Quantum computers are potential next generation computing system, and are expected to consist of millions of quantum bits (Qubits) which are typically required to operate at cryogenic temperature. Electronic devices are used to control and read out qubits, however there are limitations of wiring massive number of electronics and qubits between room temperature and cryogenic environment, which causes a bottleneck for the size-up of the quantum computer. Cryogenic temperature CMOS electronics, which are able to operate at low temperatures and be placed close to qubits have been proposed to overcome this bottleneck.
This work explores superconducting qubit, a type of widely used qubit in latest quantum computers, and the current output DAC designed for flux bias, which is a control method of superconducting qubits. The state-of-the-art DAC designs still face the problem of too high power consumption that restricts placing the control electronics close to the qubits. This thesis proposes a cryogenic current output high compliance regulated Digital-to-Analog converter, reducing the power consumption by 80% comparing to the latest studies.
This thesis proposes an architecture combining a slow DAC array and a fast switch array to realize fast speed while avoiding the high power requirement of high-speed fine DACs. The design uses an operational amplifier to build a feedback loop to regulate a cascoded current mirror array as the core of the high compliance regulated current DAC. The DAC achieves an output range of 0 to 500 𝜇A, with 8.41 nA of accuracy; a sampling rate of 250 MHz and 2.1 ns of rising and falling time. The maximum total power consumption is 179.4 𝜇W, which is less than 20% of the state-of-the-art design examples. The estimated chip area of this design is 2.7 𝑚𝑚2.

Automated diagnosis of epilepsy for differentiating epileptic EEGs without Interictal Epileptic Discharges (IEDs) from normal EEGs remains a critical challenge in clinical settings.
Current state-of-the-art methods use algorithms that can effectively detect epilepsy seizures which improves the current treatment methods for people suffer from epilepsy. Electroencephalograms (EEGs) analyzed by neurologists which are not able to meet the criteria are further looked into to obtain an efficient classification. This thesis presents an automated epilepsy diagnosis approach using a multi-algorithmic feature extraction pipeline. The final models include the development of a robust multi-processing feature extraction pipeline, the application of advanced machine learning / deep learning algorithms, and the validation of the proposed methods on comprehensive EEG datasets. The results, achieved using an XGBoost classifier with leave-one-subject-out (LOSO) cross-validation, demonstrate comparable performance to state-of-the-art epilepsy detectors. The study emphasizes the detection of epilepsy without IEDs, optimizing models through nested cross-validation, and evaluates their performance on the Temple University Hospital (TUH) and Erasmus Medical Center (EMC) Rotterdam datasets.

The device that is designed and described in this thesis aims to non-invasively stimulate neural activity in an earthworm at multiple, specific locations. This is done by inducing an electric field from two inductors with a specific electric field strength in mind. The response of the earthworm is recorded and amplified to enable analysis of this signal. Initially, requirements are set that need to be met by this device. From these requirements a design process was started and simulations were made. After this, a device was created on which the requirements could be evaluated. Due to the physical trials not being completed the requirements could not be fully verified.

To achieve greater specificity in neurostimulation, bidirectional neural interfaces are required to verify the recorded neural response after stimulation. The specific neural interface targeted in this thesis is the epiretinal implant. Due to the heterogeneity of the retinal ganglion cells (RGCs), high-fidelity vision is only possible when all the types of RGCs near the neurostimulator are mapped. This necessitates the bidirectionality of the implant and poses significant difficulties, as large stimulation artifacts obscure the small neural response that needs to be recorded. Furthermore, in order to cover a large area of the retina, a channel count in the order of 10⁴ will be required. Scaling existing neurostimulators would lead to chips that are >30mm², which is too large.

The aim of this thesis is therefore to design a neurostimulator for the epiretinal implant that is capable of implementing artifact-reducing algorithms, and is smaller than the current state-of-the-art. The proposed system makes use of a mismatch-based digital-to-analog converter (DAC), and has been optimized for an output range of 0-6 μA at an effective resolution of 8-bits. Furthermore, in order to decrease the amount of stimulation units required, waveform interleaving has been proposed, where the anodic and cathodic stimulator are separated. A voltage compliance monitor is also designed to ensure proper stimulation output. The designed system has been fabricated and occupies 0.0003mm² for two channels. Scaling this directly to 10⁴ channels would result in an area of 1.645mm². This area can be reduced even further via electrode multiplexing, which the designed system readily allows for. An output availability (i.e. how many input codes are possible after calibrating) of 99.2% and 97.3% is reported for the anodic and cathodic stimulator at an 8-bit resolution over the full output range.

