Fig. 2. Morphology visualization of the partial occlusions using the proposed tool.

(A) Design of the ADS with the programmed magnetization profile. (B) Morphological visualization principle of partial occlusions. (C) Influence of the magnetization segment length l_m and minimum diameter of the channel inside occlusions d_{c_min} on visualization error E_avg in phantoms A to E. E_avg was the average of absolute deviations of the visualized morphology (red dotted lines) from the real boundaries. E_avg increased as l_m decreased and d_{c_min} increased. (D) Two typical cases of morphology visualization. The black lines represent ADS at different positions during retraction. The pulsatile flow rate Q into the inlet was set to around 10.0 ml/min. (E) Investigation on the influence of Q and the distance between the magnet and the lumen l_z on the maximum deformation height h_max of the ADS. To successfully visualize occlusions in a lumen, h_max should not be smaller than d_{c_min}. (F) Experimental and theoretical results of the investigation in phantom F. Error bars represent the standard deviation (SD) (N = 5). The retraction speed v_r of the tool was 1.0 mm/s in (C) and (F), and l_z was 50 mm in (C).