Die Entwicklung innovativer Therapiekonzepte für die Knochenregeneration erfordert validierte segmentale Knochendefekt-Tiermodelle. Dabei ist das Mausmodell für die präklinische Testung von zentraler Bedeutung, jedoch fehlen in der wissenschaftlichen Literatur bislang Angaben zu validierten, extern stabilisierenden kritischen segmentalen Knochendefektmodellen an der immundefizienten Maus. Das Ziel dieser Arbeit war daher die Entwicklung und in vivo Evaluierung eines zuverlässigen und einfach zu handhabenden Modells für extern stabilisierte kritische Knochendefekte an der immundefizienten Maus. Dreißig männliche nu/nu-Mäuse (40,7±2,8 g, 95±2,6 d) wurden mittels Isofluraninhalation narkotisiert und anschließend ein externer Fixateur (MouseExFix, RISystem, AO Research Institute Davos, Schweiz) am rechten Femur angebracht. Femorale Knochendefekte der Länge 1 mm (n=10), 2 mm (n=10) und 3 mm (n=10) wurden erzeugt. Der Wundverschluss erfolgte mit Einzelknopfnähten. Röntgenaufnahmen wurden unmittelbar postoperativ und im Folgenden alle zwei Wochen innerhalb des Beobachtungszeitraums von zwölf Wochen angefertigt und im Hinblick auf Knochenregeneration und –fusion ausgewertet. Weiterhin wurden histomorphologische, histomorphometrische, immunhistochemische und µCT-Analysen zur dreidimensionalen und zellulären Beurteilung der Knochenheilung angefertigt. Alle Tiere überlebten die Operation. Sechs Tiere starben innerhalb des Beobachtungszeitraums als Folge von starkem Blutverlust (n=1), Infektion (n=1), Pinlockerung, welche die Euthanasie erforderlich machte (n=2) und durch Komplikationen bei der Anästhesie (n=2). Die µCT-Analyse nach zwölf Wochen zeigte, dass 3/8 der 1 mm-Defekte, 5/8 der 2 mm-Defekte und 8/8 der 3 mm-Defekte eine Pseudarthrose aufwiesen. Das mittlere Defektvolumen stieg signifikant (p<0,001) mit der Größe des Defektes und betrug 0,36±0,42 mm³ (1 mm-Gruppe), 1,4±0,88 mm³ (2 mm-Gruppe), bzw. 2,88±0,28 mm³ (3 mm-Gruppe). Die mittlere Defektgröße verringerte sich entsprechend um 77,6% (1 mm-Gruppe), 56,8% (2 mm-Gruppe), bzw. 28,6% (3 mm-Gruppe). Die histomorphologischen, histomorphometrischen und immunhistochemischen Analysen zeigten keine statistisch signifikanten Unterschiede zwischen den drei experimentellen Gruppen. Das verwendete MouseExFix-System ist ein zuverlässiges und einfach zu handhabendes Verfahren zur Stabilisierung eines kritischen segmentalen Knochendefekts an der immundefizienten Maus, wenn ein 3 mm-Defekt erzeugt wird. Das im Rahmen der Studie entwickelte und validierte murine extern stabilisierte, segmentale kritische Knochendefektmodell ermöglicht die präklinische Evaluierung von Konzepten zur lokalen Knochenregeneration inklusive der Verwendung allo- und xenogener Zellen.:1 Einleitung 1 1.1 Hintergrund 1 1.2 Das Mausmodell 2 1.3 Übersicht Tierversuche mit Knochendefekten 5 1.4 Frakturheilung 6 1.4.1 Allgemeines 6 1.4.2 Räumliche Gliederung 7 1.4.3 Expression von Proteinen der extrazellulären Matrix 8 1.4.4 Das Vier-Phasen-Modell der Frakturheilung 9 1.4.5 Das anabolisch/ katabolische Modell der Frakturheilung 12 1.4.6 Beeinflussung der Frakturheilung 12 1.4.7 Das Diamantkonzept 14 1.5 Osteosynthesesysteme 14 1.6 Pseudarthrosen 15 1.6.1 Definition 15 1.6.2 Ätiologie 16 1.6.3 Klassifikation 16 1.6.4 Therapie 17 2 Material und Methoden 19 2.1 Versuchstiere 19 2.2 Operationsvorbereitung 19 2.3 Operationsablauf 20 2.4 Postoperatives Vorgehen 27 2.5 Verlaufskontrolle 28 2.6 Entnahme der Präparate 29 2.7 Anfertigung der µCT-Aufnahmen 30 2.8 Anfertigung der histologischen Schnitte 30 2.8.1 Bearbeitung der Femora 30 2.8.2 verwendete Färbungen 31 2.9 Beurteilung der Schnitte 32 2.9.1 Histologische Beurteilung 32 2.9.2 Histomorphometrische Beurteilung 33 2.10 Statistik 33 3 Ergebnisse 34 3.1 Überlebensraten und Gewichtsverlauf 34 3.2 Röntgenauswertung 35 3.3 CT-Auswertung 38 3.4 Histologische Auswertung 41 3.5 Histomorphometrische