Positron Emission Tomography (PET) scanners provide functional three-dimensional images of the body that are extremely useful in cancer and brain research. The goal of this work is the modeling, design and characterization of a CMOS-based photodetector for PET. To this aim, first a model for the energy resolution and coincidence resolution time (CRT) for digital, SPAD-based detectors is developed. Then, a top-to-bottom detector architecture is proposed, containing an innovative in-pixel com-pression technique that allows for high fill-factor (FF) and efficient readout. At the top-level of the architecture, an integrated discriminator monitors the photon flux for incoming gamma events, enabling an event-based readout scheme. The first complete implementation of this archi-tecture is described, the SPADnet-I sensor, which is composed by an 8×16 pixel array, each of around 0.6 × 0.6 mm2 with 720 SPADs, resulting in a pixel FF of 42.6%. The sensor can obtain the discrete photon flux estimation at up to 100 Msamples/s, which are used by the discriminator and also output at real-time. The complete characterization of the sensor is presented, and the best sensor configuration was found to be at 84% of the SPADs enabled (disabled starting with the highest DCR one), with 2 V SPAD excess bias and 150 ns integration time. This configuration results in an energy resolution of 10.8% and a CRT of 288 ps, the latter which was obtained with a new, hardware-friendly time of arrival (ToA) estimation algorithm, also described in this thesis. Finally, the sensor model, validated by the experimental results, is used to predict the perfor-mance of possible modifications in the sensor, and some design improvements are suggested for a future implementation of the architecture.

This doctorate thesis is focused on the design, fabrication and characterization of Micro Electro Mechanical System (MEMS) Vibrational Energy Harvesters (VEHs). The targeted field of application of such devices is the emerging Internet of Things (IoT), in particular for Ultra Low Power (ULP) autonomous applications. In order to realize the ubiquitous paradigm remote and distributed nodes have to be small and in high number. The power requirement of such nodes is generally satisfied by means of batteries, which require periodic replacement and so are not desirable in an autonomous system. To overcome this limitation devices able to harvest energy from the surrounding environment have been investigated. Among the different sources of energy that could be harvested, the vibrational one results promising due to its high power density and its spreading in most environments of interest. The devices developed convert the vibrational energy scattered in the environment into electrical energy by means of a piezoelectric material. The thesis presents studies on both the mechanical and the electric design of a MEMS piezoelectric VEH, with particular attention on multi-modal design. The thesis presents a novel multi-modal device able to extract energy from multiple resonances in a wider bandwidth. Such a design presents two enabling features for IoT application, a wider working band and the compactness, making it more attractive with respect to cantilever like devices.

Automatic monitoring of environments, resouces and human processes are crucial and foundamental tasks to improve people's quality of life and to safeguard the natural environment. Today, new technologies give us the possibility to shape a greener and safer future. The more specialized is the kind of monitoring we want to achieve, more tight are the constraints in terms of reliability, low energy and maintenance-free autonomy. The challenge in case of tight energy constraints is to find new techniques to save as much power as possible or to retrieve it from the very same environment where the system operates, towards the realization of energy neutral embedded monitoring systems. Energy efficiency and battery autonomy of such devices are still the major problem impacting reliability and penetration of such systems in risk-related activities of our daily life. Energy management must not be optimized to the detriment of the quality of monitoring and sensors can not be operated without supply. In this thesis, I present different embedded system designs to bridge this gap, both from the hardware and software sides, considering specific resource constrained scenarios as case studies that have been used to develop solutions with much broader validity. Results achieved demonstrate that energy neutrality in monitoring under resource constrained conditions can be obtained without compromising efficiency and reliability of the outcomes.

