Premium
SU‐E‐T‐680: An Interactive Graphical User Interface for Physician‐Driven Treatment Plan Tuning
Author(s) -
Shi F,
Zarepisheh M,
Gautier Q,
Moore K,
Cervino L,
Jia X,
Jiang S
Publication year - 2013
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4815107
Subject(s) - computer science , graphical user interface , plan (archaeology) , touchpad , point (geometry) , user interface , interface (matter) , process (computing) , graphical display , medical physics , human–computer interaction , simulation , computer graphics (images) , medicine , mathematics , programming language , operating system , geometry , archaeology , bubble , maximum bubble pressure method , history
Purpose: To develop an interactive graphical user interface (GUI) for physicians to conveniently and efficiently fine‐tune an IMRT treatment plan. Methods: After an IMRT plan is developed either automatically or manually by a dosimetrist, the attending physician evaluates its quality and prefer to directly modify the DVH curves and the unsatisfactory part of dose distribution. We have developed an interactive GUI for this purpose. For the unsatisfactory part of a DVH curve, the physician can drag and drop it to where it is preferred using the computer mouse or touchpad. The treatment plan is then re‐optimized in near real‐time on the background GPUs to best match the physician's desire. The physician can also click a point in the dose distribution and drag it to modify the isodose lines to a more desired configuration. This mouse movement vector is propagated to its neighboring area to form a local vector field, which is used to modulate the underlying dose distribution. The new dose distribution is used to guide the plan re‐optimization in near real‐time. The re‐optimized DVHs and isodose lines are then displayed for the physicians to edit in the next iteration. This process is repeated until a physician satisfactory plan is achieved. Results: We have tested this GUI for a series of IMRT plans. Results indicate that the proposed method provides the physicians an intuitive and efficient graphical tool to edit the DVHs and dose distributions according to their preference. The input information is used to guide plan re‐optimization in near real‐time using our GPU optimization engine. Typically, a satisfactory plan can be developed by a physician in a few minutes using this tool. Conclusion: We have developed an interactive GUI for tuning IMRT treatment plans and demonstrated its feasibility through a series of clinical tests.