Auswertung 44 3.5.1 TRAP 44 3.5.2 Osteocalcin 46 3.5.3 Osteopontin 47 3.5.4 Osteonectin 48 4 Diskussion 50 4.1 Diskussion etablierter Modelle für kritische Knochendefekte 50 4.1.1 Diskussion der Konzeption der Modelle 50 4.1.2 Diskussion der durchgeführten Anästhesieverfahren 54 4.1.3 Diskussion der Ergebnisse 55 4.1.4 Diskussion der Defektlängen 56 4.1.5 Diskussion der verschiedenen Osteosynthesesysteme 57 4.1.6 Diskussion der Ausfaller 59 4.2 Anwendung und Nutzen immundefizienter Tiermodelle 60 4.3 Vergleich des tibialen und des femoralen murinen Frakturmodells 63 4.4 Diskussion der Beurteilung der Knochenheilung mittels bildgebender und histologischer Verfahren 63 4.5 Anwendungsmöglichkeiten 65 4.6 Schlussfolgerungen 66 5 Zusammenfassung 67 5.1 In deutscher Sprache 67 5.2 In englischer Sprache 68 6 Literaturverzeichnis 70 7 Anhang 92 7.1 Danksagung 92 7.2 Lebenslauf 93 7.3 Veröffentlichungen 95 7.4 Wertetabellen 96 7.5 Erklärungen zur Eröffnung des Promotionsverfahrens 109 / The development of innovative therapies for bone regeneration requires the use of advanced site-specific bone defect small animal models. In this context, murine models are of major importance as they allow for sufficient sample sizes prior to preclinical testing using larger animals. Owing to the small dimensions of the murine femur only a few custom fabricated fixation devices have been described in the literature so far. The aim of this investigation was to develop and validate a new, externally fixated critical size bone defect model for immunodeficient mice. Thirty male nu/nu mice (40.7 ± 2.8 g, 95 ± 2.6 days old) were anesthetized by isoflurane inhalation and an external fixation device (MouseExFix, RISystem, AO Research Institute Davos, Switzerland) was attached to the right femur. Femoral bone defects of 1 mm (n=10), 2 mm (n=10) and 3 mm (n=10) were created. Wounds were closed without any additional treatment. X-ray films obtained immediately after surgery and every 2 weeks postoperatively during the 12 week postoperative observation period were evaluated for bony regeneration and fusion. Furthermore, histomorphology, histomorphometry, immunohistochemistry and µCT analysis were performed. All of the animals survived the operation. Twenty four out of 30 animals reached the twelfth postoperative week. µCT analyses after twelve weeks showed that 3/8 of the 1 mm defects, 5/8 of the 2 mm defects and 8/8 of the 3 mm defects remained as nonunions. The defect volume was 0.36 ± 0.42 mm³ (1 mm group), 1.40 ± 0.88 mm³ (2 mm group), and 2.88 ± 0.28 mm³ (3 mm group) (p<0.001, between all groups). The defect size decreased by 77.6% (1-mm group), 56.8% (2-mm group) and 28.6% (3-mm group) (p=0.152, between all groups). Our method using the MouseExFix device has proven to be a reliable and easy-to-handle external fixation system for the stabilization of critical-size segmental bone defects in immundeficient mice when 3 mm defects are generated. This mouse model allows for high-throughput translational evaluation of concepts for site-specific bone regeneration including strategies using allogenic and xenogenic cell types.:1 Einleitung 1 1.1 Hintergrund 1 1.2 Das Mausmodell 2 1.3 Übersicht Tierversuche mit Knochendefekten 5 1.4 Frakturheilung 6 1.4.1 Allgemeines 6 1.4.2 Räumliche Gliederung 7 1.4.3 Expression von Proteinen der extrazellulären Matrix 8 1.4.4 Das Vier-Phasen-Modell der Frakturheilung 9 1.4.5 Das anabolisch/ katabolische Modell der Frakturheilung 12 1.4.6 Beeinflussung der Frakturheilung 12 1.4.7 Das Diamantkonzept 14 1.5 Osteosynthesesysteme 14 1.6 Pseudarthrosen 15 1.6.1 Definition 15 1.6.2 Ätiologie 16 1.6.3 Klassifikation 16 1.6.4 Therapie 17 2 Material und Methoden 19 2.1 Versuchstiere 19 2.2 Operationsvorbereitung 19 2.3 Operationsablauf 20 2.4 Postoperatives Vorgehen 27 2.5 Verlaufskontrolle 28 2.6 Entnahme der Präparate 29 2.7 Anfertigung der