· · · Repository hosted by TU Delft Library

We present the design of an algorithm for use in an interactivemusic system that automatically generates music playlists that fit the music preferences of a user. To this end, we introduce a formal model, define the problem of automatic playlist generation (APG), and proof its NP-hardness. We use a local search (LS)procedure employing a heuristic improvement to standard simulated annealing (SA) to solve the APG problem. In order to employ this LS procedure, we introduce an optimization variant of the APG problem, which includes the definition of penalty functions and a neighborhood structure. To improve upon the performance of the standard SA algorithm, we incorporated three heuristics referred to as song domain reduction, partial constraint voting, an da two-level neighborhood structure. We evaluate the developed algorithm by comparing it to a previously developed approach based on constraint satisfaction (CS),both in terms of run time performance and quality of the solutions. For the latter we not only considered the penalty of the resulting solutions, but we also performed a conclusive user evaluation to assess the subjective quality of the playlists generated by both algorithms. In all tests, the LS algorithm was shown to be a dramatic improvement over the CS algorithm.

The just noticeable difference (JND) in black level (BL), whitelevel (WL), and chroma as visible in natural images under practical viewing conditions is determined. The potential effect of cultural background on the JNDs is evaluated conducting the same experiment in China and in the Netherlands. In general, there is a big difference in JND depending on the content of the natural image, as we expected. Only for some images,we found a difference in JND between China and the Netherlands. It seems that for these images people on average were looking to a different area in the image during their assessment

The lack of understanding of the mechanical behavior of the human skin layers makes the development of drug delivery using microneedles or microjets a challenging task. In particular, the key mechanical properties of the epidermis composed of stratum corneum and viable epidermis, should be better understood. Micro-indentation experiments were applied, using a spherical tip with a large diameter to the sample thickness ratio. The Young's moduli were derived via an analytical and a numerical method. The tests showed that the analytical method was not appropriate to assess the Young's moduli. That is why a numerical model was used to obtain the correct stiffness. When loaded perpendicularly, the stiffness of both the epidermis and stratum corneum vary between 1 and 2 MPa. No significant differences in stiffness between the stratum corneum and viable epidermis were observed.

Multi-access networks may exhibit severe unfairness in throughput,in the sense that some nodes receive structurally higher throughputthan others. Recent studies show that this unfairness is due tolocal differences in the neighborhood structure: Nodes with fewerneighbors receive better access. We study the unfairness in satura-ted linear networks,and adapt the multi-access CSMA protocol toremove the unfairness completely,by choosing the activation ratesof nodes as a specific function of the number of neighbors. We theninvestigate the consequences of this choice of activation rates onthe network-average saturated throughput, and we show that theserates perform well in non-saturated settings.

X-ray attenuation properties of matter (i.e. human body in medicalComputed Tomography) are energy and material dependent. This dependency is largely neglected in conventional CT techniques, which require the introduction of correction algorithms in order to prevent image artefacts. The exploitation of the inherent energy information contained in the x-ray spectrum allows distinguishing the two main physical causes of energy-dependent attenuation (photo-electric effect and Compton effect). Currently a number of methods exist that allow assessing the energy-dependent attenuation in conventional systems. These methods consist of using two distinct spectra (kVp switching ordual source) or by discriminating low and high energy photons by means of stacking two detectors. Further improvements can be achievedby transitioning to direct-conversion technologies and counting-modedetection, which inherently exhibits a better signal-to-noise ratio.Further including energy discrimination, enables new applications,which are not feasible with dual-energy techniques, e.g. the possibility to discriminate K-edge features (contrast agents, e.g. Gadolinium) from the other contributions to the x-ray attenuation of a human body. The capability of providing energy-resolved information withtwo or more independent measurements is referred as Spectral CT.A new proprietary photon counting ASIC (ChromAIX) has been developedto provide high count-rate capabilities while offering energy discrimination. The ChromAIX consists of a pixel array with an isotropicpitch of 300 µm. Each pixel contains independent discriminators which enable the possibility to discretize incoming photons into a number of energy levels. Extensive electrical characterization has been carried out to assess the performance in terms of count-rate performance and noise. Observed rates exceeding 10 Mcps/pixel (Poissonian, mean incoming rates > 27 Mcps). The energy resolution is better than4.1 keV FWHM and has been shown to be consistent with simulations. Pile-up behaviour and count-rate dependency have also been evaluated. Electrical crosstalk among pixels in terms of count-rate activity and threshold position has been assessed and show no measureable influences across the array. X-ray tests have also been performed onsamples directly flip-chip bonded to CdTe and CZT crystals. The pulse shape and spectrum obtained from a 241Am source is consistent with simulations.

