Sensor Technologies for On-line Monitoring of Biological Parameters During Hemodialysis

Modern hemodialysis machines make use of sensors to monitor the proper functioning of the machine, the safety of the treatment and the intra-session alterations of the hematic properties of the patient. The real-time stream of sensor data is also crucial for the implementation of biofeedback algorithms. Non-invasive sensing techniques have the advantages of avoiding contact with the patient’s blood, thus increasing safety and avoiding the use of disposable elements. The primary aim of this doctoral thesis was the development of non-invasive sensing techniques for the measurement of physiological parameters during the hemodialysis session. An additional activity concerning mathematical modeling of the dialysis process was also carried out. The thesis is composed of four chapters. Chapter 1 serves as an introduction to the field of hemodialysis, giving the reader an overview of the different aspects of this medical therapy. In Chapter 2, a new method for the contactless estimation of plasmatic conductivity is described. High-frequency impedance measurement and model fitting are used to estimate electrical conductivity of the liquid contained inside the bloodline. Good results were obtained with measurements on saline solution and a blood-mimicking fluid. In Chapter 3, a system for continuous estimation of relative blood volume loss and plasmatic sodium concentration is described. The system integrates information from multiple sensors, among which is a new fiber-optic sensor, to estimate these two variables with satisfying accuracy. In Chapter 4, a model of sodium diffusion across the hemodialysis filter is reported. The innovative aspect of the presented model is that it interconnects mathematical descriptions of both the patient and the dialyzer, at different abstraction levels, in order to give a more realistic representation of sodium exchange during the treatment. The thesis closes with some final remarks and an outlook for the possible future developments.