µCT-Aufnahmen 30 2.8 Anfertigung der histologischen Schnitte 30 2.8.1 Bearbeitung der Femora 30 2.8.2 verwendete Färbungen 31 2.9 Beurteilung der Schnitte 32 2.9.1 Histologische Beurteilung 32 2.9.2 Histomorphometrische Beurteilung 33 2.10 Statistik 33 3 Ergebnisse 34 3.1 Überlebensraten und Gewichtsverlauf 34 3.2 Röntgenauswertung 35 3.3 CT-Auswertung 38 3.4 Histologische Auswertung 41 3.5 Histomorphometrische Auswertung 44 3.5.1 TRAP 44 3.5.2 Osteocalcin 46 3.5.3 Osteopontin 47 3.5.4 Osteonectin 48 4 Diskussion 50 4.1 Diskussion etablierter Modelle für kritische Knochendefekte 50 4.1.1 Diskussion der Konzeption der Modelle 50 4.1.2 Diskussion der durchgeführten Anästhesieverfahren 54 4.1.3 Diskussion der Ergebnisse 55 4.1.4 Diskussion der Defektlängen 56 4.1.5 Diskussion der verschiedenen Osteosynthesesysteme 57 4.1.6 Diskussion der Ausfaller 59 4.2 Anwendung und Nutzen immundefizienter Tiermodelle 60 4.3 Vergleich des tibialen und des femoralen murinen Frakturmodells 63 4.4 Diskussion der Beurteilung der Knochenheilung mittels bildgebender und histologischer Verfahren 63 4.5 Anwendungsmöglichkeiten 65 4.6 Schlussfolgerungen 66 5 Zusammenfassung 67 5.1 In deutscher Sprache 67 5.2 In englischer Sprache 68 6 Literaturverzeichnis 70 7 Anhang 92 7.1 Danksagung 92 7.2 Lebenslauf 93 7.3 Veröffentlichungen 95 7.4 Wertetabellen 96 7.5 Erklärungen zur Eröffnung des Promotionsverfahrens 109

Just-In-Time Software Defect Prediction (JIT-DP) focuses on predicting errors in software at change-level with the objective of helping developers identify defects while the development process is still ongoing, and improving the quality of software applications. This work studies deep learning techniques by applying attention mechanisms that have been successful in, among others, Natural Language Processing (NLP) tasks. We introduce two networks named Convolutional Neural Network with Bidirectional Attention (BACNN) and Bidirectional Attention Code Network (BACoN) that employ a bi-directional attention mechanism between the code and message of a software change. Furthermore, we examine BERT [17] and RoBERTa [57] attention architectures for JIT-DP. More specifically, we study the effectiveness of the aforementioned attention-based models to predict defective commits compared to the current state of the art, DeepJIT [37] and TLEL [101]. Our experiments evaluate the models by using software changes from the OpenStack open source project. The results showed that attention-based networks outperformed the baseline models in terms of accuracy in the different evaluation settings. The attention-based models, particularly BERT and RoBERTa architectures, demonstrated promising results in identifying defective software changes and proved to be effective in predicting defects in changes of new software releases. / Just-In-Time Defect Prediction (JIT-DP) fokuserar på att förutspå fel i mjukvara vid ändringar i koden, med målet att hjälpa utvecklare att identifiera defekter medan utvecklingsprocessen fortfarande är pågående, och att förbättra kvaliteten hos applikationsprogramvara. Detta arbete studerar djupinlärningstekniker genom att tillämpa attentionmekanismer som har varit framgångsrika inom, bland annat, språkteknologi (NLP). Vi introducerar två nätverk vid namn Convolutional Neural Network with Bidirectional Attention (BACNN), och Bidirectional Attention Code Network (BACoN), som använder en tvåriktad attentionmekanism mellan koden och meddelandet om en mjukvaruändring. Dessutom undersöker vi BERT [17] och RoBERTa [57], attentionarkitekturer för JIT-DP. Mer specifikt studerar vi hur effektivt dessa attentionbaserade modeller kan förutspå defekta ändringar, och jämför dem med de bästa tillgängliga arkitekturerna DeePJIT [37] och TLEL [101]. Våra experiment utvärderar modellerna genom att använda mjukvaruändringar från det öppna källkodsprojektet OpenStack. Våra resultat visar att attentionbaserade nätverk överträffar referensmodellen sett till träffsäkerheten i de olika scenarierna. De attentionbaserade modellerna, framför allt BERT och RoBERTa, demonstrerade lovade resultat när det kommer till att identifiera defekta mjukvaruändringar och visade sig vara effektiva på att förutspå defekter i ändringar av nya mjukvaruversioner.

