The Exotic Palaeo-Thethys Terrane in Central Iran: New geological data from Anarak, Jandaq and Posht-e-Badam areas

Résumé Le « terrane » d'Anarak-Jandak occupe une position géologique clé au nord-ouest du Microcontinent Centre-East Iranien (CE1M), connecté avec le Bloc du Grand Kavir et la ceinture métamorphique de Sanandaj-Sirjan. Nous discutons ici l'origine de ces différentes unités, reliées jusqu'à présent à des épisodes orogéniques d'âge Précambrien à Paléozoïque inférieur, pour conclure finalement de leur affinité paléotéthysienne. Leur histoire commence par un épisode de rifting d'âge Ordovicien supérieur-Dévonien inférieur, pour se terminer au Trias par la collision des blocs Cimmériens dérivé du Gondwana avec le Bloc du Turan d'affinité asiatique (événement Eocimmérien). La plus importante unité métamorphique affleurant au sud-ouest de la région de Jandak-Anarak-Kaboudan est une épaisse séquence silicoclastique à grains fins contenant des blocs ophiolitiques (marginal-sea-type), et des associations basalte-gabbro à signatures géochimiques de type supra-subduction. Dans la région de Nakhlak, nous avons daté ces gabbros par la méthode U-Pb à 387f0.11 Ma ; les roches métamorphiques pélitiques ont donné des âges de refroidissement Ar-Ar pour la muscovite de 320 à 333 Ma. Ce complexe d'accrétion "varisque" a été métamorphisé dans le faciès schiste vert-amphibolite au cours de l'accrétion de la ceinture granitique d'Airekan, d'âge Cambrien inférieur (549±15 Ma par la méthode U/Pb), qui affleure aujourd'hui à l'extrémité nord-ouest du terrane d'Anarak-Jandak . La subduction vers le nord de l'océan Paléotéthys depuis le Paléazoïque supérieur jusqu'au Trias, a permis l'accumulation de grandes quantités de matériel océanique dans la zone de subduction. Par exemple, une succession de guyots (Anarak, Kaboudan, et Meraji Seamounts) et de hauts sous-marins, entrés en collision oblique avec le prisme d'accrétion, est à l'origine d'un léger métamorphisme de type HP qui affecte ces séries {âges Ar-Ar de 280 à 230 Ma). De plus, le magmatisme bimodal de Chah Gorbeh est caractérisé d'une part par des roches de type trondjémite-gabbros (262 Ma), d'autre part par des laves en coussin de type basaltes alcalins-rhyolites; ces roches magmatiques ont recoupé l'ophiolite d'Anarak lors de la mise en place de cette dernière dans la fosse interne de subduction. Quant au prisme d'accrétion de Doshakh, d'âge essentiellement Permien supérieur, i1 a été accrété le long de la marge continentale et métamorphisé dans le faciès schiste vert. La fermeture de la Paléotéthys s'enregistre finalement par la sédimentation dans le bassin d'avant pays du flysch de Bayazeh, d'âge probable Triasique. Le matériel issu de l'arc magmatique de la Paléotéthys est très bien préservé dans les dépôts infra-arc Dévonien supérieur-Carbonifère de Godar-e-Siah, ainsi que dans la succession d'avant-arc de Nakhlak. Pendant l'intervalle Paléozoïque supérieur-Trias, la région de Jandak a été soumise à un régime extensif de type bassin d'arrière-arc, dont un témoin pourrait être la ceinture ophiolitique d'Arusan, elle-même comparable aux écailles ophiolitiques d'Aghdarband au nord-est de l'Iran. Cet ensemble métamorphique est recoupé par des granites d'arc à collisionnel datés à 215±15 Ma. Dans la région de Yazd, témoin de la marge passive Cimmérienne, la sédimentation syn-rift Silurienne à Dévonienne inférieure a été interrompue pendant l'intervalle Trias moyen-Trias supérieur; il en a été de même pour les dépôts de plate-forme Paléozoïque supérieur. L'érosion, qui dans ce dernier cas a atteint le Permien, pourrait être liée au bombement flexural de la marge passive. La collision finale n'a pas induit de déformations trop importantes, et se caractérise par la mise en place de nappes sur la marge passive. Cet événement est scellé par des dépôts molassique du Lias. D'un point de vue régional, la zone s'étendant actuellement de la Mer Noire au Pamir a été soumise à six épisodes d'extension-compression du Jurassique inférieur (début du l'ouverture en position arrière-arc de la Néotéthys) à l'Eocène