Browsing by Title

Tämän tutkimuksen aiheena oli kehittää elektroniikkateollisuuteen liimoja, joilla on keskenään erilaiset elastiset moduulit. Tavoitteena oli muokata liimojen ominaisuuksia muuttamalla käytettyä siloksaanipohjaista polymeeriä. Ensimmäisessä vaiheessa difenyylisilaanidiolista (DPSD) syntetisoitiin dimeerejä ja trimeerejä; bis-DPSD:a ja tris-DPSD:a. Toisessa vaiheessa syntetisoitiin moolilmassaltaan pieniä kopolymeerejä, joiden synteesiin käytettiin kolmea edellämainittua DPSD:a tetrametoksisilaanin (TMOS) ja trimetoksi[2-(7-oksabisyklo[4.1.0]hept-3-yl)etyleeni]silaanin (ECHETMOS) reaktiossa. Ensimmäiseen kopolymeeriin käytettiin DPSD:a, toiseen dimeeristä bis-DPSD:a ja kolmanteen trimeeristä tris-DPSD:a. Kolmannessa vaiheessa kopolymeereistä valmistettiin liimaa lisäämällä kopolymeereihin täyteaineita, additiiveja ja katalyyttiä, jolla valmis liima saatiin kovetettua. Valmis liima saadaan ristisilloittumaan lämpökatalyytillä, kun ECHETMOS:n epoksiryhmät aukeavat ja reagoivat renkaanavautumispolymerisaatiossa. Tutkimuksen kirjallisessa osuudessa perehdyttiin siloksaani- ja silseskvioksaanipolymeerien reaktioihin, sovelluksiin ja niistä valmistettujen materiaalien ominaisuuksiin. Siloksaani- ja silseskvioksaanipolymeerejä käytetään elektroniikkateollisuudessa optisissa sovelluksissa. Ensin keskityttiin silaanien ja siloksaanien optisten ominaisuuksien tutkimukseen. Tämän jälkeen tutkittiin difenyylisilaanidiolin vaikutusta optisiin sovelluksiin käytetyissä siloksaanipolymeereissä. Dimeeristen ja trimeeristen difenyylisilaanidiolien oletettiin vaikuttavan valmiin liiman elastiseen moduuliin, koska ne ovat selvästi pidempiä ja jäykempiä yksiköitä verrattuna monomeeriseen DPSD:n. Työssä kuitenkin havaittiin, että DPSD:n kondensaatioreaktio trimeeriksi ei onnistunut toivotusti. Sen sijaan osoittautui, että tris-DPSD muodosti ainoastaan rengasmaisia rakenteita lineaaristen sijaan, jolloin tris-DPSD:ssa ei ollut jäljellä enää yhtään vapaita OH-ryhmiä, joiden avulla polymeeriä olisi voinut muodostua. Sen takia trimeerien vaikutusta polymeerin rakenteen jäykkyyteen ei päästy tutkimaan. Bis-DPSD säilyi lineaarisena, mutta tris-DPSD:a muodostui kaikissa reaktioissa, joista saatiin bis-DPSD:a. Käytännössä tämä aiheutti kovettumisongelmia liimoissa, koska liimoihin jäi tällöin aina myös monomeerista tris-DPSD:a, eli materiaalia ei saatu täysin ristisilloitettua. Monomeerista DPSD:a käyettäessä kopolymerointi TMOS:n ja ECHETMOS:n kanssa onnistui, ja tästä kopolymeeristä saatiin mekaanisesti kestävää liimaa.

Objectives: To determine if and in what ways serious patient safety incidents differ from non-serious patient safety incidents. Methods: Statistical analysis was performed on patient safety incident reports that were reported in 2015 in Finland’s largest hospital district (Helsinki and Uusimaa, HUS). Reports were divided into two groups: non-serious incidents and serious incidents. Differences between groups were studied from several types of categorically divided information. Results: Of the total amount of reports (15 863) 1 % were serious incidents (175). Serious and non-serious incidents differed significantly from each other. Serious incidents concerning laboratory, imaging or medical equipment were more common. On the other hand incidents concerning medication, infusion, blood transfusion, were less frequent. In serious incidents the proportion of doctors reporting was greater and contributing factors were better recognized, the most common of them being working of procedures. Conclusions: In the future, special attention should be given to the particular aspects of serious patient safety incidents, such as safe use of medical equipment, training and handling of procedures. Root cause analysis is an effective way to handle serious incidents and enables the prevention of their reoccurrence. However, a systematic follow of the root cause analysis should be developed.

