Comparison of two cable configurations in 3D printed steerable instruments for minimally invasive surgery.

In laparoscopy, a small incision size improves the surgical outcome but increases at the same time the rigidity of the instrument, with consequent impairment of the surgeon's maneuverability. Such reduction introduces new challenges, such as the loss of wrist articulation or the impossibility of overcoming obstacles. A possible approach is using multi-steerable cable-driven instruments fully mechanical actuated, which allow great maneuverability while keeping the wound small. In this work, we compared the usability of the two most promising cable configurations in 3D printed multi-steerable instruments: a parallel configuration with all cables running straight from the steerable shaft to the handle; and a multi configuration with straight cables in combination with helical cables. Twelve participants were divided into two groups and asked to orient the instrument shaft and randomly hit six targets following the instructions in a laparoscopic simulator. Each participant carried out four trials (two trials for each instrument) with 12 runs per trial. The average task performance time showed a significant decrease over the first trial for both configurations. The decrease was 48% for the parallel and 41% for the multi configuration. Improvement of task performance times reached a plateau in the second trial with both instruments. The participants filled out a TLX questionnaire after each trial. The questionnaire showed a lower burden score for the parallel compared to multi configuration (23% VS 30%). Even though the task performance time for both configurations was comparable, a final questionnaire showed that 10 out of 12 participants preferred the parallel configuration due to a more intuitive hand movement and the possibility of individually orienting the distal end of the steerable shaft.

1. Introduction

Laparoscopic surgery is a minimally invasive procedure in which several small incisions allow access to the human body by means of long and straight surgical tools. The reduction of the incision size reduces the post-operative pain and the recovery time for the patient, minimizes the scar tissue, thus obtaining better cosmetic results, and improves the cost-effectiveness of the procedure. Despite its great advantages, laparoscopy introduces new hurdles, e.g., due to the loss of wrist articulation and the introduction of a fulcrum effect [1,2]. Due to the pivoting point in the abdominal wall, the movement of the end-effector is inverted with respect to the handle. This so-called fulcrum effect results in a steeper learning curve. With the advent of new domains of minimally invasive surgery, such as single-port laparoscopy, transluminal, and intraluminal procedures, new challenges arise. For instance, accessing the target area becomes demanding when its optimal approach direction is not aligned with the rigid instrument shaft inserted through the incision [3].

Many robotic platforms have been proposed to overcome the limits in laparoscopy. One of the most famous platforms is the Da Vinci® robotic system offered by Intuitive Surgical Inc. (Intuitive Surgical Inc., Sunnyvale, Ca, USA) [4]. Robotic platforms give the surgeon additional degrees of freedom (DOF), three-dimensional visualization of the surgical site, and eliminate the fulcrum effect. However, they require a large footprint and high maintenance cost that makes the price-benefit ratio unfavorable for many procedures [5].

An alternative approach is the use of handheld mechanical solutions, in which the surgeon’s dexterity is enhanced by a steering mechanism with an additional two DOF close to the end-effector. Many research prototypes and commercialized instruments have been designed, and different solutions have been proposed to control the steerability of the end-effector [6,7]. The two most used control strategies in handheld instruments are wrist control, in which the movement of the wrist is used to steer the end-effector, such as found in the Laparo-Angle [8] or the LaparoFlex [9], and thumb control, in which the thumb controls the steering by means of a joystick [10], a trackball [11], or a steering wheel [12,13]. Comparative studies have been carried out on these two different control strategies to identify the most beneficial handheld control for the surgeon [14-16]. However, despite the 2-DOF steerable end-effector, the shaft rigidity of these instruments still restricts the surgeon’s workspace, limiting surgical use to procedures in which no obstacles need to be passed without being touched. To further improve maneuverability, mechanical solutions such as cable-driven mechanisms [17-19] or continuum concentric tubes [20] have been proposed to design a multi-steerable shaft enabling the surgeon to move along complex double-curved paths. Cable-driven solutions represent a valid alternative to robotic solutions due to their low maintenance cost, low noise, high sensitivity, and speed. Moreover, they directly react to the surgeon’s movements providing direct feedback and they enable simplification of the design without compromising the instrument functionalities. In cable-driven solutions, the cable control strategy plays an important role [21]. Cables can vary from a minimum of three for steering in two planes [22] to four or more as in the so-called cable-ring configuration [23], Fig 1A. In our group, we have explored two different cable control strategies for controlling cable-driven multi-steerable instruments: a control strategy based on cables straightly guided from the steerable shaft to the control handle (parallel configuration) [24], and a control strategy based on the combination of straight and helically cables placed around the backbone of the shaft and the control handle (multi configuration) [17], Fig 1B and 1C.

Fig 1

Multi-steerable strategies to control surgical instruments.

a) Cable-ring mechanism with its cross-section. Cables are placed concentrically to actuate the segments and guide each other along the shaft, adapted from [23]. b) Parallel configuration of a multi-steerable instrument, adapted from [25]. c) Multi configuration of a multi-steerable instrument, adapted from [17].

Whereas control strategy comparisons have been performed for 2-DOF instruments with only one steerable segment, a comparison in the steering and control of multi-steerable instruments with two or more segments has not yet been carried out. As a result, there is a lack of information about which way of controlling multi-steerable instruments is more convenient to the surgeon. In this study, we developed 3D printed multi-steerable instruments using parallel and multi configurations. The study aims to highlight and compare important weak and strong point of the new designs, that can help to improve the design of future multi-steerable laparoscopic instruments. For this purpose, using these instruments, we carried out an experiment with 12 participants to compare the two control strategies and identify which one has a steeper learning curve, faster task performance time, requires a lighter workload, and is preferred by the participants.

2. Cable configuration strategies

Cable-driven steerable instruments are controlled by actuation forces applied to their steering cables. In multi-steerable instruments, various deformation modes can be generated with different cable configurations, determining the behavior of the steerable shaft. In a 2D representation of one segment, we can define a generic steerable segment as an incompressible compliant backbone, with a length L, in which a rigid end plate of 2R in length is attached at the distal end, Fig 2. The proximal end of the backbone is fixed and represents the connection with the shaft. Actuation cables are attached at the outer ends of the end plate. In the case of a 2D symmetrical cable configuration, each segment can have two parallel cables (the parallel configuration), two diagonal cables, or the combination of diagonal and parallel cables (the multi configuration).

Fig 2

The three cable configurations presented in this work.

Left: 2D representation of the segment, center: Corresponding bending moment diagram, right: Segment deforming under the applied pulling force. a) Parallel configuration, b) diagonal configuration, c) multi configuration. For the explanation of the used symbols, see the text.

2.1 Parallel configuration

In the parallel configuration, cables are placed parallel to the backbone, and the pulling force Fp is parallel to the longitudinal axis of the segment, Fig 2A. Therefore, the bending moment is constant along the segment length L because the perpendicular distance R between the force application point and the backbone stays constant along the segment length L. The bending moment, therefore, defines the orientation angle of the segment (β), and the deflection mode will result in a curve with a constant bending radius. Segments with a parallel configuration can be combined by placing them on top of one another so that the base of the first segment acts as the top of the second segment and so on. The combination of the segment angles defines the position and the orientation of the end-effector, i.e., the end plate of the most distal segment, allowing different deformation modes.