Introduction: Over the past two decades deep brain stimulation (DBS) has emerged as an important therapeutic option for Parkinson’s disease (PD). However, the current DBS programming method, monopolar review (MPR), is time-consuming, requires highly trained personnel and causes discomfort for patients. This study aimed to predict the optimal stimulation contact(s) based on local field potential (LFP) recordings by the implanted leads and a sensing enabled DBS system, and as such, improve the efficiency of DBS programming in PD patients.

Methods: Level-based LFP recordings (OFF-medication) within the first two post-operative weeks in PD patients implanted with directional Sensight leads® and the Percept PC® neurostimulator in the Haga Teaching Hospital were retrospectively analysed. Time and frequency domain data were inspected for artefacts. From the individual theta (4-7 Hz), alpha (8-12 Hz), beta (13-35Hz) and gamma (≥36 Hz) bands the maximum power (Max.) and area under the curve above 1/frequency (AUC_flat) were extracted. The clinically chosen contact during MPR served as reference for all predictions. Machine learning models using AUC_flat features from frequency band combinations were evaluated using nested cross-validation. Two custom ranking methods, pattern based and decision tree, were developed for both beta band features individually. The predictive accuracy (Acc.) of the 1st and 2nd prediction combined was evaluated on a training and unseen test set, considering all data and subgroups based on amount of symptoms during MPR and amount of beta activity above 1/frequency. The ranking methods were additionally compared to an existing algorithm (DETEC).

Results: Recordings from 34 patients (68 subthalamic nuclei) were analysed. Artefacts did not overlap with frequencies of interest or were sporadic of nature. The machine learning model with the highest performance was a linear discriminant analysis combining raw beta and alpha features (AUC: Design: 0.86, Test: 0.69). For the 1st and 2nd predicted contacts combined, the two best performing ranking models were pattern based using AUC_flat (Acc. training set: 86.2%; test set: 100%) and decision tree using Max. (Acc. training set: 87.9%; test set: 100%). Correct pattern based (AUC_flat) predictions were more often 1st opposed to 2nd predictions than correct decision tree (Max.) predictions (Tr = 55.2%, T = 90% vs. Tr = 10.3%, T = 10%). Acc. obtained for subgroups based on amount of symptoms during MPR and amount of beta activity above 1/frequency were similar across all subgroups for the custom ranking methods. The Acc. of the DETEC algorithm was inferior to all custom ranking methods.

Conclusions: This study demonstrates the feasibility of using level-based LFP recordings to predict the optimal stimulation contact in patients with PD. The best results were obtained using the pattern based (AUC_flat) and decision tree (Max.) custom ranking methods. For clinical implementation the decision tree (Max.) ranking method is expected to be favoured. Although prospective research is required to identify the true Acc. of the models in clinical practice, these results show potential to halve the required DBS programming time (only two out of four contacts require evaluation), and can thus improve DBS programming efficiency.

Currently, EEG recordings are mostly made using scalp recording devices. These devices are typically difficult to put on and bulky, limiting their use to a lab environment. In-ear EEG recording is proposed as a more practical alternative to scalp EEG recording.

In-ear EEG replaces the scalp electrode array with more discrete earpieces, recording the EEG signals from the ear canal instead of the scalp. Because of their increased convenience with respect to their scalp electrode array counterpart, the earpieces allow for everyday use, enabling the possibility of long-term EEG recording, outside of a lab environment.

This work presents the design of a fully-integrated in-ear EEG device. The device has been designed to record EEG signals, process them locally, and transmit the processed data to a smartphone or personal computer. The device is made to be easily re-programmable, making it possible to change the device firmware depending on the intended application. This functionality has been integrated into a small form factor, to allow for the integration of the electronics into an earpiece. A low-power design has been created, to allow for long-term battery operation.

