Millions of people around the world rely on kidney dialysis to replace the essential function of their kidneys. In order to receive dialysis, adequate access to the circulation is needed. The golden standard, the Arteriovenous Fistula (AVF), hasn’t really changed since the 1940’s and comes with disappointing patency and reintervention rates. This is most likely due to increases in the flow pattern in the shunt and vein adjacent to the shunt. For a novel type of shunt that can regulate the blood flow, an implantable actuator needs to be developed and validated that can function as an interface and supply a force to the valve regulating the blood flow. In this study, an implantable actuator was developed using a methodological design approach that can deliver various output forces over an axial displacement using a magnetic coupling. It gives maximum control over the shunt by allowing control over the speed of the displacement and introduces multiple locking states. The actuator was then tested using various experimental tests. First the maximum force output was tested, then the reliability and durability were tested. Lastly, a goat cadaver study was performed to test the function of the device. From this study it was concluded that the actuator met all the design requirements and therefore provided an adequate proof-of-concept for the actuator as part of a dynamic arteriovenous shunt system. Next, the actuator should be implemented in the full shunt system and get connected to the valve using a suitable transmission system.

Background: Patients suffering from end-stage renal disease (ESRD) often require haemodialysis to filter the blood of waste products. Currently, an arteriovenous fistula (AVF) or graft (AVG) is used in most cases to enable sufficient blood flow for haemodialysis, but these have been associated with a large number of undesirable patient outcomes. Most of these can be linked to the resulting constant high blood-flow and the anastomosis remaining in an open position. To counter this, an implantable Dynamic Arteriovenous System (DAS) that can regulate blood flow through an anastomosis is to be developed. For this, a valve mechanism is to be designed and validated, that converts a mechanical translation from an actuator into an opening and closing motion on a synthetic graft. Objective: The goal of this research is to design a proof-of-concept DAS valve mechanism, and validate this ex vivo. Method: A methodological approach was utilised to generate concepts for a DAS valve that can fully collapse and open a synthetic graft acting as anastomosis. A Harris profile with a set of grading criteria is set up to determine the optimal concept. This concept is assessed through means of a number of test setups, that includes force measurement during collapse and opening of the graft, traction, and lifetime. Results: The result of the design study is a linkage mechanism utilising pinhole joints that converts the input into an (approximately) vertical translation on the top and bottom of the graft. A transmission ratio that approaches infinity when the graft is fully collapsed is achieved, whilst allowing an infinite number of states to control flow when open. The majority of the requirements set are met, but the necessary input force, traction andmass require some optimisation to fulfil the criteria. Conclusion: From this study it was concluded that the current design shows promise but a next design iteration is necessary to fully meet all design criteria. Future research should include the transmission of the actuator input to the valve and the assessment of foreign body response, which may require the development of a sealing method.

Introduction In the Netherlands, approximately five thousand people are dependent on maintenance hemodialysis (HD) treatment. The one-year mortality of this patient population in 2021 was as high as 18%. A significant contributor to this high mortality and burden on HD patients is the high incidence of intradialytic hypotension (IDH), which can lead to hypoperfusion of vital organs. Despite advances in fluid management, including technical innovations such as relative blood volume (RBV) monitoring and bioimpedance measurements, the incidence of IDH remains unacceptably high. Therefore, the primary goal of this master thesis is to assess the feasibility of machine learning-based real-time prediction of IDH using patient data from HD sessions at the LUMC with the aim to enhance individualized fluid assessment. In this study, the performance of a recurrent neural network, known for its ability to retain time-varying information, will be compared to other machine learning models. Methods Patient demographics and intradialytic measurements, such as blood pressure and RBV measurements, were obtained from all HD patients of the LUMC with an age ≥18 years. The two definitions of IDH that were assessed were defined as IDH1, a drop of ≥20 mmHg compared to the pre-dialytic SBP and IDH2, is an absolute intradialytic SBP of ≤90 mmHg. Performance of a logistic regression, random forest and extreme gradient boosting (XGBoost) model were assessed and compared to that of a recurrent neural network (RNN). The models provided a prediction of IDH at every 30 minutes during HD treatment. Feature importances were determined for the logistic regression, random forest and XGBoost and two subanalyses were conducted to assess predictions made 60 minutes ahead instead of 30 minutes, and the effect of removal of blood pressure measurements on the prediction. Results 37,025 HD sessions of 436 patients were included in the analysis of IDH1 and 43,722 sessions of 441 patient in the analysis of IDH2. The Area under the Curve (AUC) of the Receiver Operating Characteristics (ROC) curve for the logistic regression, random forest, XGBoost and RNN were 0.80, 0.81, 0.81 and 0.87 respectively for the prediction of IDH1 and 0.93, 0.84, 0.86 and 0.92 for IDH2. At the highest clinically acceptable false positive rate (FPR) of 0.1, the RNN demonstrated a precision, recall, and specificity of 0.56, 0.59, and 0.90, respectively, for IDH1, and 0.10, 0.76, and 0.90 for IDH2, and the logistic regression 0.63, 0.50 and 0.90 for IDH1 and 0.14, 0.80 and 0.90 for IDH2. Conclusion The machine learning algorithms showed the ability to learn from HD data to make a prediction of IDH at different intradialytic timepoints. However, the current models do not meet the criteria for clinical implementation due to their limited performance metrics. Evaluation of the subanalyses revealed that the RNN had the ability to detect patterns that did not only rely on the blood pressure. Therefore, the prospect of real-time predictions of IDH appears promising with further refinement of the RNN and the expansion of the database by incorporating additional features and sessions.