Otosclerosis is a bone disease of the middle ear that often causes conductive hearing loss. This type of hearing loss can be treated by implanting a prosthesis, which replaces the stapes bone and attaches to the adjacent bone: the incus. Current stapes prostheses produce excellent hearing results, however the procedure is difficult to perform and success is not assured. With current prostheses, safe clamping forces cannot be accurately controlled, despite being crucial to the surgery’s long-term success. Inappropriate clamping force may cause slipping of the prosthesis or necrosis of the incus.
The objective of this study was to first create an overview of current prostheses and subsequently examine their pitfalls, and to design a new stapes prosthesis that addresses these pitfalls. Throughout the literature review a classification and evaluation system was developed to catalogue existing and future prostheses based on, for example, air bone gap difference, force controllability and ease of implantation. From this study we found the category of prostheses that utilise elastic material behaviour to securely clasp around the incus to have the most potential in a new prosthesis design.
Presented in this thesis is a novel prosthesis which features a bistable, fully compliant, two-dimensional mechanism, designed to be simpler to insert and with a more predictable clamping force compared with existing prostheses. For any incus diameter within the recorded range, the force around the incus is sure to be safe by design. The new prosthesis can be placed around the bone without causing unwanted stresses in the ossicles and joints, and is easily clicked shut.

Thermal oxidation in silicon microelectromechanical systems (MEMS) induces stress in the oxide film and silicon. Such residual stresses are usually unwanted, as it can influence performance or damage components. It can also cause unwanted plastic deformation in silicon, as was observed in previous research. Currentmeasurement methods inMEMSare not able to distinguish plastic and elastic strain as a result of thermal oxidation. This thesis presents a novel method to distinguish elastic and plastic strain in silicon beams, by removing the oxide layer to show the plastic strain. A lever mechanism is used as a mechanical amplifier. The plasticity model developed by Alexander and Haassen (AH) is
used in a numerical model to predict the elastic and plastic strain. Experiments in epitaxially grown silicon show significantly less plastic strain than predicted by the model. We conclude that the AH model is not valid for epitaxially grown silicon with low or zero initial dislocations. As significant plastic deformation was observed in FZ silicon samples in previous research by P.R. Kuppens, it is
concluded that epitaxially grown silicon is the better choice when plastic deformation is to be avoided.

Advanced tools such as machine learning are slowly finding their way into the modern scientist’s toolbox . In the design of mechanical systems however hardly any machine learning applications are being used. Research into the viability of such an application is therefore necessary.
We have performed such research, using a specific type of machine learning, known as reinforcement learning, for the synthesis of kinematic mechanisms. Reinforcement learning is an experience-based learning strategy which has proven particularly successful in learning to play games, like chess, blackjack or Go. In this research it is shown that the sequentially alternating nature of game-playing between actions and reward can also be observed in mechanism design by posing design challenges in a game-like format. We have used a decision- tree based mechanism representation developed by Lipson [1] to create such a game-like world in which mechanisms can be designed. To train an actor to navigate this game-like world both Monte Carlo and Temporal Di erence learning have been applied, in combination with a neural network as nonlinear value function approximator. Moreover a kinematic simulator and scoring modules have been implemented to evaluate synthesized mechanisms.
We demonstrated the successful implementation of the framework and learning algorithm by synthesizing mechanism for two separate path tracing objectives: straight lines and figure- eights. A set of recommended algorithm settings was extracted from a parameter sweep and grid search including a total of 560 test runs. Straight line mechanisms were obtained with a fixed maximum number of 10 nodes, drawing lines with aspect ratios up to 1:1168. Additionally a mechanism was synthesized capable of drawing figure-eight patterns.
We conclude that the use of reinforcement learning in the context of mechanical system design is viable. More specifically using the presented method kinematic synthesis for path tracing objectives can be performed. The current research cultivates the land for future e orts to bridge the gap between the challenges faced by mechanical system design groups and the advancing solutions developed by the computer sciences.