Thesis (PhD (Physics))--Stellenbosch University, 2008. / The advent of high intensity short pulse lasers has opened the door to investigating buried solid-solid interfaces through the technique of optical second harmonic generation (SHG). This has led to extensive study of technologically important systems such as the Si/SiO2 interface. In this study, SHG is employed to study the interface between highly boron doped p+-type Si and its native oxide layer (SiO2). Previous studies from this laboratory have extensively investigated the photo-induced charge transfer process across the Si/SiO2 interface in the case of undoped natively oxidized Si by means of SHG, with initial SHG measurements being performed on boron doped p+-type Si. The natively oxidized p+-type Si/SiO2 sample was placed on a computer controlled positioning system which allowed for translation of the sample and rotation around the azimuth. The laser system employed was characterized in terms of spectral composition, pulse duration, pulse repetition rate, spatial pro le and pulse energy in order to ensure quantitative measurements. The SHG signal generated from the sample interface was recorded in re ection. Under the applied irradiation conditions, defects are created at the interface by the near infra red (NIR) femtosecond radiation from the laser. These defects are then populated via multi-photon processes by electrons and to a lesser extent holes. The charge transfer across the interface induces an interfacial electric eld. This photo-induced electric eld is in addition to the built-in interfacial electric eld caused by positive ionization of naturally occurring interfacial defects due to the strong doping of the bulk Si. It is this interfacial electric eld, consisting of the built-in doping induced eld and the photo-induced electron and hole elds, that is probed by SHG. The SHG signal is strongly dependent on the magnitude of this interfacial electric eld as the electric eld induced second harmonic (EFISH) signal dominates all other contributions to the observed SHG signal in the case of the Si/SiO2 system. The temporal evolution of the SHG signal is recorded for di erent intensities from virgin as well as the pre-irradiated samples. This yields information about the time scales on which the charge separation occurs as well as the in- uence of existing photo-induced trap sites on the charge separation process, since the strength of the SHG signal is an indirect measure of the interfacial electric eld strength. The angular dependence of the SHG signal (SH rotational anisotropy measurements) for both the initial signal (when the doping induced electric eld dominates) and the saturated signal (when the electron induced electric eld dominates) is measured. Both these measurements show a four fold symmetry but with a relative 45 phase shift between them. This iii is taken as con rmation of the reversal of the interfacial electric eld direction. The initial SHG signal as a function of intensity is also recorded for di erent incident wavelengths. The variation in the non-quadratic dependence of the initial SHG signal on the incident intensity is attributed to a resonant enhancement of two-photon absorption and subsequent screening of the interfacial electric eld by charge carriers. The measurement performed and the results obtained contribute to the understanding of the photo-induced charge separation process across buried solid-solid interfaces, speci cally as it applies to the important Si/SiO2 interface.