moyen. Par exemple, le terrane d'AnarakJandak, probablement situé entre le Kopeh Dagh et la plate-forme nord Afghane, s'est complètement détaché de sa patrie d'origine au début du Crétacé supérieur. Des preuves de cet événement se retrouvent dans les séries de plate-forme de Khur (préservation de séries syn-rift puis de marge passive). Les ophiolites de Nain et de Sabzevar sont de plus interprétée comme un témoin de l'existence de ce bassin d'arrière-arc. Dans l'intervalle Eocène-Oligocène, l'indentation par la plaque indienne de l'Eurasie a été contemporaine de la rotation horaire de fragments de l'ancien microcontinent Iranien et de la formation du CEIM. Cette rotation est responsable du transport du terrane d'Anarak-Jandak vers sa position actuelle en Iran Central, et de la dislocation de Terranes de moindre importance, comme le bloc de Posht-e Badam. Depuis le Miocène supérieur, et à la suite de la collision entre l'Arabie et l'Iran, le ternane d'Anarak-Jandak a subi des déformations liées à l'activité d'une zone de cisaillement dextre parallèle à la suture du Zagros, à l'arrière de l'arc magmatique d'Uromieh-Dokhtar. Résumé large public Le Microcontinent Centre-Est Iranien occupe une position géologique clé au centre de l'Iran. Les différentes unités qui le composent, reliées jusqu'à présent à des épisodes orogéniques d'âge Précambrien à Paléozoïque inférieur, sont maintenant rajeunies et liés à la fermeture de l'océean Paléotéthys. Leur histoire commence par un épisode de rifting d'âge Ordovicien supérieur à Dévonien inférieur, pour se terminer au Trias par la collision des- blocs Cimmériens, dérivés du Gondwana, avec le Bloc du Turan d'affinité asiatique. Dans la marge active asiatique de la Paléotéthys, nous avons daté les restes d'un océan marginal à 387±0.11 Ma. Ce complexe d'accrétion a été métamorphisé au cours de la réaccrétion de la ceinture granitique d'Airekan, d'âge Cambrien inférieur (549±15 Ma), qui affleure aujourd'hui à l'extrémité nord-ouest du « terrane » d'Anarak-Jandak correspondant à la plus grande partie de la région étudiée. Le matériel issu de l'arc magmatique de la Paléotéthys est très bien préservé et daté du Dévonien supérieur-Carbonifère. Pendant l'intervalle Paléozoïque supérieur-Trias, la région a été soumise à un régime extensif de type bassin d'arrière-arc, dont un témoin pourrait être la ceinture ophiolitique d'Arusan, comparable aux écailles ophiolitiques d'Aghdarband au nord-est de l'Iran. Cet ensemble métamorphique est recoupé par des granites datés à 215±15 Ma. La subduction vers le nord de l'océan Paléotéthys depuis le Paléozoïque supérieur jusqu'au Trias, a permis l'accumulation de grandes quantités de matériel océanique dans la zone de subduction. Par exemple, une succession de volcans sous-marins, entrés en collision avec le prisme d'accrétion, est à l'origine d'un léger métamorphisme de type HP qui affecte ces séries (280 à 230 Ma). Quant au prisme d'accrétion de Doshakh, d'âge essentiellement Permien supérieur, il a été mis en place le long de la marge continentale et métamorphisé dans le faciès schiste vert. La fermeture de la Paléotéthys s'enregistre finalement par la sédimentation dans le bassin d'avant pays du flysch de Bayazeh, d'âge Triasique. Dans la région de Yazd, on trouve les témoins de la marge passive Cimmérienne, la sédimentation syn-rift Silurienne à Dévonienne inférieure a été interrompue pendant l'intervalle Trias moyen-Trias supérieur, marqué par la flexuration de la marge passive lorsqu'elle rentra en collision avec la marge active asiatique. Cet événement est scellé par des dépôts molassique à charbon du Lias. Le «terrane» d'Anarak-Jandak, probablement situé à l'origine entre le Kopeh Dagh et la plate-forme nord Afghane, s'est complètement détaché de cette région au début du Crétacé supérieur lors de l'ouverture d'un bassin d'arrière-arc, engendré, cette fois, par la subduction de l'océan Néotéthys situé au sud des blocs cimmériens. Des preuves de cet événement se retrouvent dans les séries syn-rift, puis de marge passive de Khour. Les ophiolites de Nain et de Sabzevar sont interprétées comme un témoin de l'existence de ce bassin d'arrière-arc. Dans l'intervalle Eocène-Oligocène, l'indentation de l'Eurasie par la plaque indienne a été contemporaine de la rotation horaire de fragments de l'ancien microcontinent centre-Iranien. Cette rotation de près de 90° est responsable du transport du « terrane » d'Anarak-Jandak vers sa position actuelle. Abstract The Anarak-Jandaq terrane occupies a strategic geological situation at the north-western part of the Central-East Iranian Microcontinent (CEIM) and in connection with the Great Kavir Block and Sanandaj-Sirjan metamorphic belt. Our recent findings redefine the origin of these mentioned areas so far attributed to the Precambrian-Early Palaeozoic orogenic episodes, to be now directly related to the tectonic evolution of the Palaeo-Tethys Ocean, commenced by Late Ordovician-Early Devonian rifting events and terminated in the Triassic by the Eocimmerian tectonic event due to the collision of the Cimmerian blocks with the Asiatic Turan block. The most distributed metamorphic unit that is exposed from the south-west of Jandaq to the Anarak and Kaboudan areas is a thick and fine grain siliciclastic sequence accompanied by marginal-sea-basin ophiolitic blocks including basalt-gabbro association with supra-subduction-geochemical signature. These gabbros in the Nakhlak area were dated by U/Pb method at 387.6 ± 0.11 Ma and the metamorphic pelitic rocks yielded a range of 320 to 333 Ma muscovite-cooling ages based on 40Ar/39 Ar method. This "Variscan" accretionary complex was metamorphosed in greenschist-amphibolite facies during accretion to the Lower Cambrian Airekan granitic belt (549 ± 15 Ma by U/Pb method) that crops out at the northwestern edge of the Anarak-Jandaq terrane. Continued northward subduction of the Palaeo-Tethys Ocean during the entire Late Palaeozoic-Middle Triassic brought huge amount of oceanic material to the subduction zone. One chain of Carboniferous-Triassic oceanic rises and seamounts (the Anarak, Kaboudan, and Meraji Seamounts) obliquely collided with the accretionary wedge and created a mild HP metamorphic event (280-230 Ma based on 40Ar/39Ar results). Bimodal magmatism of the Chah Gorbeh area is characterized by a 262 Ma trondjemite-gabbro as well as pillow alkalibasalts-rhyolites which intruded the Anarak ophiolite when it was being emplaced within the inner-wall trench. The mainly Late Permian-Triassic Doshakh wedge was accreted along the continent and metamorphosed under lower greenschist facies and the probable Triassic Bayazeh flysch filled the foreland basin during the final closure. The Palaeo-Tethys magmatic arc products have been well preserved in the Late Devonian-Carboniferous Godar-e-Siah intra-arc deposits and the Triassic Nakhlak fore-arc succession. During the Late Palaeozoic-Triassic times, the Jandaq area has been affected by back-arc extension and probably the Arusan ophiolitic belt is the remnant of this narrow basin comparable to the Aqdarband ophiolitic remnant in north-east Iran. This metamorphic belt was intruded by 215 ± 15 Ma arc to collisional granites. In the passive margin of the Cimmerian block, on the Yazd region, the Silurian-Early Devonian syn-rift succession as well as the nearly continuous Upper Palaeozoic platform-type deposition was interrupted during the Middle to Late Triassic time, local erosion down to Devonian levels may be related to flexural bulge erosion. The collision event was not so strong to generate intensive deformation but was accompanied by some nappe thrusting onto the passive margin. It is finally unconformably covered by Liassic continental molassic deposits. Related to the onset of Neo-Tethyan back-arc opening in Early Jurassic to Mid-Eocene times, six periods of extensional-compressional events have differently influenced an elongated area, extending from the West Black Sea to Pamir. The Anarak-Jandaq terrane which was situated somewhere in this affected area, probably between the Kopeh Dagh and North Afghan platform, was completely detached from its source at the beginning of the Late Cretaceous

A modular magnetic anastomotic device for minimally invasive digestive anastomosis: proof of concept and preliminary data in the pig model.

BACKGROUND: The aim of our study was to assess the feasibility of minimally invasive digestive anastomosis using a modular flexible magnetic anastomotic device made up of a set of two flexible chains of magnetic elements. The assembly possesses a non-deployed linear configuration which allows it to be introduced through a dedicated small-sized applicator into the bowel where it takes the deployed form. A centering suture allows the mating between the two parts to be controlled in order to include the viscerotomy between the two magnetic rings and the connected viscera. METHODS AND PROCEDURES: Eight pigs were involved in a 2-week survival experimental study. In five colorectal anastomoses, the proximal device was inserted by a percutaneous endoscopic technique, and the colon was divided below the magnet. The distal magnet was delivered transanally to connect with the proximal magnet. In three jejunojejunostomies, the first magnetic chain was injected in its linear configuration through a small enterotomy. Once delivered, the device self-assembled into a ring shape. A second magnet was injected more distally through the same port. The centering sutures were tied together extracorporeally and, using a knot pusher, magnets were connected. Ex vivo strain testing to determine the compression force delivered by the magnetic device, burst pressure of the anastomosis, and histology were performed. RESULTS: Mean operative time including endoscopy was 69.2 ± 21.9 min, and average time to full patency was 5 days for colorectal anastomosis. Operative times for jejunojejunostomies were 125, 80, and 35 min, respectively. The postoperative period was uneventful. Burst pressure of all anastomoses was ≥ 110 mmHg. Mean strain force to detach the devices was 6.1 ± 0.98 and 12.88 ± 1.34 N in colorectal and jejunojejunal connections, respectively. Pathology showed a mild-to-moderate inflammation score. CONCLUSIONS: The modular magnetic system showed enormous potential to create minimally invasive digestive anastomoses, and may represent an alternative to stapled anastomoses, being easy to deliver, effective, and low cost.

Blind channel estimation and data detection using hidden Markov models theory

In this correspondence, we propose applying the hiddenMarkov models (HMM) theory to the problem of blind channel estimationand data detection. The Baum–Welch (BW) algorithm, which is able toestimate all the parameters of the model, is enriched by introducingsome linear constraints emerging from a linear FIR hypothesis on thechannel. Additionally, a version of the algorithm that is suitable for timevaryingchannels is also presented. Performance is analyzed in a GSMenvironment using standard test channels and is found to be close to thatobtained with a nonblind receiver.

Methods of Telecommunication Technology Forecasting

This master's thesis coversthe concepts of knowledge discovery, data mining and technology forecasting methods in telecommunications. It covers the various aspects of knowledge discoveryin data bases and discusses in detail the methods of data mining and technologyforecasting methods that are used in telecommunications. Main concern in the overall process of this thesis is to emphasize the methods that are being used in technology forecasting for telecommunications and data mining. It tries to answer to some extent to the question of do forecasts create a future? It also describes few difficulties that arise in technology forecasting. This thesis was done as part of my master's studies in Lappeenranta University of Technology.

Lossless compression of hyperspectral images

Technological progress has made a huge amount of data available at increasing spatial and spectral resolutions. Therefore, the compression of hyperspectral data is an area of active research. In somefields, the original quality of a hyperspectral image cannot be compromised andin these cases, lossless compression is mandatory. The main goal of this thesisis to provide improved methods for the lossless compression of hyperspectral images. Both prediction- and transform-based methods are studied. Two kinds of prediction based methods are being studied. In the first method the spectra of a hyperspectral image are first clustered and and an optimized linear predictor is calculated for each cluster. In the second prediction method linear prediction coefficients are not fixed but are recalculated for each pixel. A parallel implementation of the above-mentioned linear prediction method is also presented. Also,two transform-based methods are being presented. Vector Quantization (VQ) was used together with a new coding of the residual image. In addition we have developed a new back end for a compression method utilizing Principal Component Analysis (PCA) and Integer Wavelet Transform (IWT). The performance of the compressionmethods are compared to that of other compression methods. The results show that the proposed linear prediction methods outperform the previous methods. In addition, a novel fast exact nearest-neighbor search method is developed. The search method is used to speed up the Linde-Buzo-Gray (LBG) clustering method.

Wavelet Filter Parameterization for Compression of Spectral Images

Vaatimus kuvatiedon tiivistämisestä on tullut entistä ilmeisemmäksi viimeisen kymmenen vuoden aikana kuvatietoon perustuvien sovellutusten myötä. Nykyisin kiinnitetään erityistä huomiota spektrikuviin, joiden tallettaminen ja siirto vaativat runsaasti levytilaa ja kaistaa. Aallokemuunnos on osoittautunut hyväksi ratkaisuksi häviöllisessä tiedontiivistämisessä. Sen toteutus alikaistakoodauksessa perustuu aallokesuodattimiin ja ongelmana on sopivan aallokesuodattimen valinta erilaisille tiivistettäville kuville. Tässä työssä esitetään katsaus tiivistysmenetelmiin, jotka perustuvat aallokemuunnokseen. Ortogonaalisten suodattimien määritys parametrisoimalla on työn painopisteenä. Työssä todetaan myös kahden erilaisen lähestymistavan samanlaisuus algebrallisten yhtälöiden avulla. Kokeellinen osa sisältää joukon testejä, joilla perustellaan parametrisoinnin tarvetta. Erilaisille kuville tarvitaan erilaisia suodattimia sekä erilaiset tiivistyskertoimet saavutetaan eri suodattimilla. Lopuksi toteutetaan spektrikuvien tiivistys aallokemuunnoksen avulla.

Managing Packet Data (Sessions) in 3G Mobile Networks

Tällä hetkellä kolmannen sukupolven matkapuhelinjärjestelmät ovat siirtyneet kaupalliseen vaiheeseen. Universal Mobile Telecommunication System (UMTS) on eräs kolmannen sukupolven matkapuhelinjärjestelmä, jota tullaan käyttämään Euroopassa. Diplomityön päämääränä on tutkia, kuinka pakettivälitteistä tiedonsiirtoa hallitaan UMTS - verkoissa. Diplomityö antaa yleiskuvan toisen sukupolven matkapuhelinjärjestelmien datapalveluiden kehityksestä kolmannen sukupolven nopeisiin matkapuhelinjärjestelmiin. Pakettivälitteisen verkon verkkoarkkitehtuuri on esitetty sekä sen, diplomityön kannalta, tärkeimpien osien toiminnallisuus on selvitetty. Myös pakettipohjaisten datayhteyksien eli istuntojen muodostaminen ja vapauttaminen sekä aktiivisen yhteyden ominaisuuksien muokkaaminen on esitetty tässä diplomityössä. Yhteydenhallintaprotokolla, Session Management (SM), on yksi protokolla, joka osallistuu pakettidatayhteyden hallintaan. SM -protokolla on käsitelty työssä yksityiskohtaisesti. SM -protokollan SDL toteutus on esitetty diplomityön käytännönosassa

Lossless Compression Methods for Multispectral Images: Development and Comparisons

Main purpose of this thesis is to introduce a new lossless compression algorithm for multispectral images. Proposed algorithm is based on reducing the band ordering problem to the problem of finding a minimum spanning tree in a weighted directed graph, where set of the graph vertices corresponds to multispectral image bands and the arcs’ weights have been computed using a newly invented adaptive linear prediction model. The adaptive prediction model is an extended unification of 2–and 4–neighbour pixel context linear prediction schemes. The algorithm provides individual prediction of each image band using the optimal prediction scheme, defined by the adaptive prediction model and the optimal predicting band suggested by minimum spanning tree. Its efficiency has been compared with respect to the best lossless compression algorithms for multispectral images. Three recently invented algorithms have been considered. Numerical results produced by these algorithms allow concluding that adaptive prediction based algorithm is the best one for lossless compression of multispectral images. Real multispectral data captured from an airplane have been used for the testing.