Two prototypes have been made. The first prototype focuses on exploring different system configurations, with the small form factor less of a priority. The performance of the prototype is tested by comparing it to a commercially available scalp EEG recording device based on scalp EEG recordings. For this comparison, the auditory steady-state response (ASSR), steady-state visual evoked potential (SSVEP), and alpha-band modulation paradigms are used. The prototype has shown similar performance to the scalp EEG device in these experiments, with the scalp EEG device performing slightly better for the ASSR and alpha-band modulation paradigms, while the prototype showed the best performance for the SSVEP paradigm.

The second prototype realises the best-performing system configuration that has been found with the first prototype into a form factor that fits into an earpiece. Ear EEG experiments are performed, with electrodes placed around the ears. The ASSR, SSVEP, and alpha-band modulation paradigms have again been used in these experiments. The measured performance on the ASSR and alpha-band modulation paradigms are slightly worse than that measured using the scalp EEG experiments. The SSVEP experiments showed significantly worse performance compared to the scalp EEG experiments. These results align with the expectations, as ear EEG has been documented to have performance close to scalp EEG for the ASSR and alpha-band modulation paradigms, while the SSVEP performance is typically worse for ear EEG than for scalp EEG.

Engineered Heart Tissues (EHTs) are a valuable approach enabled by Organ-on-Chip (OoC) technology to model human cardiac tissue. These small microfluidic devices allow the culture and development of living cardiac cells in 3D structures, to reproduce tissue dynamics and functionality in-vitro, thus fostering the development of new and more precise models for human diseases and organs’ physiological response.
One of the most important gaps in this recent technology is the lack of integration of electronic devices in the platforms, such as electrodes for tissue stimulation or sensors for real-time monitoring of the tissue.

To cover this gap, a polymer-based platform with microwell and micropillars for culturing EHTs and measuring their contractile properties was developed in ECTM group at TU Delft. The contraction force exerted by the beating cardiac tissue, self-assembled around the micropillars, is quantified by measuring the displacement of the micropillars with spiral capacitive sensors embedded in the substrate. The force generated by the tissue corresponds to a capacitance change which is simulated to be in the aF range, requiring a high-precision, sensitive, and portable read-out circuitry.

The investigation and development of this readout electronics is the final goal of this Master Thesis. A literature survey about possible capacitive readout techniques was conducted to identify suitable architectures for measuring such small dynamic changes in capacitance. Two solutions available on the market (Smartec UTI and Analog Devices AD7746) implementing two of these architectures were chosen, tested, and characterised. Benchmark measurements with accurate laboratory instrumentation were performed, and noise figures of the two solutions were evaluated.
To allow the readout, EHT platforms with embedded sensors need to be transferred and assembled on a custom Printed Circuit Board (PCB): the viability of this challenging assembly process was evaluated. The multiple constraints deriving from such a complex project determined the development of a non-standard assembly process, which proved to be delicate and gave origin to multiple failure modes. Those were documented and analysed, to identify alternative or improved assembly procedures which need to be developed to fabricate reliable samples, since these weaknesses were identified as the most critical aspect at this point of the project.

The results obtained showed how the two solutions for the readout provided results in good agreement with more precise non-portable laboratory instrumentation, and promising noise figures. The platforms with embedded sensors were successfully transferred to the developed PCBs, and measurements showed good agreement with the simulated static behaviour of the sensors, thus providing a valuable proof-of-concept for the whole project.
The dynamic behaviour of the sensors was preliminarily investigated and characterized using nanoindentation tests, indicating the possibility to measure a linear relationship between the force applied and the difference in the sensed capacitance. Despite this good result, further tests need to be performed to verify it and to address some criticalities that were identified.
Cardiac tissue was cultured on platforms with integrated sensors and the biocompatibility of the entire system was proved. The behaviour of the sensors during biological measurement with cardiac cells is left to future investigation.