The use of solid-state NMR and DFT calculations to study Y₂Sn[subscript(x)]Ti[subscript(2-x)]O₇, Y₂Sn[subscript(x)]Zr[subscript(2-x)]O₇ and Y₂Ti[subscript(x)]Zr[subscript(2-x)]O₇, materials with applications for the safe encapsulation of radioactive actinides is investigated. As a result of cation or anion disorder in these materials, NMR spectra are often complex and difficult to interpret. Therefore, an investigation using a range of NMR active nuclei and measurement of a variety of NMR parameters (isotropic chemical shift, δ[subscript(iso)]; span, Ω and quadrupolar coupling, C[subscript(Q)]) were used to provide a full and detailed picture of each material. The measurement of Ω in these disordered compounds with multiple resonances in the NMR spectra, required the use of 2D CSA-amplified PASS (CAPASS) experiments to enable the separation of each of the spinning sideband manifolds. An experimental assessment of the CAPASS experiment showed that although low ν₁/Ω[subscript(Hz)] ratios (as found in ⁸⁹Y NMR) resulted in distortions in the spectra obtained, a modified fitting procedure could be utilised to compensate for this fact, which allowed the accurate measurement of Ω. Despite the difficulties in acquiring the ⁸⁹Y NMR spectra, they were found to be the most informative of the NMR-active nuclei available. ¹¹⁹Sn NMR spectra, although much easier to acquire than ⁸⁹Y, were more complex and harder to analyse, owing to the overlapping resonances. Therefore, ¹¹⁹Sn NMR could only be used to confirm or support the results obtained using ⁸⁹Y NMR. Although ¹⁷O NMR was found to be useful, a full study could not be implemented due to the lack of ¹⁷O enriched samples; an area where future investigation may prove fruitful. Finally, [superscript(47/49)]Ti and ⁹¹Zr NMR spectra were found to be the most difficult to acquire due to their low receptivities and the quadrupolar broadened lineshapes, and as a result, little additional information was obtained. As a result of this analysis, for the Y₂Sn[subscript(x)]Ti[subscript(2-x)]O₇ pyrochlore solid solution, using primarily ⁸⁹Y δ[subscript(iso)] and Ω, and additionally confirmed with ¹¹⁹Sn δ[subscript(iso)], it was found that the Sn and Ti cations were randomly ordered throughout the B-sites. Additionally, ⁸⁹Y Ω could be used to obtain approximate Y-O[subscript(48f)] and Y-O[subscript(8b)] bond lengths for each type of Y environment. The study of Y₂Sn[subscript(x)]Zr[subscript(2-x)]O₇ using ⁸⁹Y NMR showed that although the end members were single phase, pyrochlore (Y₂Sn₂O₇) or defect fluorite (Y₂Zr₂O₇), the intermediate compositions were mostly two phase mixtures, consisting of an ordered pyrochlore (with an average formula of Y₂Sn₁.₈Zr₀.₂O₇) and a disordered phase, where the proportions of the pyrochlore and disordered phases decreased and increased, respectively, with the Zr content. Additionally, although the coordination states of the Y and Sn cations were easily determined using ⁸⁹Y and ¹¹⁹Sn NMR, respectively, the coordination states of the Zr cations could not be confirmed directly by ⁹¹Zr NMR. However, using indirect analysis from results obtained with ⁸⁹Y and ¹¹⁹Sn NMR, it was determined that 6 coordinate Zr was present in each composition, and it was always present in a greater proportion than 8 coordinate Zr. Finally, although ⁸⁹Y NMR spectra of Y₂Ti[subscript(x)]Zr[subscript(2-x)]O₇ were extremely difficult to analyse, it was tentatively proposed that they could be similar to Y₂Sn[subscript(x)]Zr[subscript(2-x)]O₇ due to some similarities observed between the spectra.

