One of every eight women will in their lifetime suffer from breast cancer, making it the most common type of cancer for women. A successful treatment is very much dependent on identifying metastatic tissue which is cancer found beyond the initial tumour. Using deep learning within biomedical analysis has become an effective approach. However, its success is very dependent on large datasets. Data augmentation is a way to enhance datasets without requiring more annotated data. One way of doing this is using the cutout method which masks parts of an input image. Our research focused on investigating how the cutout method could improve the performance of Convolutional Neural Networks for classifying metastatic tissue on the Patch Camelyon dataset. Our research showed that improvements in performance can be achieved by using the cutout method. Further, our research suggests that using a non label- preserving version of cutout is better than a label- preserving version. The most improvement in accuracy was seen when we used a randomly sized cutout mask. The experiment resulted in an increase in accuracy by 3.6%, from the baseline of 82,3% to 85.9%. The cutout method was also compared- and used in conjunction with other well- established data augmentation techniques. Our conclusion is that cutout can be a competitive form of data augmentation that can be used both with and without other data augmentation techniques. / Var åttonde kvinna drabbas under sin livstid av bröstcancer. Detta gör det till den vanligaste formen av cancer för kvinnor. En framgångsrik behandling är beroende av att kunna identifiera metastatisk vävnad, vilket är cancer som spridit sig bortom den ursprungliga tumören. Att använda djupinlärning inom biomedicinsk analys har blivit en effektiv metod. Dock är dess framgång väldigt beroende av stora datamängder. Dataförstärkning är olika sätt att förbättra en mängd data som inte innebär att addera ytterligare annoterad data. Ett sätt att göra detta är genom den en metod som kallas Cutout som maskar en del av en bild. Vår studie undersöker hur Cutout påverkar resultatet när Convolutional Neural Networks klassificerar huruvida bilder från datasetet Patch Camelyon innehåler metastaser eller inte. Vår studie visar att användandet av Cutout kan innebära förbättringar i resultatet. Dessutom tyder vår studie på att resultatet förbättras än mer om även delen av bilden som kan innehålla metastaser kan maskas ut. Den största förbättringen i resultatet var när maskningen var av varierande storlek från bild till bild. Resultatet förbättrades från 82.3% korrekta klassifikationer utan någon dataförstärkning till 85.9% med den bästa versionen av Cutout. Cutout jämfördes också, och användas tillsammans med, andra väletablerade dataförstärkningsmetoder. Vår slutsats är att Cutout är en dataförstärkningsmetod med potentital att vara användbar såväl med som utan andra dataförstärkningsmetoder.

No description available.

Nowadays inspections of civil engineering structures are performed manually at close range to be able to assess damages. This requires specialized equipment that tends to be expensive and to produce closure of the bridge. Furthermore, manual inspections are time-consuming and can often be a source or risk for the inspectors. Moreover, manual inspections are subjective and highly dependent on the state of mind of the inspector which reduces the accuracy of this kind of inspections. Image-based inspections using cameras or unmanned aerial vehicles (UAV) combined with image processing have been used to overcome the challenges of traditional manual inspections. This type of inspection has also been studied with the use of machine learning algorithms to improve the detection of damages, in particular cracks. This master’s thesis presents an approach that combines different aspects of the inspection, from the data acquisition, through the crack detection to the quantification of essential parameters. To do this, both digital cameras and a UAV have been used for data acquisition. A convolutional neural network (CNN) for the identification of cracks is used and subsequently, different quantification methods are explored to determine the width and length of the cracks. The results are compared with control measures to determine the accuracy of the method. The results present low to no false negatives when using the CNN to identify cracks. The quantification of the identified cracks is performed obtaining the highest accuracy estimation for 0.2mm cracks.