This thesis discusses the basic architecture, design details, circuit implementation, and measurements of a digital class D current driver.
The driver contains two main parts: a digital control loop and analog circuits.
Parts of the important part in the digital control loop contain noise shaper, plant compensator, and cable capacitance compensator. The noise shaper has two core functions: 1. It seeks to minimize the difference between the process variable and the set point. 2. It shapes the noise that is generated in the forward loop. The plant compensator is used to compensate for the output filter, the load, and the cable. The cable capacitance compensation is a measure-based technique to generate a negative output capacitance to compensate for the cable capacitance. The analog circuits consist of a full bridge, data acquisition circuits, and floating supplies.
Multiple simulation models have been made in Matlab. According to the simulation results, the amplifier shows its flexibility and performance to handle a large range of inductive loads.
A prototype has been implemented. The measurement results show that most of the hardware performance is as intended. Due to the time limitation, the whole amplifier is not tested.

Patients that suffer from heart failure can benefit from wearable monitoring devices that continuouslymonitor the condition of the heart. One of the foremost symptoms of exacerbation of the heart is fluidcongestion in the lungs. One of the method to measure a change in biological tissue, such as thebuild­up of fluids in the lungs and chest, is bioimpedance measurements. By injecting an alternatingcurrent into the tissue, a voltage develops across the tissue that is proportional to the impedance ofthe biological tissue, the bioimpedance. Conventional bioimpedance measurement techniques are notsuitable for continuous monitoring of the patients as they are power consuming and hinder patient intheir daily living.This thesis proposes an alternative approach that is based on a pulse width modulation (PWM) inorder to convert the measured analog signals to time signals. To determine the measured bioimped­ance, the magnitude and phase should be derived. The proposed design employs one channel toconvert the measured voltage across the bioimpedance to a PWM signal. As this PWM signal con­tains the amplitude information of the measured voltage, the magnitude of the bioimpedance can bederived. Furthermore, a second reference channel is employed where a known resistive reference isconverted to another PWM signal. By comparing the two PWM signals, the phase of the bioimpedancecan be determined. The proposed system requires only a comparator and triangular wave in order toconvert the measured analog signals, compared to the complex implementation of the conventionallyused analog­to­digital converter.In order to validate the design, the circuit is simulated and implemented on a printed circuit board (PCB). The PCB operates correctly on 3.3V and, additionally, an voltage­controlled current sourceis implemented and connected externally to the PCB to provide an excitation current of 100μA and10kHz to the circuit. The circuit should be capable of measuring the voltage across a device under test(퐷푈푇) that consist of resistive and capacitive components. This is because the bioimpedance can bemodelled by the Fricke­Morse model, which consist of a resistor in parallel with a capacitor and resistorin series. The implemented PCB can measure 퐷푈푇 magnitudes up to 1kΩ and is determine the phaseshift between the two PWM signals.This work shows an important contribution towards a wearable continuous bioimpedance measure­ments system to monitor patient that suffer from heart failure. It has been shown that the presenteddesign can measure the magnitude and phase that are required to determine the measured bioimped­ance, and also reduce complexity of the measurements instrumentation.

Medical Body Area Networks (MBANs) are a cluster of possibly heterogeneous devices, communicating with each other in, on or around the human body. Through these devices, medical data is collected, processed in some way and transferred outside of the network. The IEEE 802.15.6 standard aims to govern communications between such devices. It includes a set of constraints for physical features and communication on the PHY and MAC level, as well as association/ disassociation protocols and security services that applications need to comply with. Given the high sensitivity of the medical data transmitted via MBANs, network security is crucial. This thesis consists of three main contributions: (a) a structured procedure to analyse the security features of the IEEE 802.15.6 standard by using realistic hypothetical scenarios is introduced (b) a thorough security analysis of the standard is conducted (c) recommendations on how to improve the security posture of the standard are given.

Vagus nerve stimulation (VNS) has been proven to be an effective treatment for patients suffering from drug­-resistant epilepsy. Current vagus nerve stimulators are either implantable devices or bulky, handheld devices. The implant consists of a pulse generator beneath the skin of the chest, connected to electrodes that are wired to the cervical vagus nerve. While capable of delivering localized electrical impulses, the implantation is highly invasive. Handheld devices do not ask for a risky surgery, but their imprecise targeting of the neck reduces efficiency and provokes additional side effects.

