After the performance of a pilot study of 28 W.W. II veterans who had all had lumbar intervertebral fusion at least two years previously, an in-depth retrospective evaluation was performed on one hundred (100) Workmen's Compensation Board patients, all with a similar follow-up period subsequent to lumbar intervertebral fusion. All patient characteristics were evaluated, including historical, social, physical, radiological and psychological parameters. A success index was derived by factor analysis of twenty (20) variables all representing various aspects of the patients current functional status. This was used as a basis to analyze the remaining data by multiple correlations. These correlations were studied and 55 highly "success"-correlated variables were selected for further factor analysis. This factor analysis identified eight (8) factors closely associated with success or failure of the fusion operation. In order of importance they were identified thus: "Normally"-functioning lumbar spine. Mobility of body. Freedom from neuroticism. High pain tolerance. Minimum number of surgical operations. Freedom from persistent nerve root deficit. Optimism, ambition. General health and fitness. These eight orthogonal (totally non-correlating) factors alone independently accounted for approximately 80% of the variance of "success" as determined by the statistical index. It is felt that certain of these factors may be etiological in connection with the low back problem and prospective work is necessary to shed further light on this. The methods used in patient evaluation were examined and it is felt that certain of these measures are inadequately objective and involve measurement of several factors at one time. For example, measures of range of movement as conventionally used in orthopaedics were found to show a very high correlation with certain of the psychological measures of abnormality. Some showed poor correlation with age which is contrary to published data on the range of motion in different planes determined radiologically. It is suggested that alternative methods be used which would provide greater specificity. Increased use of quantitative radiological methods is suggested as this provides an accurate way of comparing patients from a purely structural viewpoint. The psychological profile of the patient is shown to be of great importance as twothirds of the patients showed an elevation of one or more Minnesota Multiphasic Personality Inventory scales over a T-score of 70; in a random population this figure would be approximately 3%. It is suggested that psychological evaluation should be used at all stages of low back pain disease to assist in management of the patient. Pseudarthrosis was determined to exist in the presence of two out of three radiological features: Movement at the fusion site on bending x-rays. Presence of a defect in the bone mass. Persistence of the posterior apophyseal joints. The presence of pseudarthrosis did not correlate significantly with success or failure of the operation. It is stressed that only information obtainable in an accurate way of all patients was used in this study and this prevented the use of operative reports, from the patients time of surgery which was frequently inaccurate and was not uniform in the information they provided. It was decided not to use symptomatic information as remembered by the patient from before his operation as this would be coloured by too many subsequent events and would not be matched by comparable physical, radiological or psychometric results. Similarly, success was not correlated with the individual surgeon as some surgeons were represented by too few cases and some patients would have been seen by several surgeonsaat different stages in their illness. This evaluation of the clinical and statistical methods provides much information of use in the clinical and prospective experimental setting. It does not attempt to provide definitive answers regarding the causes of low back pain: prospective investigation is required for this, and the nature of the required studies for this is outlined. / Surgery, Department of / Medicine, Faculty of / Graduate

Dans cette thèse, nous étudions la segmentation d'un flux audio en parole, musique et parole sur musique (P/M). Cette étape est fondamentale pour toute application basée sur la transcription automatique de flux radiophoniques et plus généralement multimédias. L'application visée ici est un système de détection de mots clés dans les émissions radiophoniques. Les performances de ce système dépendront de la bonne segmentation du signal fournie par le système de discrimination parole/musique. En effet, une mauvaise classification du signal peut provoquer des omissions de mots clés ou des fausses alarmes. Afin d'améliorer la discrimination parole/musique, nous proposons une nouvelle méthode de paramétrisation du signal. Nous utilisons la décomposition en ondelettes qui permet une analyse des signaux non stationnaires dont la musique est un exemple. Nous calculons différentes énergies sur les coefficients d'ondelettes pour construire nos vecteurs de paramètres. Le signal est alors segmenté en quatre classes : parole (P), non-parole (NP), musique (M) et non-musique (NM) grâce à deux systèmes disjoints de classification HMM classe/non-classe. Cette architecture a été choisie car elle permet de trouver les meilleurs paramètres indépendamment pour chaque tâche P/NP et M/NM. Une fusion des sorties des classifieurs est alors effectuée pour obtenir la décision finale : parole, musique ou parole sur musique. Les résultats obtenus sur un corpus réel d'émissions de radio montrent que notre paramétrisation en ondelettes apporte une nette amélioration des performances en discrimination M/NM et P/M par rapport à la paramétrisation de référence fondée sur les coefficients cepstraux. / In this thesis, we study the segmentation of an audio stream in speech, music and speech on music (S/M). This is a fundamental step for all application based on automatic transcription of radiophonic stream and most commonly multimedia. The target application here is a keyword detection system in broadcast programs. The application performance depends on the quality of the signal segmentation given by the speech/music discrimination system. Indeed, bad signal classification can give miss-detections or false alarms. To improve the speech/music discrimination task, we propose a new signal parameterization method. We use the wavelet decomposition which allows an analysis of non-stationary signal like music for instance. We compute different energies on wavelet coefficients to construct our feature vectors. The signal is then segmented in four classes : speech (S), non-speech (NS), music (M) and non-music (NM), thanks to two apart class/non-class classification systems. These classification systems are based on HMM. We chose a class/non-class architecture because it allows to find independently the best parameters for each S/NS and P/NP tasks. A fusion of the classifier ouputs is then performed to obtain the final decision : speech, music or speech on music. The obtained results on a real broadcast program corpus show that our wavelet-based parameterization gives a significant improvement in performance in both M/NM and S/M discrimination tasks compared to the baseline parameterization using cepstral coefficients.

No description available.

<p>Mergers between companies has become more and more common in Sweden and in the rest of the world with the motive to increase the competitive strength in the current markets and has an important role within organisational development of companies. When organisations merger there is not only buildings and objects that are put together but also their individual organisational structures, people, values and cultures. This can be a difficult process and can affect many people in how they think conscious as well as unconscious, feel and react.</p><p>The purpose of this thesis is to gain a deeper understanding of how employees are affected following a merger when changes in the organisational culture come up, and how these changes are expressed.</p><p>This thesis is delimited to investigate only one company, Skruf Snus AB and only focus on their business and employees in the region of Stockholm. The thesis contains a qualitative investigation with semi structured interviews that has been carried out with eight employees at Skruf Snus AB. The thesis is partly based on a model of Schein which describes the different levels in organisational culture and partly on some theories about reactions to change in relation to the integration between the employees following a merger.</p><p>The results of the investigation indicate that with a merger the whole organisational culture is affected and therefore also the employees. Routines, communication, loyalty and relations have been changed. A gain of stress, anxiety, sadness and disappointment appears between the employees. This is often a result of lack of information and indistinct directives. Concepts like “Us and Them” are created, both unconscious and conscious and are permeated through the entire organisation.</p>

Mergers between companies has become more and more common in Sweden and in the rest of the world with the motive to increase the competitive strength in the current markets and has an important role within organisational development of companies. When organisations merger there is not only buildings and objects that are put together but also their individual organisational structures, people, values and cultures. This can be a difficult process and can affect many people in how they think conscious as well as unconscious, feel and react. The purpose of this thesis is to gain a deeper understanding of how employees are affected following a merger when changes in the organisational culture come up, and how these changes are expressed. This thesis is delimited to investigate only one company, Skruf Snus AB and only focus on their business and employees in the region of Stockholm. The thesis contains a qualitative investigation with semi structured interviews that has been carried out with eight employees at Skruf Snus AB. The thesis is partly based on a model of Schein which describes the different levels in organisational culture and partly on some theories about reactions to change in relation to the integration between the employees following a merger. The results of the investigation indicate that with a merger the whole organisational culture is affected and therefore also the employees. Routines, communication, loyalty and relations have been changed. A gain of stress, anxiety, sadness and disappointment appears between the employees. This is often a result of lack of information and indistinct directives. Concepts like “Us and Them” are created, both unconscious and conscious and are permeated through the entire organisation.

This research project centres on advancing data analysis techniques using the Magnetic Proton Recoil Upgrade Neutron Spectrometer (MPRu) for neutron emission spectroscopy during the deuterium tritium experimental campaign (DTE2) at the Joint European Torus (JET). The study aimed to address three pivotal questions, each with implications for optimizing data accuracy, quality, and utility. The first question focused on determining the optimal short-gate settings for the MPRu. Extensive analysis revealed that conventional metrics, such as the Full Width at Half Maximum (FWHM) and spatial positioning of the proton island, were inadequate for precise short-gate configuration. It was concluded that the existing settings, characterized by a +30 offset from the signal onset, proved to be sufficient. This choice was driven by the necessity to distinguish between escape events, shadow events, and the proton island when plotting signals in a 2-D graph, proving effective across all detector channels. Accurate proton counting hinges on the precise assessment of protons within the escape event region, a task complicated by the interference of background events. This study investigated the point at which the accuracy of the escape event region diminishes by comparing the relative count with simulated data. Results demonstrated that higher-energy signals, positioned farther from the background-concentrated origin, yielded more accurate counts. Additionally, a correction factor based on simulated data is suggested for the unaccounted proton signals. The third question explored was the feasibility of modelling the proton island's location based on proton energy and the characteristics of the phoswhich scintillator detector. While initially promising, the model showed of limited use. The biggest limiting factor was the inconsistencies that originate in the detector themselves. It is not possible to account for the unique characteristics of each single detector, using the methods developed here. This could be changed if the individual characteristics of the detectors are taken into account in a future analysis.

No description available.

No description available.

No description available.

No description available.