Bladder cancer (BCa) is the fourth most commonly diagnosed cancers in men and the eighth most common in women. It is an abnormal growth of tissues which develops in the bladder lining. Histological analysis of bladder tissue facilities diagnosis as well as it serves as an important tool for research. To bet- ter understand the molecular profile of bladder cancer and to detect predictive and prognostic features, microscopy methods, such as immunofluorescence (IF), are used to investigate the characteristics of bladder cancer tissue. For this project, a new method is proposed to segment cancer epithelial us- ing nuclei morphology captured with IF staining. The method is implemented using deep learning algorithms and performance achieved is compared with the literature. The dataset is stained for nuclei (DAPI) and a marker for cancer epithelial (panEPI) which was used to create the ground truth. Three popu- lar Convolutional Neural Network (CNN) namely U-Net, Residual U-Net and VGG16 were implemented to perform the segmentation task on the tissue mi- croarray dataset. In addition, a transfer learning approach was tested with the VGG16 network that was pre-trained with ImageNet dataset. Further, the performance from the three networks were compared using 3fold cross-validation. The dice accuracies achieved were 83.32% for U-Net, 88.05% for Residual U-Net and 82.73% for VGG16. These findings suggest that segmentation of cancerous tissue regions, using only the nuclear morphol- ogy, is feasible with high accuracy. Computer vision methods better utilizing nuclear morphology captured by the nuclear stain, are promising approaches to digitally augment the conventional IF marker panels, and therefore offer im- proved resolution of the molecular characteristics for research settings.

Tsunami and storm surge are two of the main destructive and costly natural hazards faced by coastal communities around the world. To enhance coastal resilience and to develop effective risk management strategies, accurate and efficient tsunami and storm surge prediction models are needed. However, existing physics-based numerical models have the disadvantage of being difficult to satisfy both accuracy and efficiency at the same time. In this dissertation, several surrogate models are developed using statistical and machine learning techniques that can rapidly predict a tsunami and storm surge without substantial loss of accuracy, with respect to high-fidelity physics-based models. First, a tsunami run-up response function (TRRF) model is developed that can rapidly predict a tsunami run-up distribution from earthquake fault parameters. This new surrogate modeling approach reduces the number of simulations required to build a surrogate model by separately modeling the leading order contribution and the residual part of the tsunami run-up distribution. Secondly, a TRRF-based inversion (TRRF-INV) model is developed that can infer a tsunami source and its impact from tsunami run-up records. Since this new tsunami inversion model is based on the TRRF model, it can perform a large number of tsunami forward simulations in tsunami inversion modeling, which is impossible with physics-based models. And lastly, a one-dimensional convolutional neural network combined with principal component analysis and k-means clustering (C1PKNet) model is developed that can rapidly predict the peak storm surge from tropical cyclone track time series. Because the C1PKNet model uses the tropical cyclone track time series, it has the advantage of being able to predict more diverse tropical cyclone scenarios than the existing surrogate models that rely on a tropical cyclone condition at one moment (usually at or near landfall). The surrogate models developed in this dissertation have the potential to save lives, mitigate coastal hazard damage, and promote resilient coastal communities. / Doctor of Philosophy / Tsunami and storm surge can cause extensive damage to coastal communities; to reduce this damage, accurate and fast computer models are needed that can predict the water level change caused by these coastal hazards. The problem is that existing physics-based computer models are either accurate but slow or less accurate but fast. In this dissertation, three new computer models are developed using statistical and machine learning techniques that can rapidly predict a tsunami and storm surge without substantial loss of accuracy compared to the accurate physics-based computer models. Three computer models are as follows: (1) A computer model that can rapidly predict the maximum ground elevation wetted by the tsunami along the coastline from earthquake information, (2) A computer model that can reversely predict a tsunami source and its impact from the observations of the maximum ground elevation wetted by the tsunami, (3) A computer model that can rapidly predict peak storm surges across a wide range of coastal areas from the tropical cyclone's track position over time. These new computer models have the potential to improve forecasting capabilities, advance understanding of historical tsunami and storm surge events, and lead to better preparedness plans for possible future tsunamis and storm surges.