Ultrasound has been widely used in clinical applications ranging from focused ultrasound (FUS) tissue ablation to medical imaging. Much research is being carried out on the use of FUS neuromodulation as a low-­cost, non-invasive alternative to electrical VNS. The combination of using ultrasound for neuromodulation as well as imaging allows for image-­guided FUS where automated tracking of the nerve location improves the targeting of the vagus nerve.

In this thesis, the design of an ultrasound vagus nerve stimulator with combined imaging capabilities is presented. The novel architecture combines FUS for vagus nerve neuromodulation and plane wave imaging (PWI) with optimized sparse receive sub­arrays to obtain the location of the nerve within the neck.

A 2D array of piezoelectric transducers with λ/2-pitch poses a strict constraint on the pixel area. The goal of the system design is to minimize the area consumption of the on­-pixel electronics by making use of already present circuit blocks. A shared delay­-locked loop generates the required continuous phases for FUS neuromodation, while a row-­column pulse control block generates a single pulse from the available phases for use in PWI. This way the on-­pixel transmit hardware is reduced to a multiplexer and high­-voltage driver.

The receiver consists of an analog front­-end and analog-­to­-digital converter and is shared among a sub-­array of receive elements. Different degrees of random sparsity are explored for the optimal implementation of the receive sub-­arrays within the full array. Synthetic aperture imaging allows the multiplexing of multiple receive elements to the shared front-­end and reduces the time spent on imaging using dynamic beamforming on a single data set.

The full system­-level architecture of a complementary metal­-oxide­-semiconductor (CMOS) interface is proposed and the receiver front-end has been designed and simulated.

The overarching goal of Engineered Heart Tissues (EHTs) is to develop functional 3D heart tissues in vitro with the potential to find drug targets, identify drug toxicity and predict the effects of the drugs in the body. This 3D model of the cardiac tissue consists of a bunch of cells, self-assembled around specific anchoring points the main aim of which is to support tissue formation. Although this model is very promising, still has to overcome an important drawback which is the lack of maturation of iPSC-derived cardiomyocytes (CM). This obstacle leads to limited recreation of the adult human cardiac tissue physiology. It has been shown that electrical stimulation of cardiac cells is one of the most important factors for cardiomyocyte’s maturation. Therefore, the principal goal of this thesis is to design and fabricate an electrical stimulator device and integrated electrodes in an existing EHT platform for electrical stimulation of cardiomyocytes. In the beginning, a literature study was conducted on different electrical stimulation methods and existing electrical stimulation devices for this purpose. For the first part of the thesis which was the design and fabrication of electrical stimulator, the electrical stimulation parameters that must be fulfilled by this device, were specified. Various hardware design approaches were studied in detail and compared before the design and implementation of the final idea. Simulations of the selected method using the LTSpice software tool were performed in order to study the behavior of the electrical stimulator design. The device is a 16-channel electrical stimulator that provides perfect rectangular biphasic pulses in the range of±15Vwith adjustable voltage amplitude, frequency and duty cycle. A user-friendly interface allows the user to select the desired channel of stimulation and the signal is distributed providing electrical stimulation to a total of EHT platforms. Therefore, this device was designed in a way to be connected with a second Printed Circuit Board(PCB) which hosts 16-EHT with integrated electrode chips. The 16-EHT holder (PCB)was also designed to suit in a 96-well plate, used by biologists for cardiac-cell culturing inside the incubator. The characterization of the electrical stimulator took place in the measurement room of the EKL lab. The results of the measurements verify the precision and efficiency of the electrical stimulator circuit as they are in excellent agreement with the simulations. The second part of the thesis includes the fabrication of integrated electrode-chips. The fabrication was conducted at EKL lab, in the Microelectronics Department of TUD using clean room microfabrication techniques. The selected electrode material was TiN due to its unique mechanical properties and the electrodes were fabricated on PDMS, encapsulated between two layers of polyimide. The main steps required for the integration of these electrodes in the existing EHT platform have been also carried out.