Objectives: Depression presents one of the biggest global health concerns today. According to the network theory, mental disorders, such as depression, reflect co-occurring intercorrelating symptom effects. Thus, studying the properties of depressive symptom networks could enhance knowledge about the etiology of depression. In this study, network structures of men and women of the general population are compared, to enhance understanding of higher prevalence rates of depression in women. Although gender differences of depression are widely studied, this is the first study comparing the depressive symptom network structures of adult men and women in the general population. Methods: The data (n = 567 men; n= 886 women) are from a national age cohort study (LASER). Partial correlation networks of BDI-II symptoms were compared in two time-points. Results: Estimated networks had distinct gender differences. Men had lower mean scores, but more changes in the network structure across time, more negative edges, and higher network density. “Agitation” was highly central only in the men’s networks. Women showed changes in mean sum scores, but network structures had few changes, and symptoms formed three distinct communities in both time-points. “Fatigue” was reported significantly more by women and was highly central only in the women’s networks. Implications: Differences in symptom networks between men and women may explain the gender-related differences in the prevalence of depression. Negative edges in the men’s networks and the symptom communities in the women’s networks are targets for more research. Fatigue could be a valuable target for preventing and treating women’s depression.

Equine asthma is a disease syndrome comprised of two diseases, mild and severe asthma. Both diseases can affect the horses performance and intended use. The diseases can be differentiated from each other based on clinical signs, bronchoalveolar lavage fluid (BALF) and tracheal wash (TW) analyses. A horse can recover from mild asthma where as severe asthma is an irreversible disease. Severe asthma is also inheritable, although the exact genes are not known. Environmental factors play an important role in both diseases, with dust, mold, noxious gases and endotoxins being recognized as influencing factors. These components are present in horses’ stable environments. Bedding material in horse stalls is used to absorb moisture and ammonia, and to provide suitable bedding for horses to lay on. The two main bedding materials used in Finland are shavings and peat. They differ in their capability to absorb moisture and ammonia, as well as in their microbiological quality. Peat has higher moisture and ammonia binding capacity but shavings has higher microbiological quality. Horses can spend a big part of their day in the stalls, which increases the importance of using high quality bedding material, which has minimum effect on the respiratory system. The purpose of this study is to compare the effects of peat and shavings on the horses’ respiratory health. 32 horses participated in the study and they were kept 35 days on each bedding material, during three time periods. Peat was used before (peat 1) and after (peat 2) the shavings period. After each 35 day period the horses were examined, BALF and TW samples were collected and analyzed. The results show a significant increase in the TW neutrophil percentage during the shavings period (peat 1: 16.7%, shavings: 32.8%, peat 2: 13.4%). Similar results were noticed in the BALF results, with a significant difference in the neutrophil percentage when comparing shavings and peat 2, but also between the two peat periods (peat 1: 2.7%, shavings: 3.4%, peat 2: 1.6%). None of the horses participating in the study were diagnosed with mild or severe asthma based on the results and clinical signs, even though the high TW neutrophil percentage during the shavings period strongly suggest towards neutrophilic airway inflammation. The results indicate a higher irritation level in the equine respiratory system when shavings was used as bedding material. The difference in the BALF neutrophil percentages during the peat periods may be caused by the horses being taken in from pasture before the peat 1 period. The higher peat 1 results can there fore represent a reaction to the change in environment rather then the actual bedding itself. More research is needed with longer time spent on different bedding materials, combined with exercise tolerance tests, to give insight on the bedding’s effect on physical performance.