2.2 Multi configuration

In the diagonal configuration (Fig 2B), cables connect the end plate to the fixed base by crossing each other forming the α angle fo Fig 2B. Forces Fd are applied along the cable direction and can be split into the Fx and Fy components. When the cables cross each other at L/2, the segment will have a symmetric bending moment and, therefore, symmetric behavior, enabling a double-curved shape deformation mode as shown in Fig 2B. In this case, the end plate translates laterally in the direction of the Fx force while the orientation of the end plate remains unchanged.

Full control with only one segment can be obtained by combining parallel and diagonal cables, Fig 2C. This combination, which will be referred to as the multi configuration [17], results in a mechanical behavior similar to the parallel configuration but with only one segment instead of multiple segments.

2.3 Three-dimensional representation

Navigation through confined anatomy requires instruments able to move in a 3D space. For instruments based on the parallel configuration, 3D motion can be achieved by using a minimum of three actuation cables per steerable segment. However, the use of four cables per segment concentrically placed at a 90 degrees angle allows antagonist movement of the cables and simplifies control [24]. The steerable segments are placed in series, one after the other, to increase the DOF of the shaft. The combination of a number of segments allows the control of the orientation as well as the position of the end-effector. In an instrument with multiple steerable segments, the actuation cables that control the end-effector run through dedicated slots of the preceding segments, the cables of the first preceding segment through the preceding ones, and so for all segments, Fig 3A. To avoid overlap of the actuation cables, each steerable segment of the shaft is rotated slightly, as shown in the close-up of Fig 4A.

Fig 3

3D model of the parallel and multi configurations.

a) The parallel cable configuration with a close-up on the shaft and the final design of the device. Each color corresponds to a segment mirrored on the control side. b) Multi configuration of the cables and the final design of the device. In this configuration, all cables are connected at the ends. The close-up shows the cable configuration in the steerable shaft.

Fig 4

Cross-section of the shaft for the parallel and the multi configurations.

a) In the parallel configuration all cables are equally distant from the central backbone Segments S are numbered from 1 to 6. In the multi configuration, cables are concentrically placed at three different radii to avoid overlapping. S represents the steerable shaft.

For instruments based on the multi configuration, 3D motion can be achieved by placing the parallel cables concentrically at a 90 degrees angle, similar to the parallel configuration but diagonal cables need a reconfiguration, Fig 3B. In fact, the diagonal cables will cross the backbone if positioned like in the 2D representation, and they will not allow any internal lumen to be included in the instrument. A possible solution, which was successfully investigated by Gerboni et al. [17], is to use helically-oriented cables that rotate 180 degrees around the central backbone [26]. Rotation of the helical cables can be either in the clockwise or in the counterclockwise direction. Using only one of the two directions would lead to an undesired torque along the segment backbone. Combining clockwise and counterclockwise helical cables in pairs cancels out this effect. In the parallel configuration, the parallel cables of different segments can be placed at the same distance from the backbone due to the straight nature of the cable slots. In the multi configuration, the helical cables would cross each other, causing overlapping of the cable slots. In order to avoid this arrangement, the three sets of cables (clockwise, counterclockwise, and parallel) are placed concentrically at three different radii as close as possible to each other, but still remaining independent by dedicated grooves Fig 4B. Differently from the parallel configuration in which we need multiple segments to determine the position of the end-effector, in the multi configuration, we can consider the steerable shaft as one long steerable segment due to the possibility of controlling both the position and orientation of the end effector with four actuation cables of each type (clockwise, counterclockwise, and parallel).

3. Instrument prototypes

3.1. Design

Two prototypes were designed with an identical outer appearance and size: one based on the parallel configuration and one on the multi configuration. Both prototypes contain three components: a compliant handle, a rigid shaft, and a compliant shaft. A detailed description of the compliant shaft and the design in the parallel configuration is described by Culmone et al. [24]. Both designs are based on a cable-driven actuation with a serial control strategy, in which the movements of the compliant handle and the one of the shaft are mirrored [21]. The compliant shaft is based on a modular compliant segment, composed of a central flexible backbone and four helicoids that run concentrically around the centerline, Fig 5. The helicoids have a T-shaped cross-section that is thin close to the backbone to ensure low bending stiffness and enlarges towards the outer side of the segment to limit the bending angle and prevent failure for excessive bending, Fig 5B. Segments with helicoids inversely placed around the backbone (clockwise and counterclockwise helicoids) are alternately placed on top of one another to form the compliant shaft and ensure equally divided torsion stiffness around the backbone. The actuation cables run through holes in the helicoids and are looped into a cross-shaped groove at the top of the steering segment to fix and control them independently, Fig 5C and 5D.

Fig 5

Steerable segment with parallel cables.

A) Four helicoids concentrically placed around the central backbone. b) Cross-Section of the steerable segments with the T-shape of the helicoids highlighted in green and cables in black. c) 3D model of the steerable segment with d) cross-section A-A showing the looped cables in the fixation point. Adapted from [24].

The rigid shaft is connected directly to the compliant shaft. The rigid shaft contains dedicated slots to guide the actuation cables from the compliant shaft to the handle. The compliant handle is based on using wrist control, in which all fingers and the wrist are used to manipulate the handle to define the desired shape of the compliant shaft. The design of the compliant handle is similar to a large version of the compliant shaft. For each compliant segment of the shaft, there is a respective segment for the handle. Similar to the compliant shaft, each segment of the handle has an inner backbone surrounded by an outer helicoid structure. The helicoid structure has the function of guiding the cables as well as creating a cable fixation point. The connection between the rigid shaft and the handle is smoothened by an amplification component, the rigid distal part of the handle that, with an amplification factor of three, guides the cables from the shaft to the handle, amplifying the movement between the handle and the shaft (Fig 3). Moreover, both designs have four lumens to insert flexible thin tools for diagnostic or treatment.

In the two prototypes, the steerable shaft shares the same design based on six steerable segments. However, in the parallel configuration, actuation cables control every single segment independently, and the cable fixation point needs to be located directly on the segment itself, in both the compliant shaft and the handle. In the multi configuration, the six segments are considered as one element. All actuation cables run through the entire body of the instrument and are fixed at its two ends: at the distal end, the end-effector, and at the proximal end, the end of the handle. Therefore, while in the parallel configuration, the position and orientation of the end-effector are controlled indirectly by controlling the orientation of the individual segments, in the multi configuration, the position and orientation of the end-effector are directly controlled as if there is only one segment. Moreover, different from the parallel configuration where cables run through straight guiding slots, in the multi configuration, the guiding slots for the cables are both on straight and helical tracks. The parallel configuration uses a serial control strategy based on mirrored movements. The segments of the shaft are mirrored in the handle, Fig 3A, resulting in the shaft moving opposite to the handle, i.e., the end of the handle moving upwards resulting in the end-effector moving downwards. Also, the multi configuration is controlled by mirrored movements: when a cable is pulled by bending the handle, it will shorten in its distal end, mirroring the handle movement. With the equal length in the steerable shaft, the two cable configurations are able to cover the same workspace and perform single curved and double curved shapes. Moreover, having an identical appearance for the two prototypes allows for comparing the cable configuration performance without influencing the participants at the test.

