Patient geometry-driven information retrieval for IMRT treatment plan quality control.

Area of Science:

• Medical Physics
• Radiation Oncology
• Radiotherapy Planning

Background:

• Intensity modulated radiation therapy (IMRT) plan quality is subjective, relying on planner experience and time.
• Planners may overlook opportunities for further sparing of organs at risk (OARs).
• A robust quality control (QC) method is needed to objectively assess IMRT plan quality.

Purpose of the Study:

• To propose and evaluate a novel QC method for IMRT treatment plans.
• To assist planners in evaluating OAR doses and identifying potential for further sparing.
• To enhance the consistency and quality of IMRT treatment planning.

Main Methods:

• Developed a QC method comparing geometric OAR-target configurations of new plans to a database of prior patients.
• Introduced the overlap volume histogram (OVH) as a shape relationship descriptor.
• Used OVH to infer likely DVHs and identify related patients for dose comparison.

Main Results:

• Applied to 32 head-and-neck patients' parotid DVHs, identifying 17 parotids for potential dose reduction.
• Replanning achieved average D50 reductions of 6.6 Gy for flagged parotids and 1.9 Gy for controls.
• The method accurately identified parotids needing dose reduction and those with marginal sparing improvements.

Conclusions:

• The proposed method serves as an effective QC mechanism for evaluating OAR DVHs.
• Adoption of this QC method can significantly advance IMRT planning quality.
• This approach promotes better treatment plan consistency and potentially improved patient outcomes.