<p>Crystal structure, morphology and crystallization kinetics of melt-crystallized polypropylene and poly(propylene-stat-ethylene) fractions with 2.7 to 11.0 mol% of ethylene were studied by differential scanning calorimeter, wide- and small-angle X-ray scattering, polarized light microscopy, transmission electron microscopy and infrared spectroscopy. With increasing ethylene content the poly(propylene-stat-ethylene) fractions showed unchanged crystallinity, increased unit cell volume and constant crystal thickness in combination with a shortened helix length. This indicated that a fraction of ethylene defects were incorporated into the crystal structure. During the isothermal crystallization both α- and γ-crystals could be formed. The γ-crystal fraction increased with increasing ethylene content and increasing crystallization temperature. For samples with α- and γ-crystal contents, multimodal melting was observed and a noticeable γ- to α-crystal conversion was observed on slow heating. The spherulitic structure of the copolymers was coarser than that for the homopolymer.</p><p>The crystalline lamellae in copolymers exhibited profound curvature in contrast to the straighter cross-hatched α-crystals typical to the homopolymer. Area dependence of electrical breakdown strength was studied for thin polypropylene homopolymer films. The measurements were performed with an automatic measurement system equipped with a scanning electrode arm. Five different electrodes having areas between 0.045 cm2 and 9.3 cm2 were used and typically 40-80 breakdowns per sample and electrode area were collected. All measurements were performed on dry samples in air at room temperature. The data was analyzed statistically and the Weibull function parameters α and β, the first one related to 63% probability for the sample to break down and the second one to the width of the distribution were fitted to the obtained data. Different features concerning the measurement system and conditions, e.g. criteria for the automatic detection of the breakdowns, effect of electrode edge design, partial discharges, DC ramp speed and humidity were critically analyzed. It was concluded that the obtained α-parameter values were stable and repeatable over several years of time. The β-parameter values, however, varied ± 10-30%, more for the large than the small electrodes, and were also sensitive to the changes both in the sample itself and in the measurement conditions.</p><p>Breakdown strengths of over 50 capacitor grade polypropylene films were analyzed. The obtained α-parameter values were between 450 and 850 V/μm, depending on the film grade and electrode area. In addition to the high breakdown strengths, reflected by the obtained α-values, another, sparse distribution consisting of low breakdown strengths was revealed when the amount of measurement points was high enough. This means that more than one Weibull distribution could be needed to describe the breakdown strength behavior of a polypropylene film. Breakdown values showed decreasing area dependence with decreasing electrode area. Breakdown strengths for larger sample areas were predicted from the small area data by area- and Weibull extrapolation. The area extrapolation led to predicted α-values 50% higher than measured at 4 m² whereas the Weibull extrapolation showed an accuracy of ±15 % when predicted and measured values were compared.</p><p>Breakdown strengths were also extrapolated for film areas similar to those in impregnated power capacitors. It turned out that the power capacitors, tested at the factory, performed much better than predicted by the extrapolation. However, a few weak spots with very low breakdown values were also found. For the poly(ethyelene terephtalate) dielectric, which is not swelled by the impregnation liquid, the large area breakdown strength was predictable. This indicates that for polypropylene film processing and impregnation led, in addition to the improved large area breakdown performance, also to sparse weak spots with low breakdown probabilities. Different Weibull distributions were responsible for the breakdown strengths for the processed and impregnated polypropylene than for the dry film samples.</p>