3.2. Fabrication

Both instrument prototypes were fabricated using vat photopolymerization as additive manufacturing technology, Fig 6. All parts were printed using Perfactory® Mini XL (EnvisionTEC GmbH, Gladbeck, Germany), with 25 μm layer height in the vertical z-axis. The printer, based on the so-called Digital Light Processing (DLP), uses a light source and a projector to harden the liquid resin layer by layer. We used R5 (EnvisionTEC GmbH, Gladbeck, Germany), an epoxy photopolymer resin, which is specifically customized for prototyping. The handle and the shaft were printed in the vertical position, with the main axis parallel to the z-axis of the printer. The shaft was printed without any support except for the raft, a layer between the built plate and the printed object. The handle was printed with a raft, and an internal support made of small pillars autogenerated by the printer software. After printing, the excess resin was removed by placing the parts in an isopropyl alcohol bath for 30 minutes. The raft and the support of the handle were manually removed. The dimensions of the printed devices slightly differ from the CAD model with a tolerance of ± 0.08mm for the shaft and ± 0.12 mm for the handle. A total of 24 ∅ 0.2 mm stainless-steel cables in the parallel configuration and 12 ∅ 0.2 mm stainless-steel cables in the multi configuration, were used to actuate the instruments. To avoid overlapping, the radii of the cable circle in the multi configuration were determined considering the printability of the instrument and the cable size. Once decided the amplification factor by fixing the diameter of the handle, the cables were placed as close as possible to each other but at the same time kept independent with dedicated grooves. Moreover, the order of the cables (clockwise, counterclockwise, and parallel) was decided considering the easiest way to mount the cables into the instrument. After placing them in the 3D printed construction, the cables were straightened using weights of 3 grams and then fixed in the handle by means of dog point set screws. At the end of the assembly, a small steel plate was glued at the end-effectors for use in the experiments.

Fig 6

Instrument prototypes employing the parallel configuration (top) and multi configuration (bottom). The close-ups show the steerable shafts. Notice that the fixation points in the shafts differ, depending on the configuration. In the parallel configuration, each segment has two fixation points whereas in the multi configuration all cables are fixed at the distal end of the shaft.

4. Functionality evaluation

4.1. Background and goal

A comparative evaluation was carried out to study the maneuverability of the instruments and investigate which cable configuration enables a faster and easier control strategy. We hypothesized that:

The parallel configuration requires less workload. The straight arrangement of the cables within the instrument generates less friction due to the lower normal forces between the actuation cables and the 3D helical printed structure as compared to the multi configuration.

The multi configuration would be faster in hitting the target, considering that all cables control the entire steerable shaft at once, whereas, in the parallel configuration, each steerable segment of the handle individually controls the corresponding steerable segment of the shaft.

The two instruments were tested in a laparoscopic simulator where targets with different orientations and positions were placed. The participants had to hit the indicated targets as fast as possible. The task was repeated 12 times (runs) per trial. Each participant attended four trials, two for each instrument. For each run, we measured the time to complete the task properly. We analyzed and compared the task performance time between the two instruments. Moreover, we examined the learning curve for each instrument and whether the order of use influences the learning curve. Finally, we analyzed the experienced workload and the individual preference of the participants using questionnaires.

4.2. Participants

Based on similar studies [13,27-29], a total of 12 participants (5 men and 7 women, aged 27.4±1.9) were recruited to take part in the experiment. All participants had no prior experience in laparoscopic or open surgery procedures, nor with laparoscopic instruments and were recruited within the BioMechanical department of Delft University of Technology (master students, PhDs, and technicians). Participants were all right handed with different videogames habits, between 0 to 10 hours a week. One participant played a musical instrument. The participants were split randomly into two groups, Group A and Group B, of 6 participants each. Each group had a different order of instrument use. Group A started with the instrument with the parallel configuration (PC), whereas Group B started with the instrument with multi configuration (MC). All participants were informed about the purpose, the type of the experiment, and the use of the collected data. The study was approved by the Human Research Ethics Committee at Delft University of Technology (ID:1408).

4.3. Experimental setup

The experiment was carried out using a laparoscopic simulator, specifically designed for this study. The simulator was designed to create, the movement that the surgeon might perform while navigating a laparoscopic procedure, in a simulated environment. Particular importance was given to the precision and orientation of the instrument, essential for preserving the surrounding critical areas. The simulator was made of clear PolyMethylMethAcrylate (PMMA) and PolyPropylene (PP) to replicate an inflated abdomen. We decided to have a transparent simulator to provide participants with direct 3D visualization. Due to their inexperience in laparoscopic surgery, using an endoscope and a monitor could have resulted in additional difficulties related to the loss of depth rather than the instrument maneuverability. Using 2D visualization of the target area, the collected data would not reflect the learning curve of the participants to properly operate the instruments, but rather their learning curve in visualizing the space in 3D from a 2D image. A silicon valve placed in the center of the simulator allowed the insertion of the instruments. A 3D printed cylindrical stand with seven targets in different orientations was placed inside the simulator. Six target tubes (20 mm long, ∅ 9 mm) were numbered and evenly placed around the stand, while a start flat target was placed center of the stand. The entrance point of the target tubes was marked with a black line. Each target tube contained two steel plates at its bottom, Fig 7E. When the end-effector was parallel oriented to the plates and therefore hit them simultaneously, electric contact was made, and a signal was measured by a Multifunction I/O device (USB-6008, National Instruments, Austin, USA) [30] that was controlled with a laptop via a LabView 2016 program (National Instruments, Austin, USA). A monitor showed the next target to be hit. The setup and the monitor were positioned in front of the participant, and their height could be adjusted to reach a comfortable position, Fig 7.

Fig 7

Experimental setup.

a) Setup and its components: 1. simulator, 2. instruments, 3. participant information letter, 4. general instruction, 5. informed consent and questionnaires, 6. user interface. b) A participant during the test. c) The user’s interface during the experiment. Each green circle represents a target. The light green circles are the targets already hit, and the dark green circles are the targets still to be hit. The status bar displays the next target that the participant has to hit. d) The instrument into one of the targets. d) A participant during the test. e) CAD model of the target with back view and cross-section. The two steel plates are represented in grey.

