Using a Digital Twin of an Electrical Stimulation Device to Monitor and Control the Electrical Stimulation of Cells in vitro

Abstract

Electrical stimulation for application in tissue engineering and regenerative medicine has received increasing attention in recent years. A variety of stimulation methods, waveforms and amplitudes have been studied. However, a clear choice of optimal stimulation parameters is still not available and is complicated by ambiguous reporting standards. In order to understand underlying cellular mechanisms affected by the electrical stimulation, the knowledge of the actual prevailing field strength or current density is required. Here, we present a comprehensive digital representation, a digital twin, of a basic electrical stimulation device for the electrical stimulation of cells in vitro. The effect of electrochemical processes at the electrode surface was experimentally characterised and integrated into a numerical model of the electrical stimulation. Uncertainty quantification techniques were used to identify the influence of model uncertainties on relevant observables. Different stimulation protocols were compared and it was assessed if the information contained in the monitored stimulation pulses could be related to the stimulation model. We found that our approach permits to model and simulate the recorded rectangular waveforms such that local electric field strengths become accessible. Moreover, we could predict stimulation voltages and currents reliably. This enabled us to define a controlled stimulation setting and to identify significant temperature changes of the cell culture in the monitored voltage data. Eventually, we give an outlook on how the presented methods can be applied in more complex situations such as the stimulation of hydrogels or tissue in vivo.

Pulsed Electrical Stimulation Affects Osteoblast Adhesion and Calcium Ion Signaling

“…The current through the medium can be considered to validate the simulation and hence the correctness of the estimated field [ 23 ]. We measured a current of 12 mA at 1V input amplitude.…”

“…The applied electric field strength was evaluated by employing numerical simulations. Under the assumption that the dielectric parameters of the cell culture medium are not frequency-dependent, the electrostatic Laplace equation is solved [ 23 ]. This equation reads …”

Bioelectric Potential in Next-Generation Organoids: Electrical Stimulation to Enhance 3D Structures of the Central Nervous System

“…In addition, computational modelling of EFs in vitro will be required, specifically to understand electrical current distribution in the 3D organoid environment and with varying media compositions. ( Zimmermann et al, 2021 ). Ultimately, the choice of electrode material and electrochemical characterisation of the set up will be critical to determine a biologically safe stimulation regime, which delivers the required amount of charge without splitting the water or producing toxic faradaic by-products ( Boehler et al, 2020 ; Mobini et al, 2022 ).…”

“…Thirdly, computational modelling and experimental measuring of stimulation experiments should be undertaken. Several analytical or theoretical methods have been developed to aid the calculation of EFs generated by different delivery platforms within a range of culture conditions ( Schopf et al, 2016 ; Abasi et al, 2020 ; Boehler et al, 2020 ; Guette-Marquet et al, 2021 ; Zimmermann et al, 2021 ). One study aimed to address this by developing a “digital twin” of an electrical stimulation device ( Zimmermann et al, 2021 ).…”

“…Several analytical or theoretical methods have been developed to aid the calculation of EFs generated by different delivery platforms within a range of culture conditions ( Schopf et al, 2016 ; Abasi et al, 2020 ; Boehler et al, 2020 ; Guette-Marquet et al, 2021 ; Zimmermann et al, 2021 ). One study aimed to address this by developing a “digital twin” of an electrical stimulation device ( Zimmermann et al, 2021 ). This is a digital representation of the device, (a computational model) which is also able to compute a real set of data, and accordingly respond or adapt the model based on the data (i.e., machine learning).…”