The functioning of the brain depends on the interplay between a large number of neurons. To understand the information processing in neuronal networks, we need tools to record the electrical activity of cells at high resolution. Microelectrode arrays (MEAs) are predominantly used to measure neuronal activity
at high spatial and temporal resolution. With the advent of complementary metal-oxide-semiconductor (CMOS) based MEAs, it has been possible to design high-density MEAs with electrode sizes comparable to that of the individual neurons, allowing sub-cellular resolution. CMOS technology has also facilitated the on-chip signal conditioning needed to record the low-amplitude bio-signals with superior signal quality. In this thesis, a readout architecture for in-vitro MEAs has been proposed for a low-noise extracellular action potential (AP) readout.

One of the challenges in MEA implementation is the design of thousands of low-noise readout channels for simultaneously recording the signals. There is a need to design low-noise ADCs with optimal power and area consumption. In this thesis, a unique low-noise oscillator based sigma-delta ADC has been designed for MEA applications. With the advantages of oversampling and time-encoding techniques, the in-band noise has been optimized, without increased hardware complexity. The simple implementation of this proposed ADC makes it efficient in terms of area and power consumption.

The integrated circuit for this oscillator based sigma-delta ADC has been implemented in 0.18 um CMOS technology to demonstrate the feasibility of high-order oscillator-based ADCs for low-noise extracellular AP readout. It was possible to obtain a noise below 5 uV_RMS (simulations) and power consumption under 3 uW using this ADC, which approximately occupies an area of 0.002 mm². The action potential readout system implemented with this ADC has been taped-out for further analysis through measurements.

Due to the earth’s population growth and the increased living age of a person, more and more people could deal with medical problems. Early detection of health related problems in both healthy and people with medical problems, will catch problems early and prevent the need for more costly procedures. One of these medical problems is concerning the hydration level. Dehydration most often occurs for elderly people. To prevent sickness and even death, monitoring of the hydration can play a key factor in prevention. Hydration monitoring can be established by using a designed CMOS permittivity sensor. However, to design a commonly used hydration monitoring device, this CMOS permittivity sensor needs to be designed in such a way that portable sensing could be enabled. To design such a portable hydration sensor from an application sensor, different steps are required. To bring previously developed sensor to enable hydration monitoring, it need to be packaged in such a way that it can sense the medium of interest and be connected to the outside world while sufficiently protected, so that it will not damage during use. Different packaging procedures could be distinguished to accomplish this. The first packaging procedure is regarding the creation of SU8 metallized pillars. The second packaging procedure is for creation of cavities in SU8 and fill these cavities with aluminium particles. Besides the packaging of the sensor chip, validation measurements are required to ensure that a relationship could be detected with the permittivity sensor between the hydration level of the human skin. To enable measurements of the skin permittivity without bulky systems, a PCB layout should be designed.

The displacement of a loudspeaker cone is not linearly proportional to the voltage/current at its input provided. This is due to electrical and mechanical limitations. The distortion is more noticeable at lower frequencies. This report is one of two reports, that will explain how to minimize this distortion. The motion of the loudspeaker cone will be measured and compared with the input voltage. This way the error in the output can be found. Finally, a system will adjust the current through the voice coil to decrease this error. An accelerometer is used as a sensor and the ADAU1777 is used to compensate for the error in output displacement. Finally, a voltage to current converter is used to convert and amplify the output voltage from the ADAU1777 into a current signal to drive the loudspeaker. This report focuses on the voltage to current amplifier. To begin the design, a problem definition and program of requirements will be given. Usually, loudspeakers are voltage driven. This report begins by first explaining why a current-driven loudspeaker delivers less distortion. Then an attempt is made to build a single operational amplifier (opamp) design that meets the requirements for a real load. For the single opamp design, a simple version of the operational transconductance amplifier (OTA) will be compared against a Howland model based on noise performance and circuit analysis, where it will be shown that the simple OTA has a larger signal to noise ratio and does not need to meet any additional criteria to maintain a high output impedance. However, the simple OTA still does not meet the Signal to Noise (SNR) . The simple OTA also did not meet the maximum Total Harmonic Distortion (THD) at 800 Hz. Then, a two opamp design will be constructed using a composite configuration to increase the SNR and decrease the THD. This design does meet the requirements given, at least for a purely resistive load. Then an electrical model of a loudspeaker, which has a complex impedance, will replace the resistive load. Now the composite design for the resistive load needs to be adjusted for the inductive characteristic of a loudspeaker at higher frequencies. The resulting design has a THD of 0.000427\%, a PM of 51$\degree$ and a SNR is 100.92 dB. All results will be given as simulations, because the current pandemic (COVID-19) does not make it possible to physically construct and measure this model. Micro-Cap 12 is used for circuit simulations and Slicap for circuit noise analysis.

