Browsing by Author "Rousu, Juho"
Now showing 1 - 20 of 296
Results Per Page
Sort Options
Item 3D-tulostuksen työkaluketjun analysointi kotikäytön näkökulmasta(2015-04-19) Hirvola, Ossi; Nurminen, Jukka; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem 5G - Viidennen sukupolven matkapuhelinverkko(2015-04-19) Patoila, Patrick; Ylä-Jääski, Antti; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem 5G ja käytettävyys(2016-01-28) Lehto, Olli; Ylä-Jääski, Antti; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Ad-hoc social interaction for sports(2016-04-17) Tainio, Ville; McGookin, David; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Adaptiivinen reaaliaikaisen videon suoratoisto(2016-04-17) Kurkinen, Antti; Siekkinen, Matti; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Agenttien ja web-tekniikoiden yhdistämisestä(2013-12-05) Burman, Michael; Järvinen, Hannu; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Ajoneuvojen reititys teiden aurausteht aviss a(2014-04-11) Kari, Miko; Rintanen, Jussi; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Algoritmvisualisering(2014-04-25) Hellström, Kasper; Malmi, Lauri; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Alternatives to Password-based Authentication in Web Services(2016-04-17) Huttunen, Max; Suoranta, Sanna; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Älykkään sähköverkon toiminta kotitalouksissa(2013-12-05) Saarela, Juho; Nurminen, Jukka; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Älykkäät agentit rajatussa ympäristössä(2014-12-07) Päivinen, Oskari; Mäkelä, Eetu; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Analys av raptexter med hjälp av ämnesmodellering(2016-04-18) Kreutzman, Emil; Malmi, Eric; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem En analys av trådlös betalning med hjälp av närfältskommunikation(2013-12-04) Lindroos, Niklas; Luukkainen, Sakari; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Analyzing Performance of Latent Tensor Methods on Large High-Rank Data Sets(2020-08-17) Mentu, Santeri; Szedmak, Sandor; Perustieteiden korkeakoulu; Rousu, JuhoLinear regression techniques are very popular in the natural sciences thanks to their scalability, well understood mathematical properties, and interpretable parameters. However including multivariate interactions in the form of feature transformation leads to an explosion in the number of model parameters. In general the problem of nonlinear regression with higher order interactions on high dimensional complex data is quite challenging with few good general purpose solutions. Impressive results have been achieved using deep neural networks and kernel methods, but both have significant shortcomings and unsolved issues in certain scenarios. In this thesis I analyze and compare the efficacy of different latent tensor models in this task. I focus on the recently published method of latent tensor reconstruction which has not yet been extensively studied on large ($10^7$ examples) real world data sets. I present a new software implementation of this model and experiment with different training configurations to achieve optimal performance. I used both synthetic and real world data to study the characteristics of the model and the limitations of the training procedure. I used the publicly available NCI-ALMANAC dataset to compare the performance of latent tensor reconstruction to higher order factorization machines, which has previously been used to achieve state of the art performance in this task, as well as deep neural networks. Latent tensor reconstruction was able to exceed the predictive accuracy of higher order factorization machines, but deep neural networks achieved the highest performance overall. The results are promising in terms of latent tensor reconstruction achieving accuracy on par with deep neural networks with further development.Item Android-mobiilipalveluissa sovellettavat k äytett ävyysohjeistot(2013-12-04) Haikala, Niko; Riihiaho, Sirpa; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Angler-hyökkäyspakin salatun verkkoliikenteen purkaminen(2016-04-17) Nurmi, Henri; Nyberg, Kaisa; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem AngularJS- ja Backbone.js-JavaScript-ohjelmistokehysten sopivuus web-sovelluksiin(2014-04-13) Stenroth, Johan; Malmi, Lauri; Tietotekniikan laitos; Perustieteiden korkeakoulu; Rousu, JuhoItem Anomaly Detection from Patient Visit Data(2016-10-27) Zhao, Yang; Hollmén, Jaakko; Perustieteiden korkeakoulu; Rousu, JuhoHospital operation cost rises due to the growing demand for outpatient services by increasing elderly population. To reduce the operation cost and serve the patients better, improvements on the efficiency in healthcare service institutes are required. Among several potential aspects of efficiency improvements, smoother patient visits are highly desired. Thanks to the digital era, patient visits to the hospital can be recorded with all details. The Oulu Hospital in Finland starts to gather patient visits data since 2011, using queue system provided by X-Akseli company. Utilizing these collected data, this thesis aims at designing a practical way of detecting anomalies from patient visits. With the help from this system, the hospital administrative staff could analyze the performance of the queue procedure in the hospital and optimize the procedure. Even better, the system can identify anomalies in real-time so that the patient can get immediate help when it is needed. The thesis explored two categories of methods: clustering methods and generative methods. Four candidate algorithms, K-Means, DBSCAN, Markov Chain, and Hidden Markov Model, are discussed. The discussion suggests that DBSCAN and Hidden Markov Model are more practical. Then we proposed a new data representation and used negative binomial distribution in Hidden Markov Model to model patient states durations. The experiment result was visualized using t- SNE and evaluated by user interpretation. The analyses show that both DBSCAN and Hidden Markov Model can effectively detect anomalies from patient visits data. But in terms of time and space complexity, and real-time detection, Hidden Markov Model is a better choice.Item Application of quantitative proteomic approaches for optimization of the ex-vivo expansion program of human mesenchymal stem cells(2016-02-04) Garcia Gonzalez, Ana; dos Santos, Sandra; Perustieteiden korkeakoulu; Rousu, JuhoHuman mesenchymal stem cells (MSCs) possess a multilineage differentiation capacity which makes them potentially interesting for therapeutic applications. Expansion protocols for obtaining clinically meaningful cell numbers usually consist of growing MSCs under ambient oxygen conditions. However, hypoxic ex-vivo cultivation has been proven to enhance MSCs proliferation, self-renewal and long-term viability, probably by mimicking the physiological conditions of the low oxygen in-vivo "stem cell niche". To build a better understanding of the molecular mechanisms underlying MSCs behavior under these conditions, a quantitative proteomics approach using two-dimensional gel electrophoresis coupled with mass spectrometry was used to analyse the expression profiles of ex-vivo expanded cultures of bone marrow (BM) MSCs and adipose-derived stromal cells (ASCs) under hypoxia (2% O2) and normoxia (21% O2). Proteins belonging to the functional categories "Structural components and cellular cytoskeleton", "Glycolysis", and "Folding and stress response proteins" are more abundant in hypoxia compared to normoxia in both cell sources, however, there were differences between the BM and adipose tissue (AT) cell proteomes. Proteins in "Cell cycle and regulation" and "Apoptosis" are slightly less abundant in hypoxia compared to normoxia. The high number of multiple size and charge isoforms with an altered content identified in this study also emphasizes the importance of post-translational modifications upon different cell culture conditions. The differential protein expression reported suggests that changes in the actin cytoskeleton structure regulate MSCs cell adhesion, motility and self-renewal, and that a shift towards the glycolytic pathway occurs for energy requirements and protection against oxidative stress. In general, no major variances were observed between BM and AT with respect to the biological categories they were clustered in. However, as revealed by the proteomic expression profiles, differences between protein forms and protein numbers identified for each category were reported along with differences in the expression levels and number of protein isoforms for each tissue type. The global genome-wide expression approach used in this study has helped obtain mechanistic insights into the response to hypoxia, and in the future might contribute to set up a proteome profiling strategy for developing quality controls to assure clinically effective expanded MSCs.Item Application of variations of non-linear CCA for feature selection in drug sensitivity prediction(2019-06-17) Shadbahr, Tolou; Uurtio, Viivi; Perustieteiden korkeakoulu; Rousu, JuhoCancer arises due to the genetic alteration in patient DNA. Many studies indicate the fact that these alterations vary among patients and can affect the therapeutic effect of cancer treatment dramatically. Therefore, extensive studies focus on understanding these alterations and their effects. Pre-clinical models play an important role in cancer drug discovery and cancer cell lines are one of the main ingredients of these pre-clinical studies which can capture many different aspects of multi-omics properties of cancer cells. However, the assessment of cancer cell line responses to different drugs is faulty and laborious. Therefore, in-silico models, which perform accurate prediction of drug sensitivity values, enhance cancer drug discovery. In the past decade, many computational methods achieved high performances by studying similarity between cancer cell lines and drug compounds and used them to obtain an accurate predictive model for unknown instances. In this thesis, we study the effect of non-linear feature selection through two variations of canonical correlation analysis, KCCA, and HSIC-SCCA, on the prediction of drug sensitivity. To estimate the performance of these features we use pairwise kernel ridge regression to predict the drug sensitivity, measured as IC50 values. The data set under study is a subset of Genomics of Drug Sensitivity in Cancer comprise of 124 cell lines and 124 drug compounds. The high diversity between cell lines and drug compound samples and the high dimension of data matrices reduce the accuracy of the model obtained by pairwise kernel ridge regression. This accuracy reduced by employing HSIC-SCCA method as a dimension reduction step since the HSIC-SCCA method increased the differences among the samples by employing different projection vectors for samples in different folds of cross-validation. Therefore, the obtained variables are rotated to provide more homogeneous samples. This step slightly improved the accuracy of the model.