4.4. Task and procedure

The task consisted of positioning and orienting the multi-steerable shaft to reach the six targets. The experiment started when the participant hit the start target. Subsequently, the participant was asked to move the shaft towards the indicated target (randomly chosen among the six) and insert the tip into the tube. A low-frequency buzzer indicated that the participant hit the two steel plates of a wrong target, whereas a high-frequency buzzer indicated that the correct target was hit and the participant could move the shaft towards the new target. The time was recorded and was measured from the moment the participant hit the start target until the last target was hit. In each run, the participant hit the start target and the six other targets in a randomized order. Each trial consisted of 12 randomized runs. Each participant performed four trials: two trials for each of the two cable configuration instruments, allowing a comparison between the two groups over the two configurations and resulting in a total of 48 runs (12 runs x 2 trials x 2 cable configurations) per participant. The number of attempts per trial has been based on previous works where similar tests for steerable instruments have been performed [13]. Prior to the start, a short demonstration and an instruction sheet were given to the participant. The participant filled up an intake questionnaire with general information such as gender, age, educational phase, dominant hand, and video game or musical instruments experience. Before each of the four trials, the participants had two minutes to practice and familiarize themselves with the instrument. For participants of Group A, the experiment sequence was PC instrument followed by MC instrument, and again PC and MC. For Group B, the experiment sequence was MC-PC-MC-PC. In the supplementary materials, S1 Fig. shows the flow chart of the experiment and the two instruments order for the two groups.

At the end of each trial, the participant had a break of around 10 minutes to fill a self-evaluation questionnaire based on NASA’s Task Load Index (TLX) [31]. The six subscales (mental demand, physical demand, temporal demand, performance, effort, and frustration) of NASA TLX were rated from -10 to 10, in which a high score indicated that the task was highly demanding and a low score that was easy to perform. At the end of the fourth trial, the participant filled out a final questionnaire to express a preference between the two instruments, considering the ease of steering and control. All data were analyzed using Matlab R2020a scripts (accessible in the data availability). The S1 Video in the supplementary material shows the execution of one run for each instrument.

5. Results

Fig 8 shows the task performance time per instrument in the two trials. Yellow represents the PC, and blue the MC. The plot depicts the results as box and whiskers, where the bottom edge of the box indicates the 25th percentile, the top edge the 75th percentile, and the red central line the median. The median time for trial one was 102.05 s for the PC and 106.60 s for the MC. In Trial 2, the median time was 74.15 s and 76.75 s for both configurations, respectively. The median decreased for both instruments between the first and the second trial. Due to the asymmetry of the data calculated with the Shapiro-Wilk test (p<0.001), we performed the Mann-Whitney U test for independent groups of non-parametric data. The test revealed no significant difference (Z = -0.72, p = 0.44>0.05) on the task performance time of the two devices in each trial. Moreover, we compared the two trials of the same cable configurations. In both cases, the Wilcoxon Signed-Rank test for two dependent groups of non-parametric data showed a significant difference between the two trials (Z = 8.32, p<0.05), and therefore a significant reduction in time between the first and the second trial for both configurations as the participants got more experienced with the instruments after some training.

Fig 8

Box and whisker plots of the task performance time for the two cable configurations.

Yellow represents the parallel configuration (PC), and blue the multi configuration (MC). For each instrument, the participants performed two trials. The red line in the box represents the median and the red crosses, the outliers.

Looking at the trend of the runs within the trials, Fig 9 shows the learning curve of the participants per each instrument within the 12 performed runs of each of the four trials. The average time shows a reduction of 48% for the PC and 41% for the MC calculated as the difference of the average time between the first and the last run of the first trial. Data stabilized in the second trial for both instruments with an average time reduction of 24% for the parallel and 14% for the MC. The time performance for the PC and the MC in the last run of the second trial shows similar results: 76.42±19.87 s for the PC and 74.99±21.99s for the MC.

Fig 9

Box and whisker plots of the average time per run in the two trials performed by each participant for each instrument.

Yellow represents the parallel configuration (PC), and blue the multi configuration (MC). Each box and whisker plot represents the median, the upper and the bottom quartile of the average time for 12 participants. The outliers above 350 s have been cut off in the figure. The full picture can be found in the supplementary material.

The minimum task performance time average was 49.92±8.92 s and was achieved by 9 out of 12 participants in the last performed trial, Trial 2 with the MC for Group A, and Trial 2 with the PC for group B. Two participants of Group B achieved the minimum task performance using the MC; one participant in Trial 1 and one in Trial 2. In Group A, one participant achieved the minimum task performance time in Trial 2 with the PC.

The maximum task performance time, with an average of 272.15±124.85 s, was achieved in the first performed trial for 11 out of 12 participants, independently from the instrument. Only one participant of Group A achieved the maximum task performance time in Trial 1 with the MC.

Moreover, we looked at the influence of one instrument over the other, considering their order. Fig 10 shows the box and whisker plots of the task performance time for the 12 participants in the four trials for each run. We compared the task performance time of the first run of the two groups, A and B, in Trials 1 and 2 for the PC and MC, Fig 10. We performed the Mann-Whitney U test for independent groups of non-parametric data. The test revealed a significant difference (Z = 2.85, p<0.05) between Group A and Group B in Trial 1 for the PC. Group A (which started with the PC) required more time with an average time of 254.60±119.12 s than Group B (which started with the MC), which required 113.98±9.65 s for the same task in Trial 1. Also, for the MC there was a significant difference (Z = 2.43, p<0.05) between both groups in Trial 1. Group A required 124.30±13.98 s, which is less than Group B, which required 210.47± 84.53 s. Among the four trials, the learning curve of the participants shows a decrease in average task performance time. The curve dropped by more than 55% in task performance time between the average time of the first and the last run performed by all the 12 participants with for the very first instrument used, no matter which configuration, and flattened to a decrease of 3–7% for the very last instrument used in both groups.

Fig 10

Box and whisker plots of the average time per run in the two trials performed by each participant for each instrument in the two different groups.

Orange represents Group A and purple Group B. Each box and whisker plot represents the median, the upper, and the bottom quartile of the average time for the six participants of Group A and the six of Group B.

The responses of the TLX self-evaluation that ranged from -10 to 10 were transferred to a percentage scale. High percentages express a high workload, and low percentages express a low workload, i.e., -10 expresses 0% workload, whereas 10 expresses 100% workload. The overall Raw TLX score was 34% (SD = 22) for the parallel and 40% (SD = 23) for the MC in Trial 1. In Trial 2, the overall Raw TLX score was 23% (SD = 22) and 30% (SD = 23) for the parallel and the MC, respectively, Fig 11. We performed the Mann Whitney U test for independent groups of non-parametric data. The test revealed no significant difference (Z = -1.55 p = 0.12>0.05) between the overall workload in the first trials of the two instruments. In the second trial, the overall workload was significantly higher (Z = -2.18, p<0.05) for the MC compared to the PC. The Wilcoxon Signed-Rank test revealed a significant reduction (Zp = 4.92 Zm = 4.58, p<0.05) in the overall workload of the two instruments from Trial 1 to Trial 2. Participants expressed the maximum workload for both instruments in the effort subscale of Trial 1, 49% (SD = 21) for the PC and 57% (SD = 25) for the MC, respectively.

Fig 11

Average and standard deviation of the Raw TLX score for the six subscales (mental demand, physical demand, temporal demand, performance, effort, and frustration) in Trials 1 and 2.

The average was calculated over the score given by the 12 participants. Yellow represents the parallel, and blue the multi configuration.

