Real-time monitoring device for 4d bioprinting based on electrical impedance tomography

• Main Author: Morcelles, Kaue Felipe
• Publication Date: 2021
• Format: Master thesis
• Language: por
• Source: Repositório Institucional da Udesc
• Download full: https://repositorio.udesc.br/handle/UDESC/16034
• topic: Engenharia elétrica; Bioengenharia; Impedância bioelétrica; Diagnóstico não invasivo

Human tissue bioprinting is one of the most exciting fields in tissue engineering and regenerative medicine. It consists in the automatic manufacture of three-dimensional biological structures, which can then be applied in patient specific tissue replacement, drug development and more reliable in vitro biological models. One limitation of tissue bioprinting is the lack of non-destructive, real-time and remote methods to monitor the evolution of cells and tissue during the maturation process. This thesis describes the development of a non-invasive and remote sensor to evaluate bioprinted tissue evolution in real-time using EIT. The prototype consists of an electrode sensor, an analog front-end and a digital unit. The electrode sensor is made of 16 gold-surface signal electrodes arranged in a circular configuration, with a ground electrode at the center, and a polypropylene cylinder fixed around the array to act as a culture well. The analog front-end is composed of three stages: the commutation module, the excitation circuit and the voltage sensor circuit. The commutation module is based on the AD75019 cross-point switch, allowing arbitrary selection of electrodes for excitation and sensing via serial control. The excitation circuit consists of a Differential Howland Current Source with a transconductance of 150 µS, optimized for low output common-mode operation. The source operates with square-wave signals at 10 and 100 kHz. The voltage sensing circuit is composed of two input buffers, a differential VGA and a fourth-order antialiasing filter, based on the differential multiple-feedback topology. The sensing circuit provides selectable gain from 0 to 80 dB with cut-off frequency of 1 MHz. The STM32F303ZE microcontroller was the core of the digital unit, providing analog-to-digital conversion, gain control, channel selection, data preprocessing and communication with the host via UART protocol. To validate the prototype, EIT experiments were performed with biological and bioprinted phantoms. Carrots and apples were used as primary biological materials, and a saline solution with 1.5 S/m was used as background media. A dedicated bioprinter was develop to print the hydrogel phantoms, using an alginate-gelatin bioink. To collect the boundary voltages, the adjacent measurement protocol was used, with injection currents of approximately 310 µA at 10 and 100 kHz. Both TDEIT and FDEIT were tested. A MATLAB interface was developed to control the prototype and reconstruct the images, using the Iterative Total Variation Regularization algorithm. Both interface and reconstruction algorithms were developed on top of the EIDORS library. The proposed system was capable of reconstructing the EIT images for all the phantom structures. From the TDEIT images, information about the position, size and orientation of the phantom could be obtained. Moreover, it was possible to differentiate the biological tissue from the bioprinted hydrogel structure using FDEIT, which is fundamental to track the cell growth inside the scaffold. However, the use of a 2D EIT limited the performance of the device, resulting in poor shape identification capabilities, and the use of static vegetable phantoms is not optimal to represent tissue growth in vitro. Therefore, future works will focus on developing 3D EIT algorithms and testing the device with bioprinted human tissue.