Dynamical properties of non-equilibrium systems, similar to equilibrium ones, have been shown to obey robust time scaling laws which have enriched the concept of physical universality classes. In the first part of this Dissertation, we present the results of our investigations of some of the critical dynamical properties of systems belonging to the Voter or the Directed Percolation (DP) universality class. To be more precise, we focus on the aging properties of two-state and three-state Potts models with absorbing states and we determine temporal scaling of autocorrelation and autoresponse functions. We propose a novel microscopic model which exhibits non-equilibrium critical points belonging to the Voter, DP and Ising Universality classes. We argue that our model has properties similar to the Generalized Voter Model (GVM) in its Langevin description. Finally, we study the time evolution of the width of interfaces separating different absorbing states. The second part of this Dissertation is devoted to the applications of Machine Learning models in physical systems. First, we show that a trained Convolutional Neural Network (CNN) using configurations from the Ising model with conserved magnetization is able to find the location of the critical point. Second, using as our training dataset configurations of Ising models with conserved or non-conserved magnetization obtained in importance sampling Monte Carlo simulations, we investigate the physical properties of configurations generated by the Restricted Boltzmann Machine (RBM) model. The first part of this research was sponsored by the US Army Research Office and was accomplished under Grant Number W911NF-17-1-0156. The second part of this work was supported by the United States National Science Foundation through grant DMR-1606814. / Doctor of Philosophy / Physical systems with equilibrium states contain common properties with which they are categorized in different universality classes. Similar to these equilibrium systems, non-equilibrium systems may obey robust scaling laws and lie in different dynamic universality classes. In the first part of this Dissertation, we investigate the dynamical properties of two important dynamic universality classes, the Directed Percolation universality class and the Generalized Voter universality class. These two universality classes include models with absorbing states. A good example of an absorbing state is found in the contact process for epidemic spreading when all individuals are infected. We also propose a microscopic model with tunable parameters which exhibits phase transitions belonging to the Voter, Directed Percolation and Ising universality classes. To identify these universality classes, we measure specific dynamic and static quantities, such as interface density at different values of the tunable parameters and show that the physical properties of these quantities are identical to what is expected for the different universal classes. The second part of this Dissertation is devoted to the application of Machine Learning models in physical systems. Considering physical system configurations as input dataset for our machine learning pipeline, we extract properties of the input data through our machine learning models. As a supervised learning model, we use a deep neural network model and train it using configurations from the Ising model with conserved dynamics. Finally, we address the question whether generative models in machine learning (models that output objects that are similar to inputs) are able to produce new configurations with properties similar to those obtained from given physical models. To this end we train a well known generative model, the Restricted Boltzmann Machine (RBM), on Ising configurations with either conserved or non-conserved magnetization at different temperatures and study the properties of configurations generated by RBM. The first part of this research was sponsored by the US Army Research Office and was accomplished under Grant Number W911NF-17-1-0156. The second part of this work was supported by the United States National Science Foundation through grant DMR-1606814.

Yes / Aggressive posts containing symbolic and offensive images, inappropriate gestures along with provocative textual comments are growing exponentially in social media with the availability of inexpensive data services. These posts have numerous negative impacts on the reader and need an immediate technical solution to filter out aggressive comments. This paper presents a model based on a Convolutional Neural Network (CNN) and Binary Particle Swarm Optimization (BPSO) to classify the social media posts containing images with associated textual comments into non-aggressive, medium-aggressive and high-aggressive classes. A dataset containing symbolic images and the corresponding textual comments was created to validate the proposed model. The framework employs a pre-trained VGG-16 to extract the image features and a three-layered CNN to extract the textual features in parallel. The hybrid feature set obtained by concatenating the image and the text features were optimized using the BPSO algorithm to extract the more relevant features. The proposed model with optimized features and Random Forest classifier achieves a weighted F1-Score of 0.74, an improvement of around 3% over unoptimized features.

Companies in the transporting sector are undergoing an important transformation of electrifyingtheir fleets to meet the industry’s climate targets. To meet customer’s requests, keep its marketposition, and to contribute to a sustainable transporting industry, Scania needs to be in frontof the evolution. One aspect of this is to attract customers by providing accurate information anddetecting customer’s opportunities for electrification. Understanding the natural behavior of weatherparameters and their impact on energy consumption is crucial for providing accurate simulations ofhow daily operations would appear with an electric truck. The aim of this thesis is to map weatherparameters impact on energy consumption and to get an understanding of the correlations betweenenergy consumption and dynamic weather data. ML and deep learning models have undergone training using historical data from operations per-formed by Scania’s Battery Electric Vehicles(BEV). These models have been assessed against eachother to ensure that they are robust and accurate. Utilizing the trained models ability to providereliable consumption predictions based on weather, we can extract information and patterns aboutconsumption derived from customised weather parameters. The results show several interesting correlations and can quantify the impact of weather parametersunder certain conditions. Temperature is a significant factor that has a negative correlation withenergy consumption while other factors like precipitation and humidity prove less clear results. Byinteracting parameters with each other, some new results were found. For instance, the effect ofhumidity is clarified under certain temperatures. Wind speed also turns out to be an importantfactor with a positive correlation to energy consumption.