Finally, Fig 12 shows the result of the final questionnaire on the subjective participant preference. The participants expressed a strong overall preference for the PC, 10 out of 12. All participants preferred the PC when considering the response in steering.

Fig 12

Results of the final questionnaire on personal preference.

6. Discussion

6.1. Experimental findings

The difference in task performance time was not significant when comparing the parallel and multi configurations to each other over Trials 1 and 2. A significant decrease appeared over time within the two trials when using the same configuration. This data was also confirmed by the learning curve of the two configurations. The two learning curves showed that the task performance time decreased quickly in Trial 1 during the first runs, and the participants reached a plateau after the first runs of Trial 2 for both instruments. It is interesting that for the parallel configuration, the minimum average task performance time was reached during Run 9 of Trial 2, instead of the last run, with a slight increase in average time for the subsequent runs. This effect is probably due to the tiredness of the participants at the end of the test. The flattening of the curves also showed its effect on the decrease in the workload perceived by the participants. The time performance for the parallel and the multi configurations in the last run of the second trial shows similar results for the parallel and the multi configuration, rejecting our second hypothesis.

The workload strongly decreased from Trial 1 to Trial 2 for both configurations. However, even though the task performance time did not show significant differences between the parallel and the multi configuration, the decrease in workload was significantly higher for the parallel configuration. This result can also explain the net difference in the preference of the parallel configuration over the multi configuration.

Looking at the alternation between the two cable configurations over the four trials, it becomes clear that the instruments influence each other over the first trials. In Trial 1, Group A started with the parallel configuration, and the average task performance time is significantly higher than the one in Trial 1 of Group B (which started with the multi configuration) for the same configuration. The same result can be observed for the opposite: the task performance time of Group B with the multi configuration in Trial 1 is significantly higher than the one of Group A. In the very first run, when they used their first instrument for the first time, the participants needed not only to learn to use the instrument and gain dexterity but also needed to familiarize themselves with the setup and the target positions. When they used their second instrument for the first time, they only needed to get used to the different cable configurations.

We also analyzed the performance of the participants within each run. An interesting outcome was the target that required the longest time to be hit and its occurrence within all 48 runs. The analysis revealed that Target 6 was the most difficult target to be hit, 195 times out of the total of 576 recorded runs. This can be explained by the location of Target 6, which was located the closest to the participant, requiring the instrument tip to be pointed towards the participant, mirroring its motion and thus adding an extra layer of difficulty in the maneuverability. The analysis becomes even more interesting when Target 6 is compared to Target 4. Target 4 has the same orientation angle but has an opposite location of Target 6. Target 4 recorded only 66 times the longest time to hit the target, the lowest occurrence among all targets. This result is probably due to its convenient location at the front right of the participant.

The net preference of the parallel over the multi configuration was briefly explained by four participants in the comments at the end of the final questionnaire and was mainly related to steering possibility. The parallel configuration gives the possibility of steering the segments independently, which is especially convenient for the most distal segment. By individually controlling the most distal segment, the participants felt more control over the final shaft orientation during insertion into the target. Therefore, the parallel configuration showed easier maneuverability over the multi configuration as hypnotized. On the other side, the multi configuration was noticed to be faster in reaching the initial position to hit the target, but less precise when aligned the tip to the target. Moreover, the multi configuration was preferred for the higher stiffness of the entire instrument, which allows for stronger haptic feedback of the steerable shaft during the test. The higher stiffness perceived by the participants was probably caused by the friction generated by the higher normal forces between the tensioned helical cables and the 3D printed helical structure in the handle as compared to the parallel cables. This observation is also interesting considering that the total number of cables in the multi configuration is half of the one in the parallel configuration. The different characteristics in speed, stiffness, and precision of the two cable configurations might open two different paths for the instruments. Whereas the multi configuration is used when the speed of the task is the main challenge, the parallel configuration is used when precision is fundamental to success in the task. It is also important to notice that all participants completed the test and no significant increase in the performance time was recorded for any of them at the end of the test. This consideration is important for the evaluation of the instrument maneuverability in view of possible future studies.

6.2. Limitation of this study and future recommendation

Additive manufacturing (AM) represents a significant innovation in terms of fast prototyping and the complexity of the design. In our work, AM allowed us to print highly complex compliant structures enabling advanced instrument maneuverability with very limited assembly time—the instruments were printed and assembled in less than one and a half days. Our study was mainly focused on device maneuverability and functionality. Therefore, the instruments were fabricated with an acrylic-based polymeric resin, which was non-biocompatible but specifically designed for easy and precise prototyping. Future work should focus on investigating the use of biocompatible materials able to guarantee the same compliant characteristics of the material used in this study. We think that our instruments should, in the end, be used as disposable devices, opening possibilities for the patient and surgeon-specific designs.

We used the same instrument for more than one participant, and, to always have fully functional instruments, in our experiment, we decided to use a new instrument every time we noticed signs of failing. All participants used the same two devices from the beginning to the end of the test (for all the 48 attempts) except for one participant for whom the multi configuration device needed to be replaced due to breakages on the end-effector side. Most of the time, the breakages were associated with excessive force applied by the participant to hit the target, and they were mainly on the end-effector side. Another reason for failure was due to the wear of the polymeric-based material induced by the stainless-steel cables again on the end-effector side. The stainless steel cables showed no signs of fatigue when straightened, however, especially during the first few attempts when the participant familiarized with the tool, excessive bending of the handle resulted in local bends.

The test was performed under direct 3D vision due to the inexperience of the participants with laparoscopic procedures. This choice was made as no experienced particpants were available, due to unforeseen limitations due to the global pandemic. Performing the test with an endoscope and a monitor would improve the resemblance of the task with the clinical setting. Moreover, it would be interesting to compare the performance and the preference of the novices with the ones of trained operators. The previous knowledge might affect it positively by making it faster in reaching the plateau of the learning curve, as shown in previous studies [28], or negatively affected it due to the mismatch in the movements to manipulate the instruments. The preference is expected to match the novice’s preference due to the similarities with the two DOF laparoscopic instruments currently used in the field. Another aspect that would be interesting to further investigate is the possible applications of our instruments by using the available lumens to insert flexible instruments to grasp tissues or perform biopsy procedures.

By comparing two cable configurations in 3D printed steerable instruments, this study explores new possibilities for additive manufacturing technology in medical instruments where complex geometries for the single parts simplify the overall design while maintaining, if not enhancing, the instrument’s functionalities.

7. Conclusion

The goal of this study was to compare parallel and multi cable configurations in multi-steerable laparoscopic instruments in terms of task performance time and workload. Our experiment showed that there was no significant difference in the task performance time for the two configurations. In the used NASA TLX scale, however, the participants expressed a lower workload for the parallel configuration as compared to the multi configuration. Overall, 10 out of 12 participants preferred the parallel configuration. The preference was mainly determined by the increased possibility of individually orienting the most distal segment.

Flow chart of the experiment for each participant.

Each trial consists of 12 runs and the order of the instruments used for the two groups. Parallel configuration (PC), multi configuration (MC).

(TIF)

Click here for additional data file.