In recent years, Silicon Photomultipliers (SiPMs) have been increasingly used as photo-detectors in Positron Emission Tomography (PET) application, which is a nuclear imaging tech-nique that is used to accurately image biochemical processes inside the human body. A SiPM is composed by an array of parallel connected micro-cells of Single Photon Avalanche Diodes (SPADs), and can be classified mainly into two categories Digital-SiPMs (D-SiPMs) and Analog-SiPMs (A-SiPMs). In A-SiPMs, all the microcells share the same bias voltage and have a common readout line. Through custom manufacturing process, the performance of A-SiPMs can be exten-sively optimized. With respect to A-SiPMs, D-SiPMs are composed by many SPAD pixels, each one containing one SPAD and local front-end circuitry. The digital SiPM takes advantage of CMOS technology to perform a 1-b direct A/D conversion per SPAD thus providing a fully digi-tal implementation. On the other hand, SPADs fabricated in CMOS process typically suffer high noise since the critical SPAD implants can hardly be optimized by using the standard CMOS process flow. The main activities carried out within this PhD thesis have been focused on two critical as-pects relevant for the optimization of PET systems performance: (I) the improvement of the per-formance of SPAD in D-SiPMs and (II) the development of high-performance A-SiPM readout application specific integrated circuit (ASIC). Concerning the first point, novel SPADs have been fabricated in CMOS 150-nm technology targeting at low noise, high sensitivity and excellent timing jitter. Three structures with different shapes, geometries and diameters, have been implemented in three test chips. Measurement re-sults of one p+/n-well SPAD array show a 0.4Hz/µm2 dark count noise, 0.85% afterpulsing for a dead time of 150ns at 3V excess bias. The photon detection probability is about 31% at 450nm wavelength at 5V excess bias. The SPAD exhibits a timing jitter of 82ps (FWHM) under a blue laser, which is potentially suitable for D-SiPMs in PET application. The second objective of this PhD work was to develop A-SiPM readout ASIC for PET appli-cation. To utilize the high intrinsic time resolution of A-SiPMs, the development of specialized, highly integrated readout electronics is required. Therefore, two ASICs, first chip with test struc-tures and 16 channels and the second chip with 32 channels, have been developed in 150-nm CMOS technology, with the aim of developing a compact A-SiPM module. The performance of the second chip has been validated by using 3 × 3 × 5 mm3 LYSO crystals coupled to 4 × 4 mm2 SiPMs (FBK-NUV-HD). The measurements show an energy resolution of 14.7% FWHM for the detection of 511 keV photons and the coincidence time resolution is 433ps (FWHM). To improve the timing resolution, part of the PhD work was carried on Stanford University, focused on char-acterization of A-SiPMs and analysis of noise contribution.

In the past few years, several interesting developments in microstructured solid-state thermal neutron detectors have been pursued. These devices feature high aspect-ratio cavities, filled with neutron converter materials, so as to improve the neutron detection efficiency with respect to coated planar sensors. In the framework of the INFN HYDE (HYbrid Detectors for neutrons) project, we have designed new microstructured sensors aimed at thermal and fast neutron detection. Owing to the different cross section, neutron imaging is complementary to X-ray imaging allowing for a high contrast in soft materials. To this purpose, the possibility to have pixelated neutron detectors compatible with existing read-out chips (e.g., those from the Medipix/Timepix family) is an important goal that was achieved in this thesis. In this thesis the entire workflow will be described in detail, covering the design, simulations, fabrication and characterization of 3D neutron detectors for imaging produced at FBK. As a related topic, new 3D sensors for the "Phase-2" upgrades at High Luminosity LHC have been developed, and some aspects relevant to the device simulation and characterization are also reported in this thesis.