Tutkielma käsittelee asenteellista kielenkäyttöä talouspolitiikkaa koskevissa keskusteluissa kahden amerikkalaisen talousblogin kommenttiosioissa. Tutkimuskohteena on asenteellisen kielenkäytön tyyppien sekä polaarisuuden suhteellinen määrä, blogien kommentoijien eroavaisuudet sekä se, mihin kommenteissa ilmaistut asenteet kohdistuvat. Aineisto koostuu yhteensä 194:stä kommentista (22 405 sanaa) jotka on kerätty kahdesta eri blogista (Paul Krugmanin The Conscience of a Liberal ja Donald Boudreaux’n Café Hayek). Blogit valikoituivat niiden toisistaan poikkeavien taloustieteellisten lähestymistapojen ja ideologisten maailmankuvien vuoksi. Kommentit on valittu viittä eri julkistalouden osa-aluetta käsittelevistä blogikirjoituksista. Koska Krugmanin blogikirjoitukset sisältävät huomattavasti enemmän kommentteja, niistä valittiin kommenttien saamien suositteluäänten määrän mukaan neljään osaan jaettu ositettu satunnaisotos. Kommentit analysoitiin Martinin ja Whiten (2005) appraisal-teorian attitude-järjestelmän mukaisesti. Tutkielmassa toteutettu aineiston tarkka lähiluku vertautuu mm. Banksin (2005) mini-korpusanalyysiin. Tämän tutkielman pääanalyysissa kommenteista tunnistettiin ja niihin merkittiin asenteellisen kielen esiintymät, niiden tyyppi sekä asenteellisen kielen kohde ja lähde. Tähän käytettiin UAM Corpus Tool -ohjelmaa. Tämän jälkeen tuloksia käsitellään sekä kvantitatiivisesti että kvalitatiivisesti. Tutkielmassa painottuvat kvalitatiiviset tulokset, koska aineiston asenteellisen kielen kontekstisidonnainen monikerroksisuus on aineiston mielenkiintoisin osa. Aineistossa asenteellista kieltä esiintyy yhteensä 1 882 kertaa, ja se painottuu asioiden ja ilmiöiden (appreciation, 61%), sekä ihmisten tekojen ja ominaisuuksien arviointiin (judgement, 34%). Tunteiden suorat ilmaukset (affect, 5%) ovat huomattavasti vähemmän yleisiä. Tulos viittaa siihen, että asenteellinen kielenkäyttö kommenttiosioissa on pelkkien tunnereaktioiden ilmaisua perustellumpaa. Aineiston asenteellinen kieli on myös 68-prosenttisesti negatiivista. Krugmanin kommenteissa esiintyi 41% epäsuoria asenteen ilmaisuja (invoked attitude), merkittävästi Hayek-otosta enemmän (17%). Krugman-aineistossa käsitellään Yhdysvaltain Demokraatti- ja Republikaanipuoluetta huomattavasti Hayek-aineistoa enemmän, ja markkinavapautta arvotetaan negatiivisemmin (-14/+11). Molemmissa otoksissa käsiteltiin eniten politiikkapäätösten seurauksia sekä Paul Krugmania, molempia pääosin negatiivisesti. Krugmanin kritisointi otoksissa johtuu kuitenkin eri syistä. Hänen omassa blogissaan kritiikkiä synnyttää Krugmanin liian negatiivinen suhtautuminen presidenttiehdokas Bernie Sandersiin ja se, että jotkut pitävät Krugmania liian markkinamyönteisenä. Hayek-yhteisössä kritiikkiä aiheuttaa taas Krugmanin taipumus suosia valtion kontrollia talouden ohjailussa vapaiden markkinoiden sijaan, eli syy on päinvastainen kuin osalla Krugmanin lukijoista. Tutkielman keskeisimpiä huomioita on se, että vaikka taloustiede tieteenalana nojaa rationaalisuuteen ja objektiivisuuteen, keskusteluissa näkyy vahvasti myös talouspolitiikan yksilönläheisempi ja tunteellisempi puoli. Aineiston kommentit sisältävät paljon asenteellista kieltä, joka on luonteeltaan kontekstisidonnaista ja monikerroksista. Vaikka Internet-kommentointi on usein matalatasoista ja polarisoitunutta, niin kommenttien tarkastelu tarjoaa näkökulman niiden kirjoittajien maailmankuviin tavalla, joka on vertaansa vailla maailman historiassa.

Työssä esitellään ääretönharmonisten funktioiden ominaisuuksia. Ääretönharmoniset funktiot ovat äärettömän Laplacen yhtälön viskositeettiratkaisuja. Työn päämääränä on todistaa, että ääretönharmoniset funktiot ovat derivoituvia. Aluksi tutustutaan viskositeettiratkaisun määritelmään. Sen jälkeen esitellään vertailuperiaate kartiofunktioiden suhteen, ja osoitetaan, että ääretönharmoniset funktiot noudattavat kyseistä vertailuperiaatetta. Tällä ominaisuudella on useita hyödyllisiä seurauksia, jotka johdetaan kappaleessa 3. Erityisesti voidaan osoittaa, että funktiot, jotka noudattavat vertailuperiaatetta kartiofunktioiden suhteen ovat lokaalisti Lipschitz-jatkuvia. Kappaleessa 4 tutkitaan muokattua versiota äärettömästä Laplacen yhtälöstä. Muokatun version etuna on se, että ratkaisujen olemassaolo, yksikäsitteisyys ja säännöllisyys seuraavat standardista kvasilineaaristen osittaisdifferentiaaliyhtälöiden teoriasta. Kappaleessa 5 todistetaan, että ääretönharmoniset funktiot ovat derivoituvia. Sitä varten osoitetaan ensin, että ääretönharmoniselle funktiolle löydetään tangenttitaso jokaisessa määrittelyjoukon pisteessä. Tangenttitason olemassaolo seuraa pitkälti kappaleessa 3 johdetuista tuloksista. Lopuksi todistetaan, että ääretönharmonisen funktion tangenttitaso on yksikäsitteinen. Derivoituvuus seuraa suoraan tangenttitason yksikäsitteisyydestä. Kappaleen 4 tulokset muokatun version ratkaisusta ovat keskeisiä työkaluja todistettaessa tangenttitason yksikäsitteisyyttä.

Accumulating evidence indicates that the plasticity-inducing effects of conventional antidepressant drugs like fluoxetine are mediated by their direct binding to TrkB. TrkB is the receptor of the brain-derived neurotrophic factor (BDNF), a neurotrophic factor of critical importance for neuron survival and synaptic plasticity. In addition, it has recently been reported that LSD and psilocybin, two psychedelic compounds with therapeutic potential, also bind to TrkB with higher affinity than antidepressants. It has been proposed that the differences in binding affinity between conventional antidepressants and psychedelics may help explain the much faster and longer-lasting antidepressant effects of psychedelics. Psychedelics and classical antidepressants bind to the transmembrane domain of TrkB dimers, where they act as positive allosteric modulators by potentiating the action of endogenous BDNF. The transmembrane binding sites of LSD and fluoxetine, despite being partially overlapping, are distinct and induce different conformational changes when bound to TrkB dimers. However, it is still unknown whether there are differences in the TrkB dimerization dynamics and neurotrophic signalling pathways induced by psychedelics when compared to conventional antidepressants. In this study, we investigated whether psychedelics and classical antidepressants promote TrkB dimerization and neurotrophic signalling in a differential manner. The effects of psychedelics on the TrkB dimerization dynamics and neurotrophic signalling associated with plasticity were studied treating N2a cells and primary cortical neuronal cultures with LSD or fluoxetine. Dimerization of the TrkB receptor in the presence of experimental compounds is assayed by protein-fragment complementation assay (PCA). Results show a significant dimerization in cells treated with LSD, whereas non-significant response in the ones treated with fluoxetine. The phosphorylation state of the neuronal TrkB receptor in three different tyrosines (Y515, Y706, and Y816) was checked as a marker of its activation by Western blot. Primary cortical cultures were treated with classical antidepressant fluoxetine (10uM) or psychedelic LSD (100nM) for 1 hour, when their effects on TrkB phosphorylation were compared. This experiment showed a significant increase of phosphorylation in TrkB Y816 after LSD treatment in cortical neuronal cultures, while fluoxetine treatment showed no significant effect. This indicates that LSD is able to activate the BDNF-TrkB signalling pathway associated with PLCg1 recruitment and induction of plasticity at an early time point and