Fig 1.
Scala Tympani (ST) model generation.
Generation of the ST model follows several steps to preserve the common anatomical features of the cross-sectional geometry [51], as this defines the contact area between electrode and phantom. Cross sections can vary in size and orientation between individuals (a,b) [3, 34] and were rotated to the cochlear angle dependent average orientation of the basilar membrane (c). Then manual segmentations points were redistributed evenly and consistent for all cross sections of the n = 15 datasets and datapoints were averaged at the respective angular location (d).
Fig 2.
Phantom orientation and test setup.
For the orientation of the ST model (a) the CCS [52] is adopted with the y-axis being the insertion axis. The model is transferred to the physical insertion phantom (b) and an idealized cochlea opening is created. The model is placed in the test setup (c) for CI electrode array insertion. To control the boundary conditions and create a steady state during insertion, the electrode array is inserted through a guide tube (d) into the phantom. Electrode insertion starts at EID4 = 4mm = ystart (e).
Table 1.
Parametrization of mean insertion phantom.
The parameters AST (basal ST diameter, measured from the center of the round window through the modiolus to the opposite wall of the ST) and BST (basal turn width of the ST orthogonal to AST) describe the dimensions of the basal turn of the mean ST. HST (distance from the lowest to the highest point f the ST lumen along z) describes its overall height and HS, ST the height of its lateral wall (LW) spiral [49]. The Cochlear duct lengths CDLLW,ST and CDLST are measured along the LW of the phantom from the center of the round window to the most apical point and express the corresponding length as metric and angular length respectively [55]. The parameter “Area cochleostomy” describes the surface area of the basal ST phantom opening for CI array insertion.
Fig 3.
Insertion force profile of the three FLEX28 electrodes.
Three electrodes in (a)-(c); v = 0.5 mm/s, lubrication 90% soap solution. Qualitatively only the very first insertion of each electrode appears to differ from the subsequent and last (ins. 21) insertion. Differences between the electrodes appear to become negligible after the conditioning cycle (d).
Table 2.
Insertion work for different electrodes.
Insertion work W for the first insertion (Ins. 1) of each one of the three electrodes El. 1 to El. 3 is 28% higher than all subsequent ones (Ins. 2, 3 and 21). Only minor changes in work occur afterwards.
Fig 4.
Mean electrode insertion forces.
Forces averaged over all electrodes and each of the three insertions per cycle (n = 9 insertions). Maximum forces differ by insertion speed and concentration of the lubricant. Note that a snap in the force profile can be observed in all conditions.
Table 3.
Mean maximum insertion forces by speed and lubrication.
Maximal forces increase with decreasing soap solution of lubricant and decreasing insertion velocity. N = 9 insertions.
Fig 5.
Snap is quantified by (1). Lower insertion speeds result in smoother force profiles and smaller snap. Values for larger insertion speeds v = 1 mm/s and v = 2 mm/s increase towards the end of insertion. The effect is more pronounced the lower the concentration of the lubricant. At 25–28 mm, the RMS of the snap increases significantly with increasing speed for all neighboring values (see S3 Appendix. Statistical Analysis. Table 9).
Fig 6.
Results for influence of insertion speed and lubrication onto insertion behavior.
(a) for higher lubrications of 50% and 90% soap solution Fmax decreases with increasing insertion velocity. (b) Insertion work W shows almost constant values (10%), decreasing work (50%), and increasing work (90%) for rising insertion speeds depending on the concentration of the lubricant. (c) Snap, representatively only shown for 25–28 mm EID, is more pronounced for the lower concentrated lubrication and large insertion speeds.
Fig 7.
Different states of friction during CI electrode array insertion.
(a) In boundary friction the lubricating film is collapsed or only of molecular thickness. (b) In mixed friction the surface asperities of the array and ST insertion phantom are indirect contact. (c) In hydrodynamic friction array and ST phantom surface are completely separated by the lubricant. Due to different local normal forces FN [22], different frictional states can occur along the array simultaneously [47]. The effect is further driven by insertion speed and concentration of the lubricant.