience building a learning health system and decision support in radiation oncology - T McNutt, Presenting Author
8:00 AM : Learning and decision-making from BIG Data - L Xing, Presenting Author
8:15 AM : FAIR data-sharing: federated learning in healthcare - E Roelofs, Presenting Author

TU-A-605-0 (Tuesday, August 1, 2017) 7:30 AM - 8:30 AM Room: 605

In Part 1 of this Big Data Panel we present current resources and technology employed in radiation oncology. A primary component of a learning health system is data capture in the clinical setting, which is paramount to the success. Methods of collection and aggregation of treatment data, physician assessments, and patient reported outcomes will be presented with future needs to better support it. Relational database design for radiation therapy data with the intent of analysis and data mining will also be presented. Novel decision support models, and how they may be used clinically will be presented in terms of causality, predictive power, and strategies to interpret the results.

The session will conclude with an overview of the challenges in pooling data from multiple centers. By starting with proper local data governance and using a new global, distributed learning network and introduce data-driven medicine, an example of model-based decision support is presented.

Learning Objectives:
1. To understand strategies for data capture in the clinical setting, including methods of collection and aggregations of treatment data, physician assessments, and patient outcomes.
2. To become familiar with decision support models, and how they may be used clinically in terms of causality, predictive power, and strategies to interpret the results
3. To understand the commonly used strategies and machine learning techniques for big data analysis and for meaningful decision-making.

Handouts

127-35411-418554-125539.pdf (L Xing)

127-35412-418554-126875.pdf (E Roelofs)

127-38277-418554-126654.pdf (T McNutt)

Contact Email: