Ultrasound sensors based on integrated photonics devices provide a new solution to meet the miniaturization demand for the detection functionality of transducers. An important example is the silicon ring-resonator (RR) ultrasound sensor developed in our department. This sensor combines a small footprint with high sensitivity and a low detection limit. Compared with piezoelectric transducers, it has additional advantages of low cost, mass producibility, and immunity to electromagnetic interference. It is also possible to build an array of these RR sensors on a single chip. These merits make the sensor appealing for ultrasound detection in general. However, there are three challenges that have arisen from this previous work on the RR ultrasound sensor. These challenges relate to specific properties of such silicon RR ultrasound sensors and to the high sensitivity of silicon integrated photonics devices to fabrication variations. More specifically, difficulties in interrogating RR ultrasound sensors need to be overcome. Further, inherent drawbacks of RR ultrasound sensors need to be addressed, suggesting to look for an alternative silicon integrated photonics ultrasound sensor. Finally, design methods need to be developed for making RRs and Mach-Zehnder interferometers (MZIs), which are widely used in silicon photonics, robust to fabrication variations. This thesis addresses the three challenges sequentially. We developed an integrated photonics interrogator to better read out the RR sensor, a silicon MZI based ultrasound sensor to overcome the RR sensor's drawbacks, and design methods for silicon RRs and MZIs with a free spectral range insensitive to fabrication variations.

Magnetic resonance imaging (MRI) is the primary modality for the imaging of soft tissues (e.g. brain, muscle, liver). Therefore, it is an essential radiological tool for diagnosis and surgical planning. The contrast in MR images is due to tissues responding differently to the magnetic fields generated by the scan-ning system. This response can be de-scribed by the physical properties of the tissue (e.g. proton density, magnetic relaxation) and the magnetic fields. These physical properties are represent-ed by multiple parameters that can be estimated through quantitative MRI (qMRI) methods. The parameters are considered more reproducible than con-ventional MR images, which simplifies the comparison of MR data from differ-ent subjects or scanning systems. Esti-mating multiple parameters simultane-ously is needed to reduce error from system imperfections and deliver accu-rate estimates of the physical tissue pa-rameters...

This thesis explores a novel optical architecture for Whole Slide Imaging (WSI). This new architecture allows for multi-focal (3D) image acquisitions in a single scan pass. The multi-focal imaging capability is used to demonstrate 3D phase imaging and 3D imaging of thick tissue sections on a prototype scanner. Further, instrumentation for the extension of WSI to fluorescence imaging is developed: a technologically robust and cost effective method based on LEDs and a highly efficient method based on a multi-line laser illumination source. A WSI system is an optical instrument aimed at creating digital images of bio- logical samples mounted on a microscopy slide at a high throughput. WSI systems image tissues over large fields of view (∼ few cm), in 2D or in 3D (up to hundred layers of μm thickness), and at cellular resolution (∼1 μm). They are applied in high throughput screening in biology, and for novel computer aided medical diagnoses in the field of digital pathology. The optical architecture explored in this thesis is based on a tilted multi-line image sensor concept, originating from Philips. The goal of multi-line image ac- quisition is to enable closed-loop autofocus scanning, but it also allows for multi- focal image acquisitions. The core of this scanner concept lies in a novel design for a multi-line image sensor. The sensor is experimentally characterized for gain and noise, and a system model is developed to find the optimum signal-to-noise ratio (SNR) given the available photo-electron flux. This showed that images with a very high SNR of 292 can be acquired, provided that a sufficiently high photon flux can be realized. Two major issues with the sensor were found in our exper- imental characterization. At high line rates, the sensor showed missing symbols, leading to non-linearities in the read out. Further, the sensor showed high frequent fluctuations in the gain. 3D phase imaging and 3D imaging of thick specimens are two novel contrast modalities based on computational imaging techniques and are enabled by the availability of multi-focal images. For both techniques simplified algorithms were developed compatible with parallel processing at very high speeds. For 3D imaging of thick specimens a deconvolution technique is developed for improving the in- herently low axial contrast. 3D phase imaging is realized by a simplified algorithm for Quantitative Phase Tomography (QPT). QPT imaging is found to be able to im- age the sites labeled for Fluorescence in situ Hybridization (FISH) imaging, and provide additional structural information on unlabeled tissues or tissues stained for immunofluorescence. A system design study is presented showing that the in- plane transfer function has the character of a band-pass spatial frequency filter. The major opportunity for WSI systems to become compatible with fluores- cence imaging is addressed by the development of two imaging modalities. First, a widefield fluorescence WSI system with an LED illumination source is developed and built. A color sequential illumination strategy in combination with multi-band dichroics is used for multi-color imaging using a single monochromatic sensor. The main speed limitation is formed by the exposure time required to capture enough photo-electrons for a decent SNR. Based on the experimental results, a system with 96 Time Delayed Integration (TDI) lines is estimated to achieve a rea- sonable throughput of about 130 kPixel/s. This makes scanning possible of an area of 15 × 15 mm2 in three colors in about 23 min. Second, a novel optical architecture for multi-focal fluorescence image acqui- sitions based on a laser illumination source is proposed and realized in a proto- type. Illumination PSF engineering using diffractive optics is applied to generate a set of parallel scan lines in object space, that span a plane conjugate to a tilted im- age sensor. An important new element in the design is the use of higher order astig- matism to improve the uniformity of peak intensity and line width along the scan lines. Focusing the illumination on the sample provides a very high illumination efficiency and a confocal suppression of background. This optical architecture is projected to ultimately achieve a throughput of several hundreds MPixel/s, which would enable scanning an area of 15 × 15 mm2 in 8 layers in less than a minute. This thesis is concluded with an outlook to opportunities for future research in WSI systems. The challenges and some potential solutions for using a general purpose scientific CMOS (sCMOS) camera for multi-line scanning of a tilted ob- ject plane, and some opportunities for extension of WSI techniques to Light Sheet Microscopy (LSM) and Structured Illumination Microscopy (SIM) are discussed. In summary, this thesis investigates the imaging qualities and extension to computational imaging modalities of a brightfield WSI system and describes two approaches for fluorescence WSI.

An important aspect in increasing our health and safety is the development of new sensors for screening drinking water samples for the presence of microbiological contaminants.The main problem associated with the detection and identification of these microbial contaminants is the long process time. The majority of the time is required fortheir purification and multiplication due to the low concentrations in which they arefound. These steps can be avoided by using spectroscopic identification rather than biochemical techniques. Raman spectroscopy provides a fingerprint based on the vibrational states of molecular bonds from which the bio-particles containing these can be identified. The Raman effect is weak, but rich in information and flexible with respect to its excitation wavelength.When working in an aqueous environment it can take advantage of the absorption minima of water to out perform for instance IR spectroscopy. Optical trapping allows the immobilization of particles without additional preparation steps and provides much added value to, and is highly compatible with Raman spectroscopy. Lab-on-a-chip techniques allow for integration and large-scale parallelization of processes, whichis unavoidable when performing large-scale identification of microbial contaminants on the level of single cells. To take advantage of established CMOS processing techniques for mass production in electronics, a CMOS compatible technology for integrated photonics providing waveguides transparent at a Raman suitable wavelengths, is needed. TripleX is such a waveguide technology. The research presented in this thesis shows the realization of an integrated dual-waveguide optical trap and the feasibility of its use for the identification of micro-organisms based on their Raman spectrum induced by the same on-chip optical beams used for trapping. It does so in fourmain steps. Firstly, laser tweezers Raman spectroscopy is used to classify the closely related yeast species Kluyveromyces lactis and Saccharomyces cerevisiae from measurements at the single cell level. Laser tweezers Raman spectroscopy combines optical trapping of a cell and generation of its characteristic Raman spectrum using one laser-beam focus. This enables fingerprinting of a cell’s molecular composition in a harmless fluidic environment within minutes. Laser tweezers Raman spectroscopy is considerably faster than well-known biological techniques based on streak plating or PCR, and requires far less biological material. For each yeast species a training set and a test set were measured. Visual inspection of the spectra showed intra-species variations obstructing division into two classes by eye. Application of a classification rule based on Fisher’s criterion nevertheless led to the successful blind classification of the test-set cells. Finally, a Kolmogorov-Smirnov test indicated that the difference between the distributions of the species was statistically significant, implying biological origin of the classification. This successful extension of laser tweezers Raman spectroscopy to classification of the aforementioned yeasts underlinesits applicability in microbiology and will hopefully contribute to the process of its adoption in this discipline. Laser tweezers Raman spectroscopy is not limited to rapid classification of single cells, but may also include e.g. study of the cell metabolism. Secondly, a new approach to the dual-beam geometry for on-chip optical trapping and Raman spectroscopy, using box shaped-waveguides microfabricated in TripleX technologyis demonstrated. These waveguides consist of SiO2 and Si3N4, so as to provide alow index contrast with respect to the SiO2 claddings and low signal loss, while retaining the advantages of Si3N4. The waveguides enable both the trapping and Raman functionality with the same dual beams. Polystyrene beads of 1µm diameter can be trapped with this device. In the axial direction discrete trapping positions occur, owing to the intensity pattern of the interfering beams. Interpretation of the trapping events on the basis of simulated optical fields and calculated optical forces indicate that a strong trap is formed by the beams emitted by the waveguides. Furthermore, the acquisition of Raman spectraof a single trapped bead is demonstrated. The spectra obtained in this manner show distinct polystyrene Raman peaks for integration times as short as 0.25 seconds. Thirdly, usual procedure of background subtraction is found to be less effective for Raman spectra obtained with the dual-waveguide trap, due to its specific geometry. The differences in the Raman generating properties between four dual-waveguide traps with varying distances between their waveguide facets are explored using a saturated ascorbic acid solution. Furthermore, the origin of a periodic background observed in the ascorbic acid spectra is investigated. Finally, an alternative method of signal acquisition and processing is presented to deal with the lack of fluidic control in the device and the periodic background and low signal-to-noise ratio observed in the spectra. The 10 µm and 5µm traps are used in trapping and Raman generation experiments with biological relevant particles in the form of Bacillus subtilis spores. These experiments result in noisy spectra for many and few spores in both traps. Using the presented processing the spectra are identified as Bacillussubtilis spore spectra. A comparison of the obtained signal-to-noise values to literature benchmarks shows the feasibility of micro-organisms identification with the dual-waveguide trap.

Disease model systems, such as the zebrafish, play an important role in understanding the onset of diseases like cancer and monitor the efficacy of new drugs. In the past, non-invasive methods for screening, diagnostics and treatment monitoring were intrinsically from the outside. In the past decades, there has been a strong drive to look inside these model systems, which resulted in the development of many small animal tomographic imaging techniques. Due to the absence of ionizing radiation, high-resolution, and cost efficiency, optical tomography is a popular imaging technique to study disease model systems such as zebrafish. The main obstacles in obtaining high-resolution imaging suitable for tissue characterization are the scattering of light in tissue and diffraction of optical waves. Scattering of light in tissue degrades the resolution of optical tomography systems, especially for thick samples. In this thesis, transmission optical coherence tomography (OCT) is used to select ballistic, non-scattered, from non-ballistic, scattered, light. We demonstrate that transmission optical coherence tomography is a versatile tool to measure optical properties of liquids, solids, and particle suspensions. The developed technique is used to perform quantitative optical tomography of the refractive index and attenuation coefficient. A good agreement is observed between our measurements and literature values for group refractive index, group velocity dispersion, and attenuation coefficient. Based on the tomographic reconstruction of transmission OCT measurements, the median attenuation coefficient, group refractive index and volumes of various organs of an adult zebrafish are segmented and quantified in optical coherence projection tomography reconstructions. In optical tomography light is imaged by a lens onto the camera. Due to the focusing of light onto the camera, this light is collected non-uniformly along the propagation direction from the sample. Consequently, the straight-ray assumption as in standard (pre-) clinical X-ray CT reconstruction is violated. Reconstruction of optical tomography images with standard filtered back projection (FBP) causes radial blurring and tangential blurring that becomes stronger with increasing distance to the rotation axis. We present 2D and 3D tomographic reconstruction algorithms that include the point spread function (PSF) of the imaging system. For emission optical projection tomography, these methods show greatly reduced radial and tangential blurring over the entire field of view 113 114 Summary and a significantly improved signal-to-noise ratio compared to FBP. The 3D PSF-based algorithm is evaluated using different initializations. When initialized with the 2D PSF-based reconstruction result, the 3D PSF-based reconstruction gives an improved signal-to-background and image quality in a useful timeframes. Besides including the physical point spread function (PSF) in the 2D tomographic reconstruction, the effect of the PSF also can be reduced by deconvolution of the FBP reconstructed image or filtering the sinogram before FBP reconstruction. We compared the performance of these techniques with each other based on simulations and the signal-to-noise ratio and the sharpness in reconstructed fluorescent beads and zebrafish OPT images. We demonstrate that the sinogram filtering performs poorly on data acquired with high numerical aperture optical imaging systems. We show that the deconvolution technique performs best for highly sparse, low signal-to-noise ratio objects. The PSF-based reconstruction method is superior for non-sparse objects and data of high signal-to-noise ratio. In this thesis, we developed novel algorithms for transmission OCT signal processing and PSF-based tomographic reconstruction. Our algorithms allow for high-resolution quantitative imaging in turbid media. These techniques can be used for quantitative optical imaging of disease model systems. Potentially this may lead to more insight in tissue development and disease onset, progression, and treatment.

The brain’s white matter mainly consists of (myelinated) axons that connect different parts of the brain. Diffusion-weighted MRI (DW-MRI) is a technique that is particularly suited to image this white matter. The MRI signal in DW-MRI is sensitized to diffusion of water in the microstructure by introducing strong bipolar gradients in the MRI pulse sequence. By measuring the diffusion in different directions, the local diffusion profile of water molecules is obtained which reflects microstructural characteristics of the white matter.
The focus of this thesis is on the analysis of conventional DW-MRI data acquired in the context of the Rotterdam Scan Study. This is a prospective population-based cohort study with more than 10.000 participants to investigate causes of neurological disease in elderly people. Conventional DW-MRI is defined as diffusion data acquired with a single diffusion-weighting factor and a small number of diffusion-sensitizing gradient orientations. The objectives of this thesis are (1) to enhance our insight in the relation between tissue structure and the DW-MRI signal from conventional DW-MRI sequences, and (2) to develop methods to quantify diffusion properties in the brain as accurately and precisely as possible based on conventional DW-MRI data.
To gain insight into the relation between tissue structure and the DW-MRI signal, simulated DW-MRI signals based on Monte Carlo simulations of spins between randomly packed cylinders are compared to experimentally acquired data from a hardware phantom. The hardware phantom consists of solid fibers and acts as a model for the extra-axonal diffusion. The simulated DW-MRI signal is in good agreement with the experimentally acquired data. Furthermore, simulations show that the DW-MRI signal from spins between randomly packed cylinders is relatively independent of the cylinder diameter for b-values up to 1500 s/mm². For b-values higher than 1500 s/mm², substrates with a smaller cylinder diameter yield a larger attenuation of the diffusion-weighted signal (chapter 2).