Cenozoic evolution of eastern Iberia: structural data and dynamic model.

Iberia underwent intraplate deformation during the Mesozoic and Cenozoic. In eastem Ibena, compression took place during the Palaeogene and early Miocene, giving rise to the Iberian Chain, and extension started during the early Miocene in the coastal areas and the Valencia trough; during early Miocene compression continued in the western Iberian Chain whereas extension had started in the eastern Iberian Chain. From the kinematic data obtained from the major compressional and extensional structures formed dunng the Cenozoic, a simple dynamic model using Bott's (1959) formula is presented. The results show that both extension and compression may have been produced assuming a main horizontal stress-axis approximately N-S, in a similar direction that the convergence between Europe, Ibena and Afnca dunng the Cenozoic.

Cost Optimisation of Mobile Advertising Client Data Transfer

Data traffic caused by mobile advertising client software when it is communicating with the network server can be a pain point for many application developers who are considering advertising-funded application distribution, since the cost of the data transfer might scare their users away from using the applications. For the thesis project, a simulation environment was built to mimic the real client-server solution for measuring the data transfer over varying types of connections with different usage scenarios. For optimising data transfer, a few general-purpose compressors and XML-specific compressors were tried for compressing the XML data, and a few protocol optimisations were implemented. For optimising the cost, cache usage was improved and pre-loading was enhanced to use free connections to load the data. The data traffic structure and the various optimisations were analysed, and it was found that the cache usage and pre-loading should be enhanced and that the protocol should be changed, with report aggregation and compression using WBXML or gzip.

Promoting Customer Loyalty Through a Telecommunication Specific CRM Framework

The objective of this thesis is to find new methods to promote customer loyalty in the telecommunication industry. The study was made on the appointment of Tecnotree Corporation, which operates in a telecommunication software business. This thesis comprises of a theoretical section, which includes the presentation of relevant theoretical background and of an empirical section, which is mostly based on interviews and professional opinions. One of the most important aspects of customer relationship management (CRM) is building customer loyalty. Telecommunication markets are becoming saturated in an increasing number of markets. Customer acquisition is no longer a way to grow; companies need to hold on to their existing customers. Companies are always looking for new ways to avoid churning customers. These new methods are searched from recent studies and they are discussed with marketing professionals from operators. Current CRM lacks the ability to extract data efficiently from databases. Social media’s utilization is also in its early stages. It contains a lot of possibilities, but also things that can be risky for companies. In order to obtain full advantage of the new methods companies are required to invest considerable amounts of time and money to develop new systems, but not all companies are ready for the change.

Heat transfer inside internal combustion engine: modelling and comparison with experimental data

The paper is devoted to study specific aspects of heat transfer in the combustion chamber of compression ignited reciprocating internal combustion engines and possibility to directly measure the heat flux by means of Gradient Heat Flux Sensors (GHFS). A one – dimensional single zone model proposed by Kyung Tae Yun et al. and implemented with the aid of Matlab, was used to obtain approximate picture of heat flux behavior in the combustion chamber with relation to the crank angle. The model’s numerical output was compared to the experimental results. The experiment was accomplished by A. Mityakov at four stroke diesel engine Indenor XL4D. Local heat fluxes on the surface of cylinder head were measured with fast – response, high – sensitive GHFS. The comparison of numerical data with experimental results has revealed a small deviation in obtained heat flux values throughout the cycle and different behavior of heat flux curve after Top Dead Center.

Substitution-Newton-Raphson method applied to the modeling of a vapour compression refrigeration system using different representations of the thermodynamic properties of R-134a

This paper gives a detailed presentation of the Substitution-Newton-Raphson method, suitable for large sparse non-linear systems. It combines the Successive Substitution method and the Newton-Raphson method in such way as to take the best advantages of both, keeping the convergence features of the Newton-Raphson with the low requirements of memory and time of the Successive Substitution schemes. The large system is solved employing few effective variables, using the greatest possible part of the model equations in substitution fashion to fix the remaining variables, but maintaining the convergence characteristics of the Newton-Raphson. The methodology is exemplified through a simple algebraic system, and applied to a simple thermodynamic, mechanical and heat transfer modeling of a single-stage vapor compression refrigeration system. Three distinct approaches for reproducing the thermodynamic properties of the refrigerant R-134a are compared: the linear interpolation from tabulated data, the use of polynomial fitted curves and the use of functions derived from the Helmholtz free energy.