Pulse-resonance sounds like vowels or instrumental tones contain acoustic information about the physical size of the sound source (pulse rate) and body resonators (resonance scale). Previous research has revealed correlates of these variables in humans using functional neuroimaging. Here, we report two experiments that use magnetoencephalography to study the neuromagnetic representations of pulse rate and resonance scale in human auditory cortex. In experiment 1, auditory evoked fields were recorded from nineteen subjects presented with French horn tones, the pulse rate and resonance scale of which had been manipulated independently using a mucoder. In experiment 2, fifteen subjects listened to French horn tones which differed in resonance scale but which lacked pulse rate cues. The resulting cortical activity was evaluated by spatio-temporal source analysis. Changes in pulse rate elicited a well-defined N1m component with cortical generators located at the border between Heschl's gyrus and planum temporale. Changes in resonance scale elicited a second, independent, N1m component located in planum temporale. Our results demonstrate that resonance scale can be distinguished in its neuromagnetic representation from cortical activity related to the sound's pulse rate. Moreover, the existence of two separate components in the N1m sensitive to register information highlights the importance of this time window for the processing of frequency information in human auditory cortex. Keywords:

The paper presents the synthetic procedure and the structural characterisation of Pr3+-doped eulytite double phosphates Sr3La(PO4)3 and Ba3Lu(PO4)3. The luminescence properties of these materials were studied employing time-resolved VUV spectroscopy upon excitation with synchrotron radiation. The 5d-4f emission of Pr3+ ions was detected and assigned. It was shown that energy transfer from host to Pr3+ 5d states is quite inefficient. At the same time the materials demonstrate unwanted defect-related emission that presents main path for relaxation of host relaxation excitations.

An execution view is an important asset for developing large and complex systems. An execution view helps practitioners to describe, analyze, and communicate what a software system does at runtime and how it does it. In this paper, we present an approach to define viewpoints to guide the construction and use of execution views for an existing large and complex software-intensive system. This approach hasenabled the customization and extension of a set of predefined viewpoints identifying and addressing the requirements for execution views of a specific development organization. The validation and application of the defined viewpoints in practice has helped us to document and transfer them to practitioners to support the construction anduse of execution views of an MRI system, a large software-intensivesystem in the healthcare domain.

Gender recognition is one of fundamental face analysis tasks. Most of the existing studies have focused on face images acquired under controlled conditions. However, real-world applications require gender classification on real-life faces, which is much more challenging due to significant appearance variations in unconstrained scenarios. In this paper, we investigate gender recognition on real-life faces using the recently built database, the Labeled Faces in the Wild (LFW). Local Binary Patterns (LBP) is employed to describe faces, and Adaboost is used to select the discriminative LBP features. We obtain the performance of 94.81% by applying Support Vector Machine (SVM) with the boosted LBP features. The public database used in this study makes future benchmark and evaluation possible.

The local structure of the Ce3+ ion in the yellow emitting YAG:Ce phosphor has been studied by Extended X-ray Absorption Fine Structurespectroscopy in the 300−20 K temperature range. It has evidenced that the dopant Ce3+ replaces Y3+ in the garnet structure, giving rise to a significant expansion of the distorted cubic nearest neighbours coordination sphere. The distortion created by the dopant islimited to first coordination sphere. The experimental distances obtained in this study are in good agreement with the ones derived from periodic-boundary-conditions Density Functional theory calculations.

Introduction: Autocompletion supports human-computer interaction in software applications that let users enter textual data. We will be inspired by the use case in which medical professionals enter ontology concepts, catering the ongoing demand for structured and standardized data in medicine. Objectives: Goal is to give an algorithmic analysis of one particular autocompletion algorithm, called multi-prefix matching algorithm, which suggests terms whose words' prefixes contain all words in the string typed by the user, e.g., in this sense, opt ner me matches optic nerve meningioma. Second we aim to investigate how well it supports users entering concepts from a large and comprehensive medical vocabulary (snomed ct). Methods: We give a concise description of the multi-prefix algorithm, and sketch how it can be optimized to meet required response time. Performance will be compared to a baseline algorithm, which gives suggestions that extend the string typed by the user to the right, e.g. optic nerve m gives optic nerve meningioma, but opt ner me does not. We conduct a user experiment in which 12 participants are invited to complete 40 snomed ct terms with the baseline algorithm and another set of 40 snomed ct terms with the multi-prefix algorithm. Results: Our results show that users need significantly fewer keystrokes when supported by the multi-prefix algorithm than when supported by the baseline algorithm. Conclusions: The proposed algorithm is a competitive candidate for searching and retrieving terms from a large medical ontology.

A quantitative theory for the contrast-to-noise ratio (CNR) in differential phase contrast imaging (DPCI) is proposed and compared to that of images derived from classical absorption contrast imaging (ACI). Most prominently, the CNR for DPCI contains the reciprocal of thespatial wavelength to be imaged, the fringe visibility, and a tunable factor dependent on the system geometry. DPCI is thus potentiallybeneficial especially for the imaging of small object structures. We demonstrate CNR calculations for mammography, finding optimal imaging energies between 15 and 22 keV for ACI, and between 20 and 40 keV for DPCI.

Today’s state-of the art clinical computed tomography (CT) scannersexclusively use energy-integrating, scintillation detector technology, despite the fact that a part of the information carried by the transmitted x-ray photons is lost during the detection process. Roomtemperature semiconductors, like CdTe or CZT, operated in energysensitive photon-counting mode provide information about the energy of every single x-ray detection event. This capability allows novel, promising approaches to selectively image abnormal tissue types like cancerous tissue or atherosclerotic plaque with the CT modality. In thisarticle we report on recent dual K-edge imaging results obtained inthe domain of pre-clinical, energy-sensitive photon counting CT. Inthis approach, the tuning of threshold levels in the detector electronics to the K-edge energy in the attenuation of contrast agents (CA) offers highly specific, quantitative imaging of the distributionof the CA on top of the conventional, morphological image information. The combination of the high specificity of the K-edge imaging technique together with the powerful tool of targeting specific diseases in the human body by dedicated contrast materials might enrich theCT modality with capabilities of functional imaging known from thenuclear medicine imaging modalities, e.g., positron-emission-tomography but with the additional advantage of high spatial and temporal resolution. We also discuss briefly the technological difficulties tobe overcome when translating the technique to human CT imaging andpresent the results of simulations indicating the feasibility of theKedge imaging of vulnerable plaque using targeted gold nano-particles as contrast materials. Our experiments in the pre-clinical domainshow that dual-K edge imaging of iodine and gold based CAs is feasible while our simulations for the imaging of gold CAs in the clinical case support the future possibility of translating the technique to human imaging.

Snoring is a prevalent disorder affecting 20-40% of the general population. The mechanism of snoring is vibration of anatomical structures in the pharyngeal airway. Flutter of the soft palate accounts forthe harsh aspect of the snoring sound. Natural or drug-induced sleep is required for its appearance. Snoring is subject to many influences such as body position, sleep stage, route of breathing and the presence or absence of sleep-disordered breathing. Its presentation may be variable within or between nights. While snoring is generallyperceived as a social nuisance, rating of its noisiness is highly subjective and, therefore, inconsistent. Objective assessment of snoring is important to evaluate the effect of treatment interventions. Moreover, snoring carries information on the site and degree of obstruction of the upper airway. If evidence for monolevel snoring at thelevel of the soft palate is provided, the patient may benefit frompalatal surgery. These considerations have inspired researchers to scrutinize the acoustic characteristics of snoring events. Similarlyto speech, snoring is produced in the vocal tract. Because of this analogy, existing techniques for speech analysis have been applied toevaluate snoring sounds. It appears that the pitch of the snoring sound is in the low frequency range (< 500 Hz) and corresponds to a fundamental frequency with associated harmonics. The pitch of snoringis determined by vibration of the soft palate, while nonpalatal snoring is more noise-like, and has scattered energy content in the higher spectral sub-bands (> 500 Hz). To evaluate acoustic propertiesof snoring, sleep nasendoscopy is often performed. Recent evidencesuggests that the quality of snoring is markedly different in drug-induced sleep as compared with natural sleep. Most often, palatal surgery alters sound characteristics of snoring, but is no cure for this disorder. It is uncertain whether the perceived improvement afterpalatal surgery, as judged by the bed partner, is due to an alteredsound spectrum. Whether some acoustic aspects of snoring, such as changes in pitch, have predictive value for the presence of obstructive sleep apnea is at present not sufficiently substantiated.

BACKGROUND: We present a handheld integrated device based on a novel magnetic-optical technology for the sensitive detection of cardiactroponin I, a biomarker for the positive diagnosis of myocardial infarct, in a finger-prick blood sample. The test can be performed with a turn-around time of 5 minutes and can detect concentrations in the picomolar range in a sample volume of less than 25 &#61549;L. METHOD: The test was completed in a compact, injection molded plasticdisposable with a 0.5 &#61549;L assay chamber containing integrateddry magnetic nanoparticles and reagents. We have developed a 1-stepassay in which all reaction processes were precisely controlled bythe 3 electromagnetic coils positioned above and below the disposable. In the on-sensor assay, troponin molecules were sandwiched between capture antibodies attached to the detection surface and 500 nm superparamagnetic particles functionalized with tracer anti-troponinantibodies. Nanoparticles bound to the sensor surface were sensitively detected using the optical technique of frustrated total internalreflection (f-TIR). RESULT: A calibration function measured in 100%human plasma using our integrated test demonstrates a limit of detection (mean of blank plus 3 SD) of 0.03 ng/mL (1 pM) cTnI for a turn-around time of approximately 5 minutes. A linear regression analysis of the region 0.03-6.5 ng/mL yields a slope of 37 ± 4, intercept of 0.22±0.01 and linear correlation coefficient of R2=0.98. The measuring range could be extended substantially, to 100 ng/mL, by simultaneously imaging a second spot with lower capture antibody concentration. CONCLUSION: The combination of magnetic particles and their actuation with electromagnets permits the sensitive detection of cTnIto be performed rapidly. Because of the speed, ease-of-use and highanalytical sensitivity of the test, it is well suited for demandingpoint-of-care medical diagnostic applications.

The possibility of automatic understanding of customers' shopping behavior and acting according to their needs is relevant in the marketing domain, attracting a lot of attention lately. In this work, we focus on the task of automatic assessment of customers' shoppingbehavior, by proposing a multi- level framework. The framework is supported at low-level by different types of cameras, which are synchronized, facilitating effcient processing of information. A fish-eyecamera is used for tracking, while a high-definition one serves forthe action recognition task. The experiments are performed on both laboratory and real-life recordings in a supermarket. From the videorecordings, we extract features related to the spatio-temporal behavior of trajectories, the dynamics and the time spent in each region of interest (ROI) in the shop and regarding the customer-products interaction patterns. Next we analyze the shopping sequences usinga Hidden Markov Model (HMM). We conclude that it is possible to accurately classify trajectories (93%), discriminate between different shopping related actions (91.6%), and recognize shopping behavioral types by means of the proposed reasoning model in 95% of the cases.

We demonstrate a new probing technique to measure physicochemical interactions between particles and a substrate in a fluid. The technique is based on the measurement of field-induced rotation of individual magnetic particles in contact with the substrate. The parallel measurement of many particles with single-particle resolution gives reliable statistics in a short time. Using this technique, the interaction between streptavidin-coated magnetic particles and a glass substrate is measured for various buffer conditions. Increased binding is quantified for increasing ionic strength and decreasing pH. The results are found to be in agreement with calculations of electrostatic interactions. We also apply the technique to study how binding isreduced by blocking the substrate with albumin.

Abstract: An overview of the scientific literature in the medical field tells us that a large part of electronic nose applications is devoted to breath analysis. A network based study can help testing thevalidity of this strategy when using many different devices based on identical or different technologies, in view of a use in real clinical practice. The first step is the instrument calibration using aset of key-compounds. In this work a gas sensor array based on Quartz Micro Balance (QMB) transducers functionalized with metallo-porphyrins (ROTV e-nose), and a Cyranose are used simultaneously in a calibration experiment with three ad hoc selected compounds: ethanol, hexane and ethyl acetate, at concentration levels around 1 ppm. Thesetests have demonstrated that LODs down to tens of ppb are possible.Moreover, a mapping between the two instruments has been performed through the calculation of a model based on Cyranose data, and applied to the ROTV e-nose data, for the prediction of compound concentrations. This test has shown a good ability in concentrations prediction, with an error lower than 10 ppb.

Real Time Array PCR is a recently developed biochemical technique that measures amplification curves (like quantitative real time Polymerase Chain Reaction (qPCR)) of a multitude of different templates ina sample. It combines two different techniques to profit from theadvantages of both techniques, namely qPCR (real time quantitative detection) with microarrays (high multiplex capability). This enablesthe quantitative detection of many more target sequences than can be done by qPCR. Thereby, the concentration of the many different target molecules originally present in a sample can be measured. Labeled primers are used that are first elongated to form labeled amplicons in the bulk and these can hybridize to capture probes immobilizedon the surface of the microarray. During each PCR cycle, there is atime window available during which the formed labeled amplicons canhybridize to the target sequences on the microarray surface. By detection of the fluorescence of the spots on the microarray, amplification curves comparable to real time PCR can be obtained, which can be used to deduce the information needed on the presence and the amount of targets originally present in the sample. We present a mathematical model that provides fundamental insights in the different steps of Real Time Array PCR and that can be used to optimize the different biochemical processes taking place. At the microarray surface specific molecules are captured and taken away from the solution, causing a concentration gradient that powers a material flow towards themicroarray surface. Only the labeled strand of the amplicon is captured by the probes on the microarray surface and as a result locallythe PCR process is not symmetric anymore. Moreover, in course of the process more and more ssDNA renatures, leaving relatively less strands and complexes available for hybridization. We found that to a large extent, however, the surface fluorescence scales with the bulkconcentration. Important parameters to optimize are the enzyme concentration and degradation, the primer concentration and the capture probe decay rate. Also the surface hybridization time is critical since the time to reach a steady state is at least one order of magnitude longer compared to the timing of the bulk processes in qPCR.