Click here to

Session: AI-Empowered Clinical Decision Support for Personalized Radiation Therapy [Return to Session]

AI-Empowered Clinical Decision Support for Personalized Radiation Therapy

C Shen1*, X Qi2*, Q Wu3*, J Deasy4*, J Deng5*, (1) University of Texas Southwestern Medical Center, Dallas, TX, (2) UCLA School of Medicine, Los Angeles, CA, (3) Duke University Medical Center, Chapel Hill, NC, (4) Memorial Sloan Kettering Cancer Center, New York, NY, (5) Yale University School of Medicine, New Haven, CT


WE-H-BRA-0 (Wednesday, 7/13/2022) 4:30 PM - 6:00 PM [Eastern Time (GMT-4)]

Ballroom A

A trove of patient data is generated routinely in radiation oncology clinics nationwide. These include the clinical data stored in the electronic medical record (EMR) system, the radiotherapy data generated in the treatment planning system (TPS) and recorded in the radiation oncology information system (ROIS), and the image data saved in the Picture Archiving and Communication System (PACS). With continuous accumulation of this multi-platform information in the clinical patient care, the ever-increasing patient data has made clinical decision-making highly challenging to the clinicians.

To achieve personalized radiotherapy, various challenging issues in the process of radiation treatments need to be accurately and efficiently accounted for, such as the time to achieve re-planning, the patient outcome trajectory, the accuracy of automatic contouring and dose prediction, the incorporation of radiobiology and radiomics into treatment optimization. Furthermore, the concept of digital twins of cancer patients has recently drawn a lot of attention as we are moving from personalized oncology into predictive oncology.

This session will discuss some of the most important and novel applications of artificial intelligence (AI) in clinical decision support for personalized radiotherapy in terms of real-time adaptation, outcome and response modeling, auto-segmentation and dose prediction, radiomics and radiobiological optimization, as well as digital twin modeling of cancer patients.

Learning Objectives:
1. Understand the challenges in achieving personalized radiotherapy for individual patients
2. Identify the opportunities of AI applications in supporting clinical decision-making
3. Understand the trend of AI applications for personalized radiotherapy

Funding Support, Disclosures, and Conflict of Interest: Research was supported by NIBIB, NSF, NCI, and DOE awarded to Jun Deng.



Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email