Integrated circuit (IC) layer topology analysis methods are the main research topic of this doctoral thesis. Multiple methods are presented for IC layer feature analysis along with a software system where they are implemented and tested. Each of different IC layers has distinct features therefore it is very difficult to use universal algorithm for their analysis. A specialized software system was developed to test various analysis algorithms. The system and its architecture is a part of this thesis. Main tasks solved during research of this these were: finding or developing of optimal methods suitable for IC layer structure recognition, software system design and implementation, experimental testing of implemented methods accuracy and efficiency. Thesis consists of introduction, four chapters and the chapter of conclusions. In the introduction scientific novelty of the work is described as well as the aims and tasks of the work are formulated and the author’s publications and structure of the thesis are presented. The first chapter is dedicated to literature review. It covers existing IC layer structure analysis systems and algorithms which are used for this task. Generic image processing and analysis algorithms and methods are also covered as they are used as part of methodology developed in this thesis. The second chapter details different types of IC layers and their properties. Image processing and analysis methods suitable for each of these layer types are... [to full text] / Disertacijoje nagrinėjama integrinių grandynų (IG) topologijos elementų atpažinimo sistemos metodai ir algoritmai, jų taikymas bei pačios sistemos architektūra. Integrinių schemų projektavimo ir gamybos pramonėje problema yra automatinis kiekvieno technologinio lusto sluoksnio vaizdinės informacijos apdorojimas ir analizė, tiksliai išskiriant gamybos proceso metu suformuotas struktūras, tam kad šių duomenų pagalba galima būtų atlikti gamybos proceso tikslumo patikrinimą. Šio disertacijos darbo tyrimų objektas yra puslaidininkinių integrinių schemų sluoksniuose suformuoti elementai. Kiekvieno iš skirtingų sluoksnių struktūros pasižymi skirtingomis savybėmis, todėl labai sunku sukurti universalius analizės metodus. Dėl šios priežasties buvo sukurta speciali programinės įrangos (PĮ) sistema. PĮ architektūra yra disertacijos tyrimų objektas. Pagrindiniai disertacijoje sprendžiami uždaviniai: IG elementų struktūrų atpažinimo metodų pritaikymas ir kūrimas, PĮ sistemos projektavimas ir įgyvendinimas, eksperimentinis įdiegtų metodų efektyvumo ir tikslumo tyrimas. Disertaciją sudaro įvadas, keturi skyriai ir rezultatų apibendrinimas. Įvade nagrinėjamas problemos aktualumas, formuluojamas darbo tikslas bei uždaviniai, aprašomas mokslinis darbo naujumas, pristatomi autoriaus pranešimai ir publikacijos, disertacijos struktūra. Pirmasis skyrius skirtas analitinei literatūros apžvalgai. Jame nagrinėjamos žinomos IG topologijos elementų atpažinimo sistemos ir jose naudojami metodai... [toliau žr. visą tekstą]

Context. Reports suggest that defects in code cost the US in excess of $50billion per year to put right. Defect Prediction is an important part of Software Engineering. It allows developers to prioritise the code that needs to be inspected when trying to reduce the number of defects in code. A small change in the number of defects found will have a significant impact on the cost of producing software. Aims. The aim of this dissertation is to investigate the factors which a ect the performance of defect prediction models. Identifying the causes of variation in the way that variables are computed should help to improve the precision of defect prediction models and hence improve the cost e ectiveness of defect prediction. Methods. This dissertation is by published work. The first three papers examine variation in the independent variables (code metrics) and the dependent variable (number/location of defects). The fourth and fifth papers investigate the e ect that di erent learners and datasets have on the predictive performance of defect prediction models. The final paper investigates the reported use of di erent machine learning approaches in studies published between 2000 and 2010. Results. The first and second papers show that independent variables are sensitive to the measurement protocol used, this suggests that the way data is collected a ects the performance of defect prediction. The third paper shows that dependent variable data may be untrustworthy as there is no reliable method for labelling a unit of code as defective or not. The fourth and fifth papers show that the dataset and learner used when producing defect prediction models have an e ect on the performance of the models. The final paper shows that the approaches used by researchers to build defect prediction models is variable, with good practices being ignored in many papers. Conclusions. The measurement protocols for independent and dependent variables used for defect prediction need to be clearly described so that results can be compared like with like. It is possible that the predictive results of one research group have a higher performance value than another research group because of the way that they calculated the metrics rather than the method of building the model used to predict the defect prone modules. The machine learning approaches used by researchers need to be clearly reported in order to be able to improve the quality of defect prediction studies and allow a larger corpus of reliable results to be gathered.