Video of the execution of one run with the parallel configuration in strument and one run with the multi configutation instrument.

(MP4)

Click here for additional data file.

6 Jul 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Antonio Riveiro Rodríguez, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

"Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

4. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

5. Please include your full ethics statement in the ‘Methods’ section of your manuscript file. In your statement, please include the full name of the IRB or ethics committee who approved or waived your study, as well as whether or not you obtained informed written or verbal consent. If consent was waived for your study, please include this information in your statement as well.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This work is interesting and clearly shows the results of their research. However, I have some comments and doubts about this study:

1.- Why was a direct view used instead of a 2D view on a monitor? Although this increases the complexity of the trials, it is necessary to recreate all the conditions of the minimally invasive procedures to obtain an idea of ​​what is happening.

2.- Do the cables that attach to the links of the instrument suffer from any fatigue? How often did you change all the cables or the instrument? Every 3 or 4 attempts? Understanding that many articulating instruments suffer from fatigue or lack of tension in the cable arrangement after 10 attempts, it would be interesting to comment on this point of the investigation.

3.- Although no differences were shown between the two configurations, parallel and multi, would crossing the cables allow to cancel the fulcrum effect at the tip of the instrument controlled by the handle?

Reviewer #2: The authors compared two cable configurations for laparoscopy instruments fabricated via 3D printing. The topic is interesting, the paper is well-written and free from significant flaws. However, I suggest to accept it only after minor revisions. There are three points that can be improved:

1) In general, the authors should clearly indicate all the dimensions of the devices presented in the paper.

2) Section 3.2: it would be interesting to add more details on the 3D printing process. For example, the authors should provide the 3D models and the exact dimensions of the 3D printed parts that constitute the devices. How were these parts printed? Did the authors use printing supports? Did they postprocessed the 3D printed parts?

3) Section 4.2: can the authors comment on the choice of recruiting participants with no experience in laparoscopy? Is it expectable a difference between untrained and trained operators?

Reviewer #3: What is the size of the instruments? Scales (diameter, length, etc.) are missing in the figures.

For the multi configuration, how were the radii determined to avoid overlapping the cables?

Cable configuration in cable-driven flexible devices is of an issue in manipulation and control. However, there are many other scientific ways (some studied by same group of authors) to tackle the problem rather than user’s intuitive performance.

Motivations for the study is not well laid out. Why is this study needed, and where is the significance? Does the study intent to present superiority of one design over another?

What did this study aim to confirm or conclude?

Statistical analyses to decide the number of participants and number of tests are not provided. Without this, no conclusions can be drawn.

The learning curve could be observed with unexperienced participants. Clarification needed on how workload and learning curve were quantified.

It is intended that these tools will eventually be used in the hands of experienced surgeons, so what preference is shown by surgeons?

What criteria were taken into consideration designing the tests, repetitions, etc.

How was the data analyzed?

There was a preference for the parallel config because the distal joint could be oriented more reliably. There is no reason given for why this was important in the context of the experiment though...In the video it appears that the tip orientation needed to be very precise in order to activate the target- it was likely harder to align the tip with such precision by using the MC, but it appeared that the initial positioning to reach the target (before the fine-tuning motions) using the MC was faster. How does that align with real laparoscopic practices? Would the effect of a small lack of precision in the tip angle be negligible or drastic for surgeons?

Clarification needed on “amplification component”

Figure 8 is unnecessary

Figure 9: please choose more contrasting colors for your plots. It’s difficult to see the difference between PC and MC plots

Figure 10: Trial is incorrectly spelt

Fig. 2a: Orientation angle is mentioned, but not shown.

Fig. 2b: All used parameters in that figure should be explained (alpha is missed)

In center parts of Fig. 2, the vertical axis is not named.

Lines 193 to 208 discuss differences in control and segments, but the purpose/result is not clear.

Clinical application: The materials and manufacturing needs improvement and the device presented is only a prototype – this research is maybe at too early for user trials.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

2 Sep 2022

Dear Editor and Reviewers,

We would like to thank you for the efforts in reading our manuscript and providing useful feedback. We have carefully gone through the reviewers’ comments and created an improved version of the paper. We are glad to resubmit the new manuscript for evaluation, together with this accompanying letter answering in detail to each reviewer comment. All the modifications introduced in the revision process are highlighted in red in the text.

Kind regards,

The Authors

Reviewer 1

This work is interesting and clearly shows the results of their research. However, I have some comments and doubts about this study:

Comment 1

Why was a direct view used instead of a 2D view on a monitor? Although this increases the complexity of the trials, it is necessary to recreate all the conditions of the minimally invasive procedures to obtain an idea of what is happening.

Authors

We would like to thank the reviewer for this comment. We choose of using a transparent phantom to address the difficulties that users with no knowledge of laparoscopic procedures might encounter in properly visualizing the operating space through a 2D view on a monitor. Differently, by using laparoscopic view settings, the collected data would have reflected the learning curve of the participants in visualizing the space in 3D from a 2D image, more than the capacity to properly operate the device. 2D visualization would instead be recommended when testing the device with experts. For clarity, we added this consideration in Section 4.3 Experiment setup, lines 286-288.

Comment 2

Do the cables that attach to the links of the instrument suffer from any fatigue? How often did you change all the cables or the instrument? Every 3 or 4 attempts?

Understanding that many articulating instruments suffer from fatigue or lack of tension in the cable arrangement after 10 attempts, it would be interesting to comment on this point of the investigation.

Authors

We thank the reviewer for this comment. The cables were preliminarily tested before the experiments to check if they were suffering fatigue. The test did not show any particular fatigue when straightened. Once mounted on the device, cables were never changed during each individual test. Therefore, they were able to withstand a total of 48 attempts. We noticed that, due to the space between the helical elements of the handle, the cables showed possible local bends during the use of the instrument. We believe that the bending was due to the excessive force applied by the participant during the first attempts, while trying to familiarize themselves with the devices. When the instrument was used with the appropriate force, no bending or fatigue was noticed. All participants used the same device from the beginning to the end of the test, except for one participant for who damaged the device end-effector by applying excessive force. We have now added these details for clarity in the Discussion, lines 490-497.

Comment 3

Although no differences were shown between the two configurations, parallel and multi, would crossing the cables allow to cancel the fulcrum effect at the tip of the instrument controlled by the handle?

Authors

We would like to thank the reviewer for the comment. All cables for both the multi and the parallel configuration run from the tip's distal end to the proximal end of the handle. Rotating the cables 180 degrees in the shaft would let the tip move in the opposite direction, mirroring the current tip movement. However, it will not eliminate the fulcrum effect caused by the rigid shaft, pivoting in the abdominal wall.

Reviewer 2

The authors compared two cable configurations for laparoscopy instruments fabricated via 3D printing. The topic is interesting, the paper is well-written and free from significant flaws. However, I suggest to accept it only after minor revisions. There are three points that can be improved:

Comment 1

In general, the authors should clearly indicate all the dimensions of the devices presented in the paper.

Authors

We would like to thank the reviewer for this suggestion. We have now reported the dimensions of the devices in details in Figure 3.