Geiger-Mode avalanche photodiodes (GM-APDs) are diodes designed to operate at a reverse voltage that exceeds the breakdown voltage. Their ability to detect single photons combined with their excellent timing resolution make them ideal for applications in which low amplitude signals need to be detected with sub-ns timing resolution. In the research activity reported in this dissertation two different applications of Geiger-Mode Avalanche Photo Diode arrays have been analysed: a two-tier GM-APD array with in-pixel coincidence for particle tracking and a direct Time-of-Flight range meter with a SiPM-based receiver. The increasing complexity of particle tracking systems at collider experiments demands for high performance detectors with high granularity and position accuracy, low material budget and power consumption. Both Hybrid and monolithic detectors have been proposed as particle trackers. Hybrid detectors have the advantage of allowing an independent customization of sensors and readout electronics, but they exhibit a higher input capacitance that increases readout noise, thus limiting their minimum thickness (100 μm). Monolithic detectors have lower input capacitance, thus a better minimum detectable charge, with a charge time collection of few tens of ns. Thanks to their high gain, particle detectors based on GM-APDs have the potential for combining low material budget, low power consumption and an excellent timing resolution in the range of 100 ps. However, their use as particle tracking detectors has been prevented by their Dark Count Rate, since it is not possible to discriminate a particle-generated event from a dark event. To overcome this issue, the use of GM-APDs in coincidence has been proposed. The activity reported in this dissertation has been conducted in the framework of APiX2 and ASAP projects funded by Istituto Nazionale di Fisica Nucleare (INFN). A two-tier sensor based on avalanche detectors in coincidence has been designed and fabricated in standard 150 nm CMOS technology. A charged particle crossing both GM-APDs integrated in a pixel triggers an avalanche in both devices. The output signal from each detector reaches the coincidence electronics, that allows the device to discriminate the particle detection from dark events. The performance of the proposed detector has been evaluated in a complete electrical and functional characterization campaign. The feasibility to reduce the substrate thickness has been studied in some devices thinned down to 50 μm and 25 μm. The electronics for coincidence detection was also used to directly measure optical crosstalk, a phenomenon that is getting greater importance as stacked optical and image sensors are becoming common. The functional characterization of the APiX prototype was performed with a beta-source and an evaluation of the radiation hardness of the devices was carried out in an irradiation campaign with neutrons at the INFN Laboratori Nazionali di Legnaro (LNL). SiPMs take advantage of the characteristics of GM-APDs such as high sensitivity, high efficiency and very low time jitter, and overcome the problem related to the dead time connecting several sensitive elements in parallel, making them suitable for the simultaneous detection of more than one photon. The evaluation of a SiPM-based direct Time-of-Flight range meter has been performed at two different wavelengths: 405 nm and 810 nm. The set of measurement at 405 nm has been performed using a TCSPC module as acquisition system, while in the 810 nm measurements a low-cost FPGA-based TDC was used. The replacement of the TCSPC module with an FPGA-based TDC represents an important step towards the integration of a low-cost prototype thanks also to the low power consumption of the device. In order to evaluate the feasibility of a SiPM-based range meter in the NIR region, a collaboration with the Circuits and Systems Research Unit of the ITEE Faculty of the University of Oulu was established to set up a system with a GaAs/AlGaAs multiple quantum well laser diode with a spectral emission of 0.808 μm as transmitter and a NIR-HD SiPM with an enhanced efficiency for NIR photons recently developed at FBK as receiver. The evaluation was performed at high repetition rate (MHz range), for the perspective purpose to upgrade the system including a 2-axis scanning mirror to perform real-time 3D imaging.

Parkinson’s disease is a neurodegenerative disorder due to the death of the dopaminergic neurons of the substantia nigra of the basal ganglia. The process that leads to these neural alterations is still unknown. Parkinson’s disease affects most of all the motor sphere, with a wide array of impairment such as bradykinesia, akinesia, tremor, postural instability and singular phenomena such as freezing of gait. Moreover, in the last few years the fact that the degeneration in the basal ganglia circuitry induces not only motor but also cognitive alterations, not necessarily implicating dementia, and that dopamine loss induces also further implications due to dopamine-driven synaptic plasticity got more attention. At the present moment, no neuroprotective treatment is available, and even if dopamine-replacement therapies as well as electrical deep brain stimulation are able to improve the life conditions of the patients, they often present side effects on the long term, and cannot recover the neural loss, which instead continues to advance. In the present thesis both motor and cognitive aspects of Parkinson’s disease and basal ganglia circuitry were investigated, at first focusing on Parkinson’s disease sensory and balance issues by means of a new instrumented method based on inertial sensor to provide further information about postural control and postural strategies used to attain balance, then applying this newly developed approach to assess balance control in mild and severe patients, both ON and OFF levodopa replacement. Given the inability of levodopa to recover balance issues and the new physiological findings than underline the importance in Parkinson’s disease of non-dopaminergic neurotransmitters, it was therefore developed an original computational model focusing on acetylcholine, the most promising neurotransmitter according to physiology, and its role in synaptic plasticity. The rationale of this thesis is that a multidisciplinary approach could gain insight into Parkinson’s disease features still unresolved.

Autism Spectrum Disorders (ASDs) describe a set of neurodevelopmental disorders. ASD represents a significant public health problem. Currently, ASDs are not diagnosed before the 2nd year of life but an early identification of ASDs would be crucial as interventions are much more effective than specific therapies starting in later childhood. To this aim, cheap an contact-less automatic approaches recently aroused great clinical interest. Among them, the cry and the movements of the newborn, both involving the central nervous system, are proposed as possible indicators of neurological disorders. This PhD work is a first step towards solving this challenging problem. An integrated system is presented enabling the recording of audio (crying) and video (movements) data of the newborn, their automatic analysis with innovative techniques for the extraction of clinically relevant parameters and their classification with data mining techniques. New robust algorithms were developed for the selection of the voiced parts of the cry signal, the estimation of acoustic parameters based on the wavelet transform and the analysis of the infant’s general movements (GMs) through a new body model for segmentation and 2D reconstruction. In addition to a thorough literature review this thesis presents the state of the art on these topics that shows that no studies exist concerning normative ranges for newborn infant cry in the first 6 months of life nor the correlation between cry and movements. Through the new automatic methods a population of control infants (“low-risk”, LR) was compared to a group of “high-risk” (HR) infants, i.e. siblings of children already diagnosed with ASD. A subset of LR infants clinically diagnosed as newborns with Typical Development (TD) and one affected by ASD were compared. The results show that the selected acoustic parameters allow good differentiation between the two groups. This result provides new perspectives both diagnostic and therapeutic.

