Systems and methods for automated radiation treatment planning with decision support

Unmet Need

Cancer is a leading cause of death worldwide, with almost 10 million deaths and over 19 million new cases in 2020. These numbers are expected to grow, with 28 million new cases estimated to be diagnosed in 2040. Some of the most effective therapies for cancer are based around beams of radiation applied by an external source, such as intensity-modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT). External radiation therapy is an almost $6B a year market, which is expected to grow at a CAGR of about 6.3% to almost $8B by 2023. However, these types of radiation treatments have to be planned for each individual patient so that the beams deliver the maximum dose to the target tumor while sparing nearby healthy tissue and organs at risk (OAR), a time-consuming process managed by a team of clinical professionals. These professionals can use software solutions to more efficiently plan radiation treatment, but currently available radiation treatment planning software is usually limited to a specific and more simple cancer site and generates only one plan. There is a need for efficient radiation treatment planning software which creates multiple plans and is generalizable to various cancer sites and types.

Technology

Duke inventors and colleagues have developed software systems and methods that intake site- and type-agnostic cancer patient data and automatically design multiple radiation treatment plans based on similar cases, followed by decision support for plan selection. By using dataset partitioning to choose model training data similar to the patient case and geometric information and case-based reasoning for selecting models and determining beam angles, IMRT or VMAT plans with multi-criteria optimization for OAR sparing and target dose maximization can be automatically generated for final selection by the radiation treatment team. These system and methods have been tested on retrospective patient cases and subsequently derived applications have been applied clinically at the Duke health system.

Advantages

  • More flexible than currently available planning software, applicable to many different types of patient cases, including different cancers and outliers
  • Faster plan design, on the order of several minutes or quicker, allowing for more approaches to be analyzed more efficiently
  • Delivers clinically relevant plans with same or better performance than human-generated plans