• artificial bladder
• artificial detrusor
• hydraulic actuation
• kirigami-based actuator
• soft robotics

Bladder cancer ranks ninth among the most frequently diagnosed cancers worldwide. Disease progression often necessitates radical cystectomy, i.e., complete surgical removal of the urinary bladder, with considerable impact on patients’ quality of life. Current clinical solutions involve either urinary diversion, redirecting urine through an external collection bag, or reconstruction of a neobladder using autologous tissue to replicate the storage function of the native bladder. However, these procedures do not provide patients with sensory feedback regarding bladder fullness, and urine drainage is frequently managed through intermittent catheterization, which often results in recurrent infections. Moreover, these approaches generally offer limited storage capacity of approximately 200 mL. For these reasons, the development of a fully implantable artificial urinary system capable of replicating bladder functions, namely urine storage, fullness sensation and voiding, is highly desirable.
In this context, the aim of this thesis is to develop a novel actuation strategy to be integrated into an existing origami-based artificial bladder, enabling efficient and on-demand voiding. The proposed solution employs a soft, hydraulic, kirigami-inspired actuation system to be integrated along the lateral walls of the artificial bladder. The kirigami-patterned actuator features a planar geometry with strategically arranged triangular subunits created through patterned cuts. Rectangular cross-section channels are embedded within the regions between the cuts: when vacuum is applied, these channels collapse, thereby generating the compressive force required to drive the micturition process of the artificial bladder. Conversely, when these channels are filled with fluid, the actuators relax allowing the artificial bladder to expand and fill with urine. The soft actuators are made of silicone material to ensure safe interaction with surrounding organs in the abdominal cavity.
Initially, the actuator design was defined through finite element (FEM) simulations in Ansys Mechanical to determine optimal geometric (e.g., number of subunits) and structural parameters (e.g., channel dimension). In particular, two geometries were evaluated, incorporating 4 and 12 cuts respectively, with channel widths of 4 mm and 6 mm and heights of 1 mm and 2 mm, yielding to eight distinct design configurations. FEM simulations were analyzed in terms of actuation volume and generated force to identify the optimal configuration. Each configuration was evaluated in different elastomeric materials (Ecoflex 00-50, Moldstar 15, Dragonskin 30, and SmoothSil 936) in ascending order of stiffness to optimize the performance while minimizing resistance to bladder expansion during filling.
The optimal combination of actuator design (i.e. 12 cuts, wall thicknesses of 1 mm, channel dimensions of 4 × 2 mm, and actuation fluid volume of 9.2 mL) and material (Moldstar 15 and Dragonskin 30) were subsequently fabricated and tested on bench. Molding fabrication was performed using a two-step process in which the silicone was poured into two separate molds corresponding to one of the lateral walls and the remaining structure with the embedded channels. After curing, the two layers were bounded together using additional silicone to form a single, integrated structure. This approach allowed preventing air entrapment within the mold, which could otherwise compromise the integrity of the thin-walled structure.
Two experimental phases were conducted: (i) actuator characterization and (ii) voiding performance upon integration on the artificial bladder. The actuator characterization involved the measurement of forces generated during depressurization of the channels using a load-cell setup enabling validation of the FEM simulation results. In particular, a mean absolute error of 0.186 N was showed between the FEM simulations and the experimental results. In addition, elastic and hydraulic contributions were assessed (90% and 10%, respectively, for an elongation of 20 mm). The complete actuation system, composed of four actuators positioned in pairs on the later walls of the artificial bladder, was then integrated and tested. The actuators performed well in both the best-case configuration, with the bladder urethra aligned horizontally, achieving a mean post-void residual volume (PVR) of 12.73 ± 5.59 mL, and the worst-case configuration, with the urethra elevated of 3 cm with respect to the bladder, resulting in a mean PVR of 31.98 ± 5.59 mL. In both cases, the system achieved the clinically relevant target of a PVR below 50 ml. Similarly, the voiding time was always in line with the state of the art of artificial detrusors for artificial bladders (120 ± 4 s and 120 ± 8 s, respectively).
Future investigation should focus on evaluating the durability of the actuation system under cyclic operation conditions and more realistic in-body condition to determine its suitability for long-term implantation. Moreover, integration with artificial sphincters and a sensing system for continuous monitoring of bladder fullness, along with miniaturized electronics for actuator fluid management, would enable the development of the first fully implantable artificial bladder with restored all physiological functions.