Designing materials for fusion applications is a very challenging problem, requiring detailed understanding of the behaviour of materials under the kinds of extreme conditions expected in a fusion environment. During the lifetime of fusion-reactor components, materials will be subjected to high levels of neutron irradiation, but must still perform effectively at high operating temperatures and under significant loading conditions. Body-centred cubic (bcc) transition metals are some of the most promising candidates for structural materials in fusion because of their relatively high density, which allows for effective neutron-shielding with the minimum volume and mass of material. In this work we perform atomistic simulations on two of the most important of these, Fe and W. In this thesis we describe atomic-scale simulations of defects found in bcc systems. In part I we consider the vacancy and interstitial loop defects that are produced and accumulated as a result of irradiation-induced displacement cascades. We show that vacancy dislocation loops have a critical size below which they are highly unstable relative to planar void defects, and thus offer an explanation as to why they are so rarely seen in TEM observations of irradiated bcc metals. Additionally, we compare the diffusion rates of these vacancy loops to their interstitial counterparts and find that, while interstitial loops are more mobile, the difference in mobility is not as significant as might have been expected. In part II we study screw dislocations, which, as the rate limiting carriers of plastic deformation, are significantly responsible for the strength of materials. We present results from large-scale finite temperature molecular dynamics simulations of screw dislocations under stress and observe the thermally-activated kink-pair formation regime at low stress, which appears to be superseded by a frictional regime at higher stresses. The mobility functions fitted to the results are vital components in simulations of dislocation networks and other large-scale phenomena. Lastly, we develop a multi-string Frenkel-Kontorova model that allows us to study the core structure of screw dislocations. Subtle changes in the form of the interaction laws used in this model demonstrate the difference between the non-degenerate and degenerate core structures. We provide simple criteria to guarantee the correct structure when developing interatomic potentials for bcc metals.

Implantation requires complex molecular and cellular events involving coagulation, angiogenesis and immunological processes that need to be well regulated for a pregnancy to establish and progress normally.  The overall aim of this thesis was to study different models associated with atypical angiogenesis, impaired implantation and/or placentation, such as recurrent miscarriage (RM), oocyte donation (OD) and preeclampsia. Histidine-rich glycoprotein (HRG), a serum protein with angiogenic potential has been previously shown to have an impact on implantation and fertility.  In two retrospective case-control studies, women suffering from RM (Study I) and gestational hypertensive disorders (GHD) (Study IV) have been compared to healthy control women, regarding carriership of HRG genotypes (HRG A1042G and C633T SNP, respectively).  According to the findings of this thesis, heterozygous carriers of the HRG A1042G SNP suffer from RM more seldom than homozygous carriers (Study I).  Additionally, the presence of the HRG 633T allele was associated with increased odds of GHD (GHD IV).  Studies II and III comprised a national cohort of relatively young women with optimal health status conceiving singletons with donated oocytes versus autologous oocytes (spontaneously or via IVF).  We explored differences in various obstetric (Study II) and neonatal (Study III) outcomes from the Swedish Medical Birth Register.  Women conceiving with donated oocytes had a higher risk of GHD, induction of labor and cesarean section, as well as postpartum hemorrhage and retained placenta, when compared to autologously conceiving women.  OD infants had higher odds of prematurity and lower birthweight and length when born preterm, compared to neonates from autologous oocytes.  With regard to the indication of OD treatment, higher intervention but neverthelss favourable neonatal outcomes were observed in women with diminished ovarian reserve; the risk of GHD did not differ among OD recipients after adjustment. In conclusion, HRG genetic variation appears to contribute to placental dysfunction disorders.  HRG is potential biomarker that may contribute in the prediction of the individual susceptibility for RM and GHD.  Regarding OD in Sweden, the recipients-despite being of optimal age and health status- need careful preconceptional counselling and closer prenatal monitoring, mainly due to increased prevalence of hypertensive disorders and prematurity.

When processing materials, there are a number of factors which influence the final quality of products. The possibility of defects occurrence, both sub-surface and surface ones, is one of the factors. In the precision engineering production, the automotive, bearing or aerospace industry in particular, no level of defects is tolerable. The thesis deals with the causes and consequences of the occurrence of such defects, and the methods for their identification. For this purpose, a new contactless optical method for the fully automatic inspection of the surface defects has been developed by Mesing, spol. s r.o. In order to verify the performance of the aforementioned method, four specific products have been tested by the author using the method. Finally, there are examples of the successful implementation of the automatic machines, for a design of which the results of the model testing were utilized.