Click here to

Session: Radiomics and Outcome Prediction [Return to Session]

Radiomics and Outcome Prediction

H Aerts1*, Y Xiao2*, F Yin3*, E Deutsch4*, (1) Dana-Farber/Brigham Womens Cancer Center, Boston, MA, (2) University of Pennsylvania, Philadelphia, PA, (3) Duke University, Chapel Hill, NC, (4) Universite Paris-saclay, Villejuif, FR


1:00 PM Artificial Intelligence in Cancer Imaging - H Aerts, Presenting Author
1:30 PM Standardizing Radiomics Applications for Clinical Trials - Y Xiao, Presenting Author
2:00 PM Impact of Image Quality on Radiomics Applications - F Yin, Presenting Author
2:30 PM State-of-the-art Multi-omics - E Deutsch, Presenting Author

TU-CD-TRACK 5-0 (Tuesday, 7/27/2021) 1:00 PM - 3:00 PM [Eastern Time (GMT-4)]

Rapid development in machine learning (ML) and radiomics in the last decade has stimulated substantial interest in precision medicine as well as research activities in applying radiomics and artificial intelligence (AI) for radiation therapy treatment assessment and outcome prediction. At the same time, the radiation oncology community is eager to understand its efficacy and limitation in order to narrow down the gap between scientific research and clinical application and prepare for emerging field evolution. This symposium aims to provide fundamental knowledge of radiomics and artificial-intelligence (AI) development in precision medicine, including open-source informatics. The impact of imaging parameters and image processing techniques on radiomics/ML as well as some general and specific guidelines recently recommended by a number of medical professional societies will be discussed to improve the robustness of using medical images for the outcome research. As an extension of radiomic application in multimodality cancer treatment, multiple omics applications in cancer treatment assessment and outcome prediction will also be introduced.

Learning Objectives:
1. To learn the role of radiologic AI with other –omics data for precision medicine and open-source informatics developments
2. To learn the current recommendations for incorporating radiomics in oncology clinical studies and the impact of imaging and image processing on radiomics applications
3. To learn the methodology of incorporating biological data into radiomics for the assessment of tumor heterogeneity


Feature Selection, Modeling, Decision Theory


TH- Response Assessment: Radiomics/texture/feature-based response assessment

Contact Email