This thesis report focuses on possible methods of digital implementation of motional feedback in a bass loudspeaker. In this thesis, the control system will be created for a monopole and a dipole speaker specifically. It does so by sketching the outlines of such a system and examining possible designs for the components required to suppress distortion which is mainly created by the speaker. This thesis shows that the digital implementation indeed allows filtering with very little latency but there are limitations in accuracy due to the conversion from the analogue to the digital domain. It is shown in this thesis that a partly analogue and partly digital system is desired for better error correction. A desired input-output gain of 30=29.54dB is achieved. The controller that is suggested is a controller based on the transfer function of the speaker. By transforming the input signal with the inverse of this open-loop transfer, the closed-loop transfer will be flattened. The instrumentation amplifier used to subtract the feedback signal from the input signal is designed such that the influence of the quantization noise is minimized. Loop-gain is added by the instrumentation amplifier, the controller and voltage-to-current amplifier. By creating loop-gain as much as possible using the digital implementation, noise is suppressed from 30dB up to 90dB within the range of frequencies of interest. The designed components in the motional feedback system allow a loop-gain up to 70dB and an input to output voltage gain of 30dB before instability occurs.

Absence seizures have a real-life impact on epileptic subjects, as day-to-day tasks can by suddenly interrupted making for dangerous situations. Though a lot of work has been done on seizure detection, to limit the impact on epileptic patients, the true necessity lies in timely prediction of seizures before they manifest. Various attempts have been made using conventional algorithms to accurately predict seizures however, so far, results are not that encouraging. In this work, we applied various machine-learning algorithms, in an attempt to identify complex, multi-dimensional epileptic precursors in brain recordings. Three types of neural networks are used in this feasibility study, namely Multi-Layer Perceptron (MLP) networks, Long Short-Term Memory (LSTM) networks and Gated Recurrent Unit (GRU) networks. The used input data was annotated Electrocorticography (ECoG) data, recorded in living mutant rodents, containing epileptic events at an interval of about one minute. The data was pre-processed for better learning performance by data normalisation and by generating distinctive training features. The neural networks were configured as three-class classifiers, distinguishing among inter-ictal, pre-ictal and ictal periods. A grid-search approach was applied to determine the best set of parameters for the neural networks. Despite our best efforts, the relation between the input data and output data could not be learned in a reliable way. The maximum reached Average Prediction Rate (APR) was 0.57 with a prediction time of 3.1s when using the normalised data as input and 0.65 with a prediction time of 6.1s when using the distinctive features as input. These results essentially signify good detection but virtually no prediction of upcoming seizures. The evaluation of the experimental findings has revealed that the employed ECoG recordings were ill-selected for training our various neural-network models. Also, a non-conclusive exploratory experiment is performed by applying a Weibull Time-To-Event Recurrent Neural Network (WTTE-RNN) on a sub-set of the normalised input data. The experiment has yielded some positive results, a short-notice prediction of the upcoming seizure in some cases, encouraging for further exploration of this approach. Despite the limited success of this work, however, through its extended forensics analysis, it has paved the crucial, initial steps in the direction of seizure prediction.

