Automated elaborate resection planning for bone tumor surgery.

Purpose: Planning for bone tumor resection surgery is a technically demanding and time-consuming task, reliant on manual positioning of planar cuts in a virtual space. More elaborate cutting approaches may be possible through the use of surgical robots or patient-specific instruments; however, methods for preparing such a resection plan must be developed.

Methods: This work describes an automated approach for generating conformal bone tumor resection plans, where the resection geometry is defined by the convex hull of the tumor, and a focal point. The resection geometry is optimized using particle swarm, where the volume of healthy bone collaterally resected with the tumor is minimized. The approach was compared to manually prepared planar resection plans from an experienced surgeon for 20 tumor cases.

Results: It was found that algorithm-generated hull-type resections greatly reduced the volume of collaterally resected healthy bone. The hull-type resections resulted in statistically significant improvements compared to the manual approach (paired t test, p < 0.001).

Conclusions: The described approach has potential to improve patient outcomes by reducing the volume of healthy bone collaterally resected with the tumor and preserving nearby critical anatomy.