Assessment of the diagnostic value of diffusion tensor imaging in patients with spinal cord compression: a meta-analysis

We investigated the diagnostic value of the apparent diffusion coefficient (ADC) and fractional anisotropy (FA) of magnetic resonance diffusion tensor imaging (DTI) in patients with spinal cord compression (SCC) using a meta-analysis framework. Multiple scientific literature databases were exhaustively searched to identify articles relevant to this study. Mean values and standardized mean differences (SMDs) were calculated for the ADC and FA in normal and diseased tissues. The STATA version 12.0 software was used for statistical analysis. Of the 41 articles initially retrieved through database searches, 11 case-control studies were eligible for the meta-analysis and contained a combined total of 645 human subjects (394 patients with SCC and 251 healthy controls). All 11 studies reported data on FA, and 9 contained data related to the ADC. The combined SMDs of the ADC and FA showed that the ADC was significantly higher and the FA was lower in patients with SCC than in healthy controls. Subgroup analysis based on the b value showed higher ADCs in patients with SCC than in healthy controls at b values of both ≤500 and >500 s/mm2. In summary, the main findings of this meta-analysis revealed an increased ADC and decreased FA in patients with SCC, indicating that DTI is an important diagnostic imaging tool to assess patients suspected to have SCC.

Sentence Compression with Supervised Machine Learning

The subject of the thesis is automatic sentence compression with machine learning, so that the compressed sentences remain both grammatical and retain their essential meaning. There are multiple possible uses for the compression of natural language sentences. In this thesis the focus is generation of television program subtitles, which often are compressed version of the original script of the program. The main part of the thesis consists of machine learning experiments for automatic sentence compression using different approaches to the problem. The machine learning methods used for this work are linear-chain conditional random fields and support vector machines. Also we take a look which automatic text analysis methods provide useful features for the task. The data used for machine learning is supplied by Lingsoft Inc. and consists of subtitles in both compressed an uncompressed form. The models are compared to a baseline system and comparisons are made both automatically and also using human evaluation, because of the potentially subjective nature of the output. The best result is achieved using a CRF - sequence classification using a rich feature set. All text analysis methods help classification and most useful method is morphological analysis. Tutkielman aihe on suomenkielisten lauseiden automaattinen tiivistäminen koneellisesti, niin että lyhennetyt lauseet säilyttävät olennaisen informaationsa ja pysyvät kieliopillisina. Luonnollisen kielen lauseiden tiivistämiselle on monta käyttötarkoitusta, mutta tässä tutkielmassa aihetta lähestytään television ohjelmien tekstittämisen kautta, johon käytännössä kuuluu alkuperäisen tekstin lyhentäminen televisioruudulle paremmin sopivaksi. Tutkielmassa kokeillaan erilaisia koneoppimismenetelmiä tekstin automaatiseen lyhentämiseen ja tarkastellaan miten hyvin erilaiset luonnollisen kielen analyysimenetelmät tuottavat informaatiota, joka auttaa näitä menetelmiä lyhentämään lauseita. Lisäksi tarkastellaan minkälainen lähestymistapa tuottaa parhaan lopputuloksen. Käytetyt koneoppimismenetelmät ovat tukivektorikone ja lineaarisen sekvenssin mallinen CRF. Koneoppimisen tukena käytetään tekstityksiä niiden eri käsittelyvaiheissa, jotka on saatu Lingsoft OY:ltä. Luotuja malleja vertaillaan Lopulta mallien lopputuloksia evaluoidaan automaattisesti ja koska teksti lopputuksena on jossain määrin subjektiivinen myös ihmisarviointiin perustuen. Vertailukohtana toimii kirjallisuudesta poimittu menetelmä. Tutkielman tuloksena paras lopputulos saadaan aikaan käyttäen CRF sekvenssi-luokittelijaa laajalla piirrejoukolla. Kaikki kokeillut teksin analyysimenetelmät auttavat luokittelussa, joista tärkeimmän panoksen antaa morfologinen analyysi.