Electronic implants are becoming a valuable tool to explore and regulate neural activity, potentially overcoming neural disabilities that are still incurable. On the one hand, neural recording can provide tremendous insight into the neural system and the sometimes accompanied neural diseases. On the other hand, neural stimulation is necessary to control or adjust neural activity related to neural diseases. Detail of neural recording and stimulation is achieved with highly dense and often deep implantable electrodes (spatial resolution), while neural patterns are found with larger neural area coverage. In both stimulating and recording, a combination of spatial resolution and area coverage can be key to understanding and curing. Application-specific integrated circuits (ASICs) are often employed for the interaction with the electrodes interfacing with neural tissue but ASICs are limited in size, which consequently limits the amount of individual recording or stimulation channels. The de facto solution to circumvent this limitation is to multiplex high numbers of electrodes to a single ASIC channel. However, due to switching and signal latencies only a limited number of electrodes can be multiplexed per channel. Multiple channels on an ASIC are therefore desirable nonetheless to accommodate implants with a many electrodes. Due to recent miniaturization advances, an ever-increasing number of channels can be made available on a single ASIC and the urgency arises to investigate connecting and integrating these channels to electrodes on a flexible implantable substrate. In this work we will investigate technology for assembling a fully flexible electrode array substrate to an ASIC with a high number of channels. For such assemblies, a routing optimiser was developed especially for flexible implants, optimising for chip size and critical substrate limitations. However, with no established manufacturing methodology for such systems, technology parameters were not fully understood. Therefore, initial efforts were put into fabrication of a custom chip with which 72 chip-to-substrate gold stud contacts can be evaluated by means of 4-point measurements and daisy-chaining. This multifunctional chip allows for experiments at various chip-to-substrate contact diameters and pitches. In addition, different materials, including polycarbonate (PC), polyimide (PI) and thermoplastic polyurethane (TPU) were investigated as candidate substrate materials. A transfer methodology, using copper and FR-4 as carrier materials, proved fruitful for these polymeric materials. TPU samples showed particularly promising results, with the main advantage that it can encapsulate the entire implantable system, including chip and metallization, without the formation of additional material interfaces. This has significant advantages for the longevity of implantable systems. A second chip iteration with similar functionality was manufactured with 1008 gold electroplated contacts. The chip measures at 4x4mm² and contacts were designed to be 36x36µm² in size, pitched at 80µm apart. Contact resistances of this chip on TPU substrates were in the range of 5.5 and 73mΩ but the yield of proper contact dropped from 100% to an estimated 90% compared to experiments with the previous chip. A demonstrator product, fully embedded in TPU, was made using the same chip and connects 324 peripheral electrodes to the chip on a single Au metallization layer.

Fluid accumulation in the human body is, in many cases, a symptom of some underlying pathological condition. Most common examples include edema, ascites, pleural effusion, and internal bleeding. Currently available non-invasive methods to assess fluid accumulation in the body are mainly imaging-based (i.e. ultrasound, computed tomography). This means that monitoring is limited to spot-checks and is only performed when the presence of a particular condition is already suspected from prior clinical evaluation of the patient. There is high clinical interest in the development of a system allowing such monitoring to be performed automatically, continuously over a prolonged period of time, and independently of prior clinical evaluation. For instance, in the case of internal bleeding, hemorrhage indicators such as variation in heart rate and blood pressure are only observable after significant blood loss (up to more than 1 liter) has already taken place, therefore an early warning system can save lives, decrease hospital stay length and significantly reduce complication-related costs. In this thesis, a pioneer solution is proposed for development of such a monitoring system. The hypothesized solution is based on the observation of specific energy-describing features of the Ballistocardiogram (BCG) signal, a measure of the periodic displacements generated on the body as a result of ballistic forces produced by the heart during each cardiac cycle. Because of additional damping generated by the presence of internally accumulated fluid, the energy of this signal is expected to decrease compared to its baseline value. The BCG can be recorded unobtrusively with different sensing modalities on the surface of the body. In order to validate the research hypothesis and assess the feasibility of this novel technique, a custom experimental set-up exploiting several different BCG sensing options has been used to carry out a study on 15 human volunteers. Localized fluid accumulation along the GI tract was induced in a controlled, safe and simple fashion, by means of water intake by the participants, and the BCG signal was recorded before and after intake. Signal feature exploration and performance analysis has been used to develop an optimized feature able to accurately capture the decrease in signal energy due to fluid accumulation, as well as to identify the most suitable sensor type and monitoring body location for the present application. The developed energy-describing feature shows a significant decrease in energy value from baseline to after-intake condition with a p-value<0.001. Moreover, the selected feature was able to correctly identify presence of fluid accumulation with high sensitivity (90% in bed-based, and 100% in standing-position monitoring). Given the promising results of the present study, further research towards improvement and development of the proposed technique is highly encouraged.