Heart diseases are the leading cause of death worldwide, both for men and women. However, the ionic mechanisms underlying many cardiac arrhythmias and genetic disorders are not completely understood, thus leading to a limited efficacy of the current available therapies and leaving many open questions for cardiac electrophysiologists. On the other hand, experimental data availability is still a great issue in this field: most of the experiments are performed in vitro and/or using animal models (e.g. rabbit, dog and mouse), even when the final aim is to better understand the electrical behaviour of in vivo human heart either in physiological or pathological conditions. Computational modelling constitutes a primary tool in cardiac electrophysiology: in silico simulations, based on the available experimental data, may help to understand the electrical properties of the heart and the ionic mechanisms underlying a specific phenomenon. Once validated, mathematical models can be used for making predictions and testing hypotheses, thus suggesting potential therapeutic targets. This PhD thesis aims to apply computational cardiac modelling of human single cell action potential (AP) to three clinical scenarios, in order to gain new insights into the ionic mechanisms involved in the electrophysiological changes observed in vitro and/or in vivo. The first context is blood electrolyte variations, which may occur in patients due to different pathologies and/or therapies. In particular, we focused on extracellular Ca2+ and its effect on the AP duration (APD). The second context is haemodialysis (HD) therapy: in addition to blood electrolyte variations, patients undergo a lot of other different changes during HD, e.g. heart rate, cell volume, pH, and sympatho-vagal balance. The third context is human hypertrophic cardiomyopathy (HCM), a genetic disorder characterised by an increased arrhythmic risk, and still lacking a specific pharmacological treatment. / Le malattie cardiache e cardiovascolari sono ad oggi la causa principale di morte nel mondo. Tuttavia, i meccanismi ionici responsabili di aritmie o di altre malattie cardiache non sono ancora del tutto conosciuti: questo spesso porta a una minore o mancata efficacia delle terapie attualmente disponibili, e lascia numerose domande aperte per gli elettrofisiologi. Inoltre, la difficoltà di acquisizione dei dati sperimentali rimane ancora uno dei problemi più grandi in questo campo. Infatti la maggior parte dei dati vengono raccolti in vitro e/o utilizzando modelli animali come coniglio, ratto o cane, sebbene l’obiettivo ultimo sia quello di una più completa comprensione del comportamento elettrico del cuore in vivo e nell’uomo, in condizioni sia fisiologiche sia patologiche. In questo contesto, la modellistica computazionale costituisceuno strumento indispensabile: infatti, le simulazioni in silico permettono di superare, almeno in parte, i limiti sperimentali, e di investigare i meccanismi ionici alla base di specifici fenomeni a diversi livelli (singola cellula, tessuto, intero cuore). Una volta validati sui dati sperimentali, i modelli matematici possono essere dunque utilizzati per fare predizioni, testare ipotesi e valutare l’efficacia di eventuali interventi farmacologici. Lo scopo di questa tesi di dottorato è stato quello di applicare tecniche di modellistica matematica a problemi di elettrofisiologia cardiaca, in particolare utilizzando modelli di potenziale d’azione (PA) umano in tre diversi contesti: variazioni elettrolitiche nel sangue, effetti della terapia dialitica sul cuore e cardiomiopatia ipertrofica.

Falls are common and burdensome accidents among the elderly. About one third of the population aged 65 years or more experience at least one fall each year. Fall risk assessment is believed to be beneficial for fall prevention. This thesis is about prognostic tools for falls for community-dwelling older adults. We provide an overview of the state of the art. We then take different approaches: we propose a theoretical probabilistic model to investigate some properties of prognostic tools for falls; we present a tool whose parameters were derived from data of the literature; we train and test a data-driven prognostic tool. Finally, we present some preliminary results on prediction of falls through features extracted from wearable inertial sensors. Heterogeneity in validation results are expected from theoretical considerations and are observed from empirical data. Differences in studies design hinder comparability and collaborative research. According to the multifactorial etiology of falls, assessment on multiple risk factors is needed in order to achieve good predictive accuracy.