Comment 2

Section 3.2: it would be interesting to add more details on the 3D printing process. For example, the authors should provide the 3D models and the exact dimensions of the 3D printed parts that constitute the devices. How were these parts printed? Did the authors use printing supports? Did they postprocessed the 3D printed parts?

Authors

We would like to thank the reviewer for this comment. The 3D model of the two devices is used in Figure 3 to present the parallel and multi configurations. We improved the caption to clarify and refer to them as the 3D model. Moreover, we implemented Section 3.2 Fabrication with an improved description of the printing process: “The shaft was printed without any support except for the raft. The handle instead, was printed with the raft, and the internal support was made of small pillars autogenerated by the printer software. After the printing, the excess resin was removed by placing the parts in an isopropyl alcohol bath for 30 minutes. The support of the handle and the raft were manually removed. The dimensions of the printed devices slightly differ from the CAD model with a tolerance of �  0.08mm for the shaft and �  0.12 mm for the handle”. The integration can be found on lines 227-232.

Comment 3

Section 4.2: can the authors comment on the choice of recruiting participants with no experience in laparoscopy? Is it expectable a difference between untrained and trained operators?

Authors

We would like to thank the reviewer for this comment. The experiments were carried out in March 2021 when restrictions due to the pandemic were still tight. For this reason, access to hospitals, and therefore recruiting subjects with experience in laparoscopy, was not possible. We recognize how comparing the performance between novices and experienced operators, would represents a valuable follow up study. Two different outcomes could be expected: compared to novice users, previous experience with different laparoscopic instruments (most of them with 2 degrees of freedom at the end effector) could constitute an initial obstacle, making the learning curve slower in the first trials, or could help them instead to reach faster a plateau in the learning curve, as shown in previous studies on articulating laparoscopic instruments “The trade-off between flexibility and maneuverability: task performance with articulating laparoscopic instruments”, Martinec et al., (2009). However, independently from previous experience, we would still expect a strong preference for the parallel configuration due to the similarities in orientation of the conventional steerable laparoscopic instruments. We elaborated on this point in Section 6.2, lines 499-506.

Reviewer 3

Comment 1

What is the size of the instruments? Scales (diameter, length, etc.) are missing in the figures.

Authors

We would like to thank the reviewer for this comment. We have now indicated in details the dimension of the device in Figure 3.

Comment 2

For the multi configuration, how were the radii determined to avoid overlapping the cables?

Authors

We thank the reviewer for this comment. The radii to avoid the overlap of the cables in the multi configuration were determined considering the printability of the instrument and the cable size. Once decided the amplification factor by fixing the diameter of the handle, the cables were placed as close as possible to each other but at the same time kept independent with dedicated grooves. Moreover, the order of the cables (clockwise, counterclockwise, and parallel) was decided considering the easiest way to mount the cables into the instrument. We improved this part in the Section 2.3, lines 156-157, and Section 3.2, lines 233-238.

Comment 3

Cable configuration in cable-driven flexible devices is of an issue in manipulation and control. However, there are many other scientific ways (some studied by same group of authors) to tackle the problem rather than user’s intuitive performance. Motivations for the study is not well laid out. Why is this study needed, and where is the significance?

Authors

We would like to thank the reviewer for this comment. Fully mechanical cable-driven instruments have the advantage to have low maintenance costs, they make no noise, they have high sensitivity, and high speed and they directly react to the surgeon’s movement giving the haptic feedback that is missing in many robotic solutions, but more importantly, they allow simplification of the design without compromising the instrument functionality. For this reason, cable-driven mechanisms still represent a valid alternative for laparoscopic instruments. We improved the Introduction in order to highlight this point on lines 66-69.

Comment 4

Does the study intent to present superiority of one design over another? What did this study aim to confirm or conclude?

Authors

We thank the reviewer for this comment. The study does not present the superiority of a design over another, but rather highlight and compare important weak and strong point of the new designs, that can help to improve the design of future multi-steerable laparoscopic instruments. For example, the performed tests showed a net preference for the parallel configuration over the multi configuration, proving that the independent orientation of the end-effector is a fundamental factor for subjects with no experience in steerable laparoscopic instruments. However, the multi configuration appeared to be faster in positioning the instrument to reach the target We improved the goal of the work in order to clarify the scope of the study, on lines 82-83 of the Introduction.

Comment 5

Statistical analyses to decide the number of participants and number of tests are not provided. Without this, no conclusions can be drawn.

Authors

We would like to thank the reviewer for this comment. We agree with the reviewer on the importance of statistical analysis to decide the number of participants. However, due to the difficulties in defining the target population and the number of instruments used in the field being the instrument new in such application, we decided to base the number of participants for this study on similar studies i.e., “SATA-LRS: A modular and novel steerable hand-held laparoscopic instrument platform for low-resource settings” Lenssen et al., (2022), “The trade-off between flexibility and maneuverability: task performance with articulating laparoscopic instruments”, Martinec et al., (2009), “Comparison of Laparoscopic Steerable InstrumentsPerformed by Expert Surgeons and Novices”, Lacitignola et al., (2020), “Precision in stitches: Radius Surgical System”, Waseda, (2007), where it varies between 5 and 24. In these studies either novices, experts, or both of the categories are recruited as participants of the study. Moreover, among the 12 participants of the studies, we tried to have differences such as age or gender and different habits that could affect the performances such as playing instruments or video games. For what concerns the number of tests, a similar consideration can be drawn. We decided to organize the test in four trials per participant so that each participant could test each instrument two times and the two Groups A and B could be comparable. Moreover, we fixed the number of repetitions to 12 so that the participants had the time to first familiarized themselves with the instrument and then improve his/ her performance. A similar number of attempts have also been used by similar studies such as SATA-LRS: A modular and novel steerable hand-held laparoscopic instrument platform for low-resource settings” Lenssen et al., 2022", where the study has been conducted with novices, whereas a lower number of attempts has been used in studies that involved also expert surgeons, such as “Comparison of precision and speed in laparoscopic and robot-assisted surgical task performance” Zihni et al. (2017). We included this consideration in Section 4.2 266, 268-271, and Section 4.3 lines 315-318.

Comment 6

The learning curve could be observed with unexperienced participants. Clarification needed on how workload and learning curve were quantified.

Authors

The learning curve was quantified for two different aspects. First, it was calculated as the difference between the average time used to perform the first run and the one to perform the last run for each instrument in the two different trials. Then the learning curve was calculated as the difference between the two groups over the four performed trials. The different drop in time is reported in percentage in the study. The workload was quantified using the NASA-TLX test which is generally used to quantify the workload of a task. The test is a self-evaluation that the participants had to fill out at the end of each trial. In the form, the participant had to mark on a scale from very low to very high how demanding the test was. The results have been evaluated by considering very low as 0 % and very high as 100 % workload. The results of the test for each trial have been plotted in Figure 12 (new Figure 11). The code used to analyze all data is available in the link of “data availability”. We improved the test to make these points clear in Section 5 lines 365-366, 388-390, and 399.