Conventional DW-MRI data is commonly analyzed with a technique known as diffusion tensor imaging. Here, thewater diffusion profile is modelled by a 3D Gaussian diffusion profile. However, in white matter structures in close proximity to the cerebrospinal fluid (CSF) the use of the single diffusion tensor model is inappropriate. A novel framework is introduced to analyze white matter structures adjacent to the CSF. In this framework a constrained two-compartment diffusion model is fit to the data in which the CSF is explicitly modeled with a free water diffusion compartment. The proposed diffusion statistics are shown to be relatively independent of partial volume effects with CSF and are applied to study ageing in the fornix, a small white matter structure bordering the CSF (chapter 3).
A significant part of the white matter constitutes of ‘crossing fibers’, whereby two or more white matter tracts contribute to the DW-MRI signal in a voxel. The single diffusion tensor model cannot adequately describe the data in such voxels. To solve this issue a fiber orientation atlas and a model complexity atlas were used to analyze conventional DW-MRI data with a simple crossing fibers model, namely the ball-and-sticks model. It is shown that the application of a fiber orientation atlas and a model complexity atlas can significantly improve the reproducibility and sensitivity of diffusion statistics in a voxel-based analysis (chapter 4).
Finally, a framework is proposed that aims to specifically improve the analysis of longitudinal DW-MRI data. In this framework the ball-and-sticks model is fit simultaneously to multiple scans of the same subject. The orientations of the sticks are constrained to be the same over different scans, while all other parameters are estimated separately for each scan. The use of this framework is shown to increase the precision of estimated ball-and-sticks model parameters in longitudinal DW-MRI studies (chapter 5).
In conclusion, this thesis describes frameworks to enhance the accuracy or precision of estimated diffusion properties of the white matter by applying sophisticated diffusion models to conventional DW-MRI data. We anticipate that many diffusion MRI studies may benefit from the work described in this thesis.

The millions of filter weights in Convolutional Neural Networks (CNNs), all have a well-defined and analytical expression for the partial derivative to the loss function. Therefor these weights can be learned from data with a technique called gradient descent optimization. While the filter weights have a well-defined derivative, the filter size has not. There is currently no other way to optimize filter sizes, then with an exhaustive search over multiple CNNs trained with different filter sizes.
In this report, we propose a new filter called Structured Receptive Field of which the filter size can be optimized during the training stage. This new filter constitutes a parameterization of a normal filter as a linear combination of all 2D Gaussian derivatives up to a certain order. Instead of learning the weights of the resulting filter directly, we learn the weights of the linear combination and the sigma of the Gaussian derivatives that implicitly define the filter weights. The advantage of parameterizing normal filters in this way, is that gradient descent optimization of the continuous sigma parameter which has a well-defined derivative, can be used as a proxy for optimizing the discrete filter size which has no well-defined derivative.
The basic idea for this parameterization comes from scale-space theory in which the scale of image features is studied by parameterizing the image features with a continuous scale parameter. Thereby explicitly decoupling the spatial structure of image feature from the scale at which it occurs in the image. Structured receptive fields have several compelling advantages: they can learn bigger filters without increasing the number of learnable parameters and without increasing the complexity of the structure of the filter.
In this report, we both provide theoretical and empirical evidence that structured receptive fields can approximate any filter learned by normal CNNs. I.e. we show that structured receptive fields of order 4 can approximate the filters in all layers of a pre-trained AlexNet. Furthermore, we demonstrate empirically that structured receptive fields can indeed learn their own filter size during training, and that this extra ability over normal filters seems to give them an advantage over normal filters when used for classification tasks. I.e. by replacing normal filters in a DenseNet architecture and keeping every- thing else the same, a ~1% higher test accuracy was obtained on the highly competitive CIFAR-10 benchmark dataset.
It would not be wise to draw hard conclusions from a single training run on a single dataset, but this result definitely looks promising. In future research, we hope to demonstrate that, because of their extra ability to learn their filter size, replacing normal filters with structured receptive fields will always lead to strictly better or equal performance.