Differential evolution and cost-maps for needle path planning in Baker's cyst aspiration.

Abstract

PURPOSE The Baker's cysts appear within the popliteal fossa along with the progression of degenerative changes. Removal of its contents through aspiration is often a necessary complement to treatment at various stages of the development of gonarthritis. METHODS The paper presented a procedure for needle automatic needle path planning in cyst aspiration in transverse plane. The method was based on optimization and used a custom objective function, which utilized cost maps obtained from preprocessed, segmented images of the knee. The optimization was carried out with Differential Evolution. Furthermore, a preliminary sensitivity analysis was carried out. The obtained paths were compared to the reference paths proposed by an experienced surgeon. RESULTS The procedure was tested on 165 numerical simulations. In all of the obtained paths, the needle successfully avoided crucial objects, such as veins, arteries and nerves. Furthermore, the overall travel distance in the joint was also minimized. When compared to the reference from the surgeon, 90% of the paths were almost the same or only slightly different. Furthermore, the remaining 10% of the generated paths were viable but different. CONCLUSION Based on the obtained results, the proposed solution could be a viable solution for planning the aspiration of Baker's cyst.