Yes / Multimodal biometric systems have been widely applied in many real-world applications due to its ability to deal with a number of significant limitations of unimodal biometric systems, including sensitivity to noise, population coverage, intra-class variability, non-universality, and vulnerability to spoofing. In this paper, an efficient and real-time multimodal biometric system is proposed based on building deep learning representations for images of both the right and left irises of a person, and fusing the results obtained using a ranking-level fusion method. The trained deep learning system proposed is called IrisConvNet whose architecture is based on a combination of Convolutional Neural Network (CNN) and Softmax classifier to extract discriminative features from the input image without any domain knowledge where the input image represents the localized iris region and then classify it into one of N classes. In this work, a discriminative CNN training scheme based on a combination of back-propagation algorithm and mini-batch AdaGrad optimization method is proposed for weights updating and learning rate adaptation, respectively. In addition, other training strategies (e.g., dropout method, data augmentation) are also proposed in order to evaluate different CNN architectures. The performance of the proposed system is tested on three public datasets collected under different conditions: SDUMLA-HMT, CASIA-Iris- V3 Interval and IITD iris databases. The results obtained from the proposed system outperform other state-of-the-art of approaches (e.g., Wavelet transform, Scattering transform, Local Binary Pattern and PCA) by achieving a Rank-1 identification rate of 100% on all the employed databases and a recognition time less than one second per person.

Twitter är en av de populäraste mikrobloggarna, som används för att uttryckatankar och åsikter om olika ämnen. Ett område som har dragit till sig mycketintresse under de senaste åren är twittersentimentanalys. Twittersentimentanalyshandlar om att bedöma vad för sentiment ett inlägg på Twitter uttrycker, om detuttrycker någonting positivt eller negativt. Olika metoder kan användas för attutföra twittersentimentanalys, där vissa lämpar sig bättre än andra. De vanligastemetoderna för twittersentimentanalys använder maskininlärning.Syftet med denna studie är att utvärdera tre stycken klassificeringsalgoritmerinom maskininlärning och hur märkningen av en datamängd påverkar en klassifi-ceringsmodells förmåga att märka ett twitterinlägg korrekt för twittersentimenta-nalys. Naive Bayes, Support Vector Machine och Convolutional Neural Network ärklassificeringsalgoritmerna som har utvärderats. För varje klassificeringsalgoritmhar två klassificeringsmodeller tagits fram, som har tränats och testats på två se-parata datamängder: Stanford Twitter Sentiment och SemEval. Det som skiljer detvå datamängderna åt, utöver innehållet i twitterinläggen, är märkningsmetodenoch mängden twitterinlägg. Utvärderingen har gjorts utefter vilken prestanda deframtagna klassificeringmodellerna uppnår på respektive datamängd, hur lång tidde tar att träna och hur invecklade de var att implementera.Resultaten av studien visar att samtliga modeller som tränades och testades påSemEval uppnådde en högre prestanda än de som tränades och testades på Stan-ford Twitter Sentiment. Klassificeringsmodellerna som var framtagna med Convo-lutional Neural Network uppnådde bäst resultat över båda datamängderna. Dockär ett Convolutional Neural Network mer invecklad att implementera och tränings-tiden är betydligt längre än Naive Bayes och Support Vector Machine. / Twitter is one of the most popular microblogs, which is used to express thoughtsand opinions on different topics. An area that has attracted much interest in recentyears is Twitter sentiment analysis. Twitter sentiment analysis is about assessingwhat sentiment a Twitter post expresses, whether it expresses something positiveor negative. Different methods can be used to perform Twitter sentiment analysis.The most common methods of Twitter sentiment analysis use machine learning.The purpose of this study is to evaluate three classification algorithms in ma-chine learning and how the labeling of a data set affects classification models abilityto classify a Twitter post correctly for Twitter sentiment analysis. Naive Bayes,Support Vector Machine and Convolutional Neural Network are the classificationalgorithms that have been evaluated. For each classification algorithm, two classi-fication models have been trained and tested on two separate data sets: StanfordTwitter Sentiment and SemEval. What separates the two data sets, in addition tothe content of the twitter posts, is the labeling method and the amount of twitterposts. The evaluation has been done according to the performance of the classifi-cation models on the respective data sets, training time and how complicated theywere to implement.The results show that all models trained and tested on SemEval achieved ahigher performance than those trained and tested on Stanford Twitter Sentiment.The Convolutional Neural Network models achieved the best results over both datasets. However, a Convolutional Neural Network is more complicated to implementand the training time is significantly longer than Naive Bayes and Support VectorMachine.