Comment 7

It is intended that these tools will eventually be used in the hands of experienced surgeons, so what preference is shown by surgeons?

Authors

We would like to thank the reviewer for this comment. Due to the fact that the proposed instruments have not been tested by experienced operators, we can only make some considerations for what concern the possible scenarios. Either the previous experience with different laparoscopic instruments, most of them with none or 2 degrees of freedom at the end effector, would constitute an initial obstacle and would make the learning curve slower than for the novices in the first trials, or their knowledge would help them and the learning curve would reach a plateau faster than for novices, as it has been shown in previous studies when a comparison with conventional and articulating laparoscopic instruments was conducted with experts and novices (The trade-off between flexibility and maneuverability: task performance with articulating laparoscopic instruments, Martinec, 2009). The authors would expect a similar outcome for what concerns the type of instruments with a strong preference for the parallel configuration due to the similarities in the orientation of the conventional steerable laparoscopic instruments. We elaborated on this point in Section 6.2, lines 501-506.

Comment 8

What criteria were taken into consideration designing the tests, repetitions, etc.

How was the data analyzed?

Authors

We would like to thank the reviewer for this comment. The test was designed considering that the steerable laparoscopic instruments compared to the conventional laparoscopic instruments allow more precise positioning of the instruments and orientation of the end-effector. Therefore, in the designed platform each tube has a different orientation and position that can be reached only with a specific shape of the end-effector. The repetitions of the trials were decided based on the fact that each instrument had to be used two times to have a comparison among the two groups and, in case, notice differences or influences of one instrument over the other. For each run, all the targets need to be randomly reached so that all runs could be comparable. Each trial has 12 runs so that the participants would during the first runs familiarize themselves with the instrument and then try to improve his/her performance. Moreover, a comparable number of attempts has been used in similar studies such as “SATA-LRS: A modular and novel steerable hand-held laparoscopic instrument platform for low-resource settings” Lenssen et al., 2022", where the study has been conducted with novices, whereas a lower number of attempts has been used in studies that involved also expert surgeons, such as “Comparison of precision and speed in laparoscopic and robot-assisted surgical task performance” Zihni et al. (2017). Data have been analyzed using the Matlab script available in the data availability of this article. We now added this consideration to clarify in Section 4.3 lines 315-318, and 336.

Comment 9

There was a preference for the parallel config because the distal joint could be oriented more reliably. There is no reason given for why this was important in the context of the experiment though...

Authors

We would like to thank the reviewer for this comment. During surgery precision and accuracy are extremely important to guarantee the success of the procedure. Having a fast and reliable orientation of the surgical instrument allow reaching the target while preserving critical surrounding areas. Therefore, in the experiment, we aimed at recreating in simulating environment the movement that the surgeon might perform when navigating during a laparoscopic procedure with particular importance to the precision in the orientation of the instrument. We improved Section 4.3 in lines 278-281.

Comment 10

In the video it appears that the tip orientation needed to be very precise in order to activate the target- it was likely harder to align the tip with such precision by using the MC, but it appeared that the initial positioning to reach the target (before the fine-tuning motions) using the MC was faster. How does that align with real laparoscopic practices? Would the effect of a small lack of precision in the tip angle be negligible or drastic for surgeons?

Authors

We would like to thank the reviewer for this comment. When using conventional laparoscopic instruments, the surgeons lose dexterity, and, therefore, precision becomes also a challenge. The use of steerable laparoscopic instruments allows the surgeon to be more precise and accurate, as shown in previous studies by Weseda et al. "Precision in stitches: Radius Surgical System". In our case, the lack in the precision of the MC would probably not be an issue for standard procedures, as it can be compared to commercialized steerable laparoscopic instruments, however, the high precision of the PC might allow procedures that are nowadays not possible due to such a lack of precision. We included this consideration in the Discussion, Section 6.1, lines 463-464 and 470-473.

Comment 11

Clarification needed on “amplification component”

Authors

We would like to thank the reviewer for this comment. We specified in the text that the “amplification is the distal rigid part of the handle (Section 3.1 line 191), and we improved Figure 3 so that also the amplification component is named in it.

Comment 12

Figure 8 is unnecessary

Authors

We would like to thank the reviewer for this comment. We moved Figure 8 to the supplementary materials.

Comment 13

Figure 9: please choose more contrasting colors for your plots. It’s difficult to see the difference between PC and MC plots

Authors

We would like to thank the reviewer for this comment. We changed the color that represents the parallel configuration from green to yellow in order to improve the contrast. All figures and test have been changed accordingly.

Comment 14

Figure 10: Trial is incorrectly spelt

Authors

We would like to thank the reviewer for this comment. We corrected the spelling of Trial in the previous Figure 11, now Figure 10.

Comment 15

Fig. 2a: Orientation angle is mentioned, but not shown.

Authors

We thank the reviewer for this comment. We included the orientation angle in Figure 2. We named it beta in the text and in the figure.

Comment 16

Fig. 2b: All used parameters in that figure should be explained (alpha is missed)

Authors

We thank the reviewer for this suggestion. We added the explanation of the alpha parameter in the text, line 114.

Comment 17

In center parts of Fig. 2, the vertical axis is not named.

Authors

We thank the reviewer for this comment. We added the name of the vertical axis in the central parts of Fig 2.

Comment 18

Lines 193 to 208 discuss differences in control and segments, but the purpose/result is not clear.

Authors

We would like to thank the reviewer for this consideration. We improved Section 3.1 (lines 215-218) by explaining that, due to the equal length of the steerable shaft, the two prototypes are able to cover the same workspace. Moreover, the identical appearance of the two prototypes allows for comparing the cable configuration performance without influencing the participant in the test.

Comment 19

Clinical application: The materials and manufacturing needs improvement and the device presented is only a prototype – this research is maybe at too early for user trials.

Authors

We would like to thank the reviewer for this comment. We agree with the reviewer about the early stage of such research to directly be used in a trial. In this work, we presented a first evaluation of the maneuverability of our 3D-printed steerable instruments. The presented test is fundamental, especially for inexperienced subjects to have an immediate impression of the possible use of such instruments. In case of complete failure of our studies with subjects that would have given up in the middle of the test or an increase of time during different trials would have suggested a possible need for modification of our approach in investigating 3D printed steerable instruments. On the contrary, the feedbacks were all positive and the subjects enjoyed participating in the training. We implement this consideration in the Discussion, Section 6.1, lines 473-476.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

19 Sep 2022

Comparison of two cable configurations in 3D printed steerable instruments for Minimally Invasive Surgery

PONE-D-22-13417R1

Dear Dr. Culmone,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Antonio Riveiro Rodríguez, PhD

Academic Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: I consider that the authors have made the necessary changes to publish the document; therefore, my decision is to accept the article.

Reviewer #2: The authors satisfactorily replied to all my observations. In my opinion, the paper can now be published in its current form.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

**********

26 Sep 2022

PONE-D-22-13417R1

Comparison of two cable configurations in 3D printed steerable instruments for Minimally Invasive Surgery

Dear Dr. Culmone:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Antonio Riveiro Rodríguez

Academic Editor

PLOS ONE