Developing Biosensor Technology to Monitor Biofilm Formation on Voice Prosthesis in Throat Cancer Patients Following Total Laryngectomy

Voice prostheses (used to replace an excised larynx in laryngectomy patients) are often colonised by the yeast Candida albicans, yet no monitoring technology for C. albicans biofilm growth until these devices fail. With the current interest in smart technology, understanding the electrical properties of C. albicans biofilm formation is necessary. There has been great interest in Passive Radio Frequency Identification (RFID) for use with implantable devices as they provide a cost-effective approach for sensing. The main drawback of RFID sensors is the need to overcome capacitive loading of human tissue and, thus, low efficiency to produce a high read range sensor design. This is further complicated by the size restriction on any RFID design to be implemented within a voice prosthesis as this medical device is limited to less than 3 cm in overall size. In order to develop such a voice prosthesis sensor, we looked at three separate aspects of C. albicans colonisation on medical devices within human tissue. To understand if it is possible to detect changes within a moist environment (such as the mouth), we developed a sensor capable of detecting minute dielectric changes (accuracy of ± 0.83 relative permittivity and ± 0.05 S·m-1 conductivity) within a closed system. Once we understood that detection of dielectric changes within a liquid solution were possible, to overcome human tissue capacitive loading of RFID sensors. Adjusting backing thickness or adding a capacitive shunt into the design could limit this tissue effect and even negate the variability seen between human tissues. Without developing these methods, implementation of any RFID device would be difficult as human tissue variability would not be compensated for properly. Finally, biofilm growth in terms electrical properties. As C. albicans biofilm matures, there is a loss in capacitance (the biofilm becomes increasingly hydrophobic) prior to 24 hours after which the biofilm thickness shifts the resonance leading to a slow gain in capacitance. Understanding all of these aspects allowed us to develop two final voice prosthesis sensors producing read ranges above 60 cm and 10 cm within a tissue phantom. Ultimately, this showed the possibility of developing cost-effective passive RFID sensor technology for monitoring microbial biofilm formation within human tissue, leading to more effective real-time clinical care.