Myocardial ischaemia induced by cardioplegia is the most prominent risk during open-heart surgery. To achieve adequate protection of the cardiomyocytes during surgery, the cardioplegia must arrive at all cardiomyocytes. Local obstruction can lead to regional ischaemia. For surgeons and perfusionist, the arrested heart is a black box. They know how much cardioplegia they are administering to the heart. However, they do not know if cardioplegia is reaching all cells. This work describes which parameters can be measured during cardioplegia induced cardiac arrest and compares the methods used in the literature to detect these parameters. Following this, it is the goal to create a proof-­of-­concept sensor for the detection of myocardial ischaemia. A literature review indicated that a fluorescent optochemical pH sensor has the most potential. To be able to develop a proof-­of-­concept, optical, chemical, and medical knowledge needs to be combined. This is necessary to map what is required and what is possible for optochemical in vivo sensing. First, a framework is designed and used to select the best technique and material suited for this project. In this framework, the medical requirements and the resources available are combined. As a result, this thesis work will use a dual wavelength pH­-sensitive fluorescent dye encapsulated in a biocompatible hydrogel. In preparation for the proof-­of-­conceptsensor, multiple samples are fabricated and extensively tested to achieve the optimal sensing layer. Using the optimised concentrations, a proof-­of-­conceptsensor is created using a miniature reflection probe and an USB spectrometer. This proof-­of-­conceptsensor shows that it can measure the changes in pH and can be corrected for multiple interferences. It also shows the potential to be further miniaturised and to be used during cardiac surgery. Before this can happen, chemical optimisation of the sensing layer is needed, and the consistency of the sensing layer needs to be improved. However, besides this, the work succeeded in selecting an optochemical sensing technique and material which shows the potential to be used in cardiac surgery

Nowadays, Dopple's wireless earphones have touch buttons as a user inter-
face, however these take in quite some physical space. Since the earphones
are becoming smaller over the years, there is a need for a new type of touch-
less user interface, that is smaller. In order to create that new user interface,
research is done on the topic of remote gesture sensing with sensors that can
t onto wireless earphones. When gestures are recognized by the system,
a corresponding action can be taken, like for example pausing the music.
Small infra-red imaging sensors are chosen as a solution to the problem. Its
images are analysed by a trained image recognition neural network created
with Python and Keras. This network takes an image as input and outputs
a gesture. Each gesture is supposed be linked to an action in the new user
interface. This report focusses on the retrieval of low resolution infra-red
images and neural network training/machine learning. The APDS-9500, an
already existing moving gesture sensor, is used as a comparison with regard
to the new neural network technique. It achieves an accuracy of 92.3% with
5 dierent gestures. The AMG Grid Eye is an 8 by 8 pixel infra-red camera
for which 5 gestures are trained. For raw images, 5 gestures are recognized
with an accuracy of 79.2%. With the help of pre-processing in the form of
contrast increasing and linear extrapolation the accuracy is increased up to
92.4%. The FLUKE 279 FC is a high resolution camera mounted on a mul-
timeter of which the images are downscaled to 30 by 30 pixels. It is found
out that while the size of the model increases, accuracy also increases up to
97.2% for 5 gestures. When the FLUKE is tested with 9 dierent gestures,
while also optimizing for size, an accuracy is achieved of 98.2% with a model
of 203 kB. The study proves that 30 by 30 infra-red images contain enough
information to use gesture recognition with a small neural network.