Nonlinear prediction based on score function

The linear prediction coding of speech is based in the assumption that the generation model is autoregresive. In this paper we propose a structure to cope with the nonlinear effects presents in the generation of the speech signal. This structure will consist of two stages, the first one will be a classical linear prediction filter, and the second one will model the residual signal by means of two nonlinearities between a linear filter. The coefficients of this filter are computed by means of a gradient search on the score function. This is done in order to deal with the fact that the probability distribution of the residual signal still is not gaussian. This fact is taken into account when the coefficients are computed by a ML estimate. The algorithm based on the minimization of a high-order statistics criterion, uses on-line estimation of the residue statistics and is based on blind deconvolution of Wiener systems [1]. Improvements in the experimental results with speech signals emphasize on the interest of this approach.

Gene filtering with optimal threshold selection

Gene filtering is a useful preprocessing technique often applied to microarray datasets. However, it is no common practice because clear guidelines are lacking and it bears the risk of excluding some potentially relevant genes. In this work, we propose to model microarray data as a mixture of two Gaussian distributions that will allow us to obtain an optimal filter threshold in terms of the gene expression level.

Improving Pitch Tracking Performance in Hard Noise Conditions by a Preprocessing Based on Mathematical Morphology

In this paper we show how a nonlinear preprocessing of speech signal -with high noise- based on morphological filters improves the performance of robust algorithms for pitch tracking (RAPT). This result happens for a very simple morphological filter. More sophisticated ones could even improve such results. Mathematical morphology is widely used in image processing and has a great amount of applications. Almost all its formulations derived in the two-dimensional framework are easily reformulated to be adapted to one-dimensional context

Aluminium fouling and its impact on activated carbon based dioxin and furan removal in a wet scrubber at a co-incineration plant

Dioxins and furans, PCDD/Fs, are highly toxic substances formed in post combustion zones in furnaces. PCDD/F emissions are regulated by a waste incineration directive which relates also to co-incineration plants. Several observations of dioxin and furan enhancements in wet scrub- bers have been reported previously. This is thought to be due to the so-called "memory effect" which occurs when dioxins and furans absorb into plastic material in scrubbers and desorb when ambient circumstances alter significantly. At the co-incineration plant involved, dioxins and furans are controlled with a wet scrubber, the tower packing of which is made of plastic in which activated carbon particles are dispersed. This should avoid the memory effect and act as a dioxin and furan sink since dioxins and furans are absorbed irreversibly into the packing ma- terial. In this case, the tower packing in the scrubber is covered with a white layer that has been found to be mainly aluminium. The aim of this thesis was to determine the aluminium balance and the dioxin and furan behaviour in the scrubber and, thus, the impacts that the foul- ing has on dioxin and furan removal. The source of aluminium, reasons for fouling and further actions to minimize its impacts on dioxin and furan removal were also to be discovered. Measurements in various media around the scrubber and in fuels were made and a profile analysis of PCDD/F and mass balance calculations were carried out. PCDD/F content de- creased in the scrubber. The reduced PCDD/F was not discharged into scrubbing water. The removal mechanism seems to work in spite of the fouling, at least with low PCDD/F loads. Most of the PCDD/F in excess water originates from the Kymijoki River which is used as feeding water in the scrubber. Fouling turned out to consist mainly of aluminium hydroxides. Sludge combusted in the furnace was found to be a significant source of aluminium. Ways to minimize the fouling would be adjustment of pH to a proper lever, installation of a mechanical filter to catch the loose material from the scrubbing water and affecting the aluminium content of the sludge.

Lietteenkäsittelyn toimintaan vaikuttavien muutosten hallinta

Object of the Master’s thesis was to obtain clarification to problems related to sludge treatment at the waste water treatment plant at Stora Enso Varkaus Mill. From time to time these problems have caused emissions to exceed given limits. Case was studied by examining existing data, fiber length fractions and experimental methods. Changes at the Mill have reduced total solids emissions. At the same time the requirement for tertiary treatment has grown. Treatment of tertiary sludge is hard. In the future emission limits shall tighten and the urge for tertiary treatment will grow. The significance of thesis’s results may have a great impact to Stora Enso Varkaus Mill in the future. The results give valuable information to follow-up research and guidelines to sludge treatment. The results encourage observing researched matters at longer period of time. Future recommendations emphasize the meaning of maintenance and systematic, appropriate experiments.

Pain-evoked alterations on gingival blood flow and gingival crevicular fluid (GCF) neuropeptide SP and collagenase-2 (MMP-8) levels

Previous studies have demonstrated that clinical pulpal pain can induce the expression of pro-inflammatory neuropeptides in the adjacent gingival crevice fluid (GCF). Vasoactive agents such as substance P (SP) are known to contribute to the inflammatory type of pain and are associated with increased blood flow. More recent animal studies have shown that application of capsaicin on alveolar mucosa provokes pain and neurogenic vasodilatation in the adjacent gingiva. Pain-associated inflammatory reactions may initiate expression of several pro- and anti-inflammatory mediators. Collagenase-2 (MMP-8) has been considered to be the major destructive protease, especially in the periodontitis-affected gingival crevice fluid (GCF). MMP-8 originates mostly from neutrophil leukocytes, the first line of defence cells that exist abundantly in GCF, especially in inflammation. With this background, we wished to clarify the spatial extensions and differences between tooth-pain stimulation and capsaicin-induced neurogenic vasodilatation in human gingiva. Experiments were carried out to study whether tooth stimulation and capsaicin stimulation of alveolar mucosa would induce changes in GCF MMP-8 levels and whether tooth stimulation would release neuropeptide SP in GCF. The experiments were carried out on healthy human volunteers. During the experiments, moderate and high intensity painful tooth stimulation was performed by a constant current tooth stimulator. Moderate tooth stimulation activates A-delta fibres, while high stimulation also activates C-fibres. Painful stimulation of the gingiva was achieved by topical application of capsaicin-moistened filter paper on the mucosal surface. Capsaicin is known to activate selectively nociceptive C-fibres of stimulated tissue. Pain-evoked vasoactive changes in gingivomucosal tissues were mapped by laser Doppler imaging (LDI), which is a sophisticated and non-invasive method for studying e.g. spatial and temporal characteristics of pain- and inflammation-evoked blood flow changes in gingivomucosal tissues. Pain-evoked release of MMP-8 in GCF samples was studied by immunofluorometric assay (IFMA) and Western immunoblotting. The SP levels in GCF were analysed by Enzyme immunoassay (EIA). During the experiments, subjective stimulus-evoked pain responses were determined by a visual analogue pain scale. Unilateral stimulation of alveolar mucosa and attached gingiva by capsaicin evoked a distinct neurogenic vasodilatation in the ipsilateral gingiva, which attenuated rapidly at the midline. Capsaicin stimulation of alveolar mucosa provoked clear inflammatory reactions. In contrast to capsaicin stimuli, tooth stimulation produced symmetrical vasodilatations bilaterally in the gingiva. The ipsilateral responses were significantly smaller during tooth stimulation than during capsaicin stimuli. The current finding – that tooth stimulation evokes bilateral vasodilatation while capsaicin stimulation of the gingiva mainly produces unilateral vasodilatation – emphasises the usefulness of LDI in clarifying spatial features of neurogenic vasoactive changes in the intra-oral tissues. Capsaicin stimulation of the alveolar mucosa induced significant elevations in MMP-8 levels and activation in GCF of the adjacent teeth. During the experiments, no marked changes occurred in MMP-8 levels in the GCF of distantly located teeth. Painful stimulation of the upper incisor provoked elevations in GCF MMP-8 and SP levels of the stimulated tooth. The GCF MMP-8 and SP levels of the non-stimulated teeth were not changed. These results suggest that capsaicin-induced inflammatory reactions in gingivomucosal tissues do not cross the midline in the anterior maxilla. The enhanced reaction found during stimulation of alveolar mucosa indicates that alveolar mucosa is more sensitive to chemical irritants than the attached gingiva. Analysis of these data suggests that capsaicin-evoked neurogenic inflammation in the gingiva can trigger the expression and activation of MMP-8 in GCF of the adjacent teeth. In this study, it is concluded that experimental tooth pain at C-fibre intensity can induce local elevations in MMP-8 and SP levels in GCF. Depending on the role of MMP-8 in inflammation, in addition to surrogated tissue destruction, the elevated MMP-8 in GCF may also reflect accelerated local defensive and anti-inflammatory reactions.

Prévention des troubles du rythme cardiaque en hémodialyse: quels sont les facteurs modulables [Prevention of cardiac arrhythmias in hemodialysis: what are the modulating factors?].

La mort subite est la première cause de mortalité chez les patients souffrant d'une insuffisance rénale terminale traités par dialyse chronique. La technique de dialyse utilisée et la composition chimique du dialysat influencent l'incidence des arythmies. Des études pilotes démontrent que l'utilisation d'un dialysat sans acétate avec perfusion de bicarbonate de sodium en aval du filtre de dialyse, couplée à une modulation du profil de potassium pendant la séance de dialyse, ou acetate free biofiltration with potassium profiled dialysate, permet de réduire l'incidence des arythmies, l'intervalle QT et sa dispersion. La limitation du volume de soustraction liquidienne pendant la dialyse et l'augmentation de la concentration de calcium dans le dialysat constituent d'autres stratégies anti-arythmogènes possibles Sudden death is the first cause of mortality in patients with end stage renal disease undergoing chronic dialysis treatment. The technique of dialysis as well as the chemical composition of the dialysate can impact on the incidence of cardiac arrhythmias. Pilot studies reveal that the use of an acetate-free dialysate with a downstream filter infusion of sodium bicarbonate, coupled with a modulated potassium-profiled dialysate during hemodialysis, or acetate free biofiltration with potassium profiled dialysate, reduces the incidence of arrhythmias, the QT interval and QT dispersion. The limitation of the ultrafiltration volume during the dialysis session, and the increase in calcium concentration in the dialysate are other possible strategies to reduce cardiac arrhythmias.

On the Analysis and Control of a Linear Synchronous Servomotor with a Flexible Load

Industry's growing need for higher productivity is placing new demands on mechanisms connected with electrical motors, because these can easily lead to vibration problems due to fast dynamics. Furthermore, the nonlinear effects caused by a motor frequently reduce servo stability, which diminishes the controller's ability to predict and maintain speed. Hence, the flexibility of a mechanism and its control has become an important area of research. The basic approach in control system engineering is to assume that the mechanism connected to a motor is rigid, so that vibrations in the tool mechanism, reel, gripper or any apparatus connected to the motor are not taken into account. This might reduce the ability of the machine system to carry out its assignment and shorten the lifetime of the equipment. Nonetheless, it is usually more important to know how the mechanism, or in other words the load on the motor, behaves. A nonlinear load control method for a permanent magnet linear synchronous motor is developed and implemented in the thesis. The purpose of the controller is to track a flexible load to the desired velocity reference as fast as possible and without awkward oscillations. The control method is based on an adaptive backstepping algorithm with its stability ensured by the Lyapunov stability theorem. As a reference controller for the backstepping method, a hybrid neural controller is introduced in which the linear motor itself is controlled by a conventional PI velocity controller and the vibration of the associated flexible mechanism is suppressed from an outer control loop using a compensation signal from a multilayer perceptron network. To avoid the local minimum problem entailed in neural networks, the initial weights are searched for offline by means of a differential evolution algorithm. The states of a mechanical system for controllers are estimated using the Kalman filter. The theoretical results obtained from the control design are validated with the lumped mass model for a mechanism. Generalization of the mechanism allows the methods derived here to be widely implemented in machine automation. The control algorithms are first designed in a specially introduced nonlinear simulation model and then implemented in the physical linear motor using a DSP (Digital Signal Processor) application. The measurements prove that both controllers are capable of suppressing vibration, but that the backstepping method is superior to others due to its accuracy of response and stability properties.

Cigarette et affiche publicitaire: histoire d'une liaison dangereuse durant les Trente Glorieuses

The advertising poster's main characteristic is the ability to convey a commercial message quickly and publicly thanks to its straightforward image and text. The young people, being the tobacco industry's principal target, are particularly exposed to these messages. This kind of advertisement becomes a mean of counterstroke as soon as the cigarette's harmfulness is acknowledged. Some of the cigarette industry's strategies can be revealed by the historical analysis of a 253 posters corpus selected among the main cigarette manufacturers in Switzerland at the time of the post-war economic boom. With the misuse of sport's theme, the overvaluation of the filter's efficiency, the use of a vocabulary that implies lightness and by erasing the image of smoke in its advertisement, the industry tries to reassure the smoker wrongly.

Derivation of an observer model adapted to irregular signals based on convolution channels.

Anthropomorphic model observers are mathe- matical algorithms which are applied to images with the ultimate goal of predicting human signal detection and classification accuracy across varieties of backgrounds, image acquisitions and display conditions. A limitation of current channelized model observers is their inability to handle irregularly-shaped signals, which are common in clinical images, without a high number of directional channels. Here, we derive a new linear model observer based on convolution channels which we refer to as the "Filtered Channel observer" (FCO), as an extension of the channelized Hotelling observer (CHO) and the nonprewhitening with an eye filter (NPWE) observer. In analogy to the CHO, this linear model observer can take the form of a single template with an external noise term. To compare with human observers, we tested signals with irregular and asymmetrical shapes spanning the size of lesions down to those of microcalfications in 4-AFC breast tomosynthesis detection tasks, with three different contrasts for each case. Whereas humans uniformly outperformed conventional CHOs, the FCO observer outperformed humans for every signal with only one exception. Additive internal noise in the models allowed us to degrade model performance and match human performance. We could not match all the human performances with a model with a single internal noise component for all signal shape, size and contrast conditions. This suggests that either the internal noise might vary across signals or that the model cannot entirely capture the human detection strategy. However, the FCO model offers an efficient way to apprehend human observer performance for a non-symmetric signal.

Microtubule Associated Protein 1b (MAP1B) Is a Marker of the Microtubular Cytoskeleton in Podocytes but Is Not Essential for the Function of the Kidney Filtration Barrier in Mice.

Podocytes are essential for the function of the kidney glomerular filter. A highly differentiated cytoskeleton is requisite for their integrity. Although much knowledge has been gained on the organization of cortical actin networks in podocyte's foot processes, less is known about the molecular organization of the microtubular cytoskeleton in primary processes and the cell body. To gain an insight into the organization of the microtubular cytoskeleton of the podocyte, we systematically analyzed the expression of microtubule associated proteins (Maps), a family of microtubules interacting proteins with known functions as regulator, scaffold and guidance proteins. We identified microtubule associated protein 1b (MAP1B) to be specifically enriched in podocytes in human and rodent kidney. Using immunogold labeling in electron microscopy, we were able to demonstrate an enrichment of MAP1B in primary processes. A similar association of MAP1B with the microtubule cytoskeleton was detected in cultured podocytes. Subcellular distribution of MAP1B HC and LC1 was analyzed using a double fluorescent reporter MAP1B fusion protein. Subsequently we analyzed mice constitutively depleted of MAP1B. Interestingly, MAP1B KO was not associated with any functional or structural alterations pointing towards a redundancy of MAP proteins in podocytes. In summary, we established MAP1B as a specific marker protein of the podocyte microtubular cytoskeleton.

Teollisuussuodattimen kehikon hitsauskiinnittimen suunnittelu robottihitsausasemaan Larox Oy:lle

The purpose of this work is to support the company in the designing of a welding clamp and to switch partly from manual welding to automatic welding. The product is a stainless steel frame of an industrial filter. This work includes the theory and the case sections. The theory covers the half of the work and studies significant aspects of designing of a welding clamp. The case reveals few models for a welding clamp for the company's needs. The study of the initial conditions and possible changes to existing product line are also included.

Jännitteen reunan hallinta taajuusmuuttajissa

Diplomityössä perehdyttiin taajuusmuuttajien toimintaan ja ohjaukseen. Lisäksi työssä tarkasteltiin vaihtosuuntaajan nopeiden transienttitilojen aiheuttamaa moottorin ylijännitettä. Moottorikaapelin heijastuksia käsiteltiin vertaamalla moottorikaapelia siirtolinjaan ja todennettiin ylijännitteen syyt. Ylijännitteen vähentämiseksi on kehitetty useita suodatusmenetelmiä. Työssä vertailtiin näitä menetelmiä ja kartoitettiin kaupallisia vaihtoehtoja. Taajuusmuuttajan ohjaus on tähän päivään asti tehty yleensä käyttäen mikroprosessoria sekä logiikkapiiriä. Tulevaisuudessa ohjaukseen käytetään todennäköisesti uudelleenohjelmoitavia FPGA-piirejä (Field Programmable Gate Array). FPGA-piirin etuihin kuuluu uudelleenohjelmoitavuus sekä ohjauksen keskittäminen yhdelle piirille.

Deinking flotation of recycled linerboard for food packaging applications

Brown packaging linerboard, made entirely from recovered pulp, was subjected to deinking flotation for evaluating the possible improvements in its chemical, optical and mechanical properties. The increase in the rate of recovered paper utilisation, along with the tendency towards lower basis weights, in the packaging paper production, has created a growing need for the utilisation of secondary fibers of improved quality. To attain better quality fibers, flotation deinking of brown grades is being considered, along with the addition of primary fibers to recovered paper furnish. Numerous conducted studies, in which the flotation technology was used in the treatment of brown grades, support this idea. Most of them show that the quality of fibers is improved after flotation deinking, resulting in higher mechanical properties of the deinked handsheets and in lower amounts of chemical contaminants. As to food and human health safety, packaging paper has to meet specific requirements, to be classified as suitable for its direct contact with foods. Recycled paper and board may contain many potential contaminants, which, especially in the case of direct food contact, may migrate from packaging materials into foodstuffs. In this work, the linerboard sample selected for deinking was made from recycled fibers not submitted previously to chemical deinking flotation. Therefore, the original sample contained many noncellulosic components, as well as the residues of printing inks. The studied linerboardsample was a type of packaging paper used for contact with food products that are usually peeled before use, e.g. fruits and vegetables. The decrease in the amount of chemical contaminants, after conducting deinking flotation, was evaluated, along with the changes in the mechanical and optical properties of the deinked handsheets. Food contact analysis was done on both the original paper samples and the filter pads and handsheets made before and after deinking flotation. Food contact analysis consisted of migration tests of brightening agents, colorants, PCPs, formaldehydes and metals. Microbiological tests were also performed to determine the possible transfer of antimicrobial constituents

Filtration of effluents for microirrigation systems

Clogging, measured through head loss across filters, and the filtration quality of different filters using different effluents were studied. The filters used were: 115, 130, and 200 m disc filters; 98, 115, 130, and 178 m screen filters; and a sand filter filled with a single layer of sand with an effective diameter of 0.65 mm. The filters were used with a meat industry effluent and secondary and tertiary effluents of two wastewater treatment plants. It was observed that clogging depended on the type of effluent. With the meat industry effluent, the poorest quality effluent, disc filters clogged more than the other filter types. When the wastewater treatment plant effluents were used, the disc filters showed less frequent clogging. Several physical and chemical parameters, such as total suspended solids, chemical oxygen demand, turbidity, electrical conductivity, pH, and number of particles, were analyzed in the effluents at the entry and exit points of the filters. In general, filters did not reduce the values of the main clogging parameters to a great degree. It was found that the parameter that explained the clogging, expressed as Boucher’s filterability index, was different depending on the type of effluent and filter. The best quality of filtration was achieved with a sand filter when the meat industry effluent was used. No significant differences were observed between the quality of filtration of disc and screen filters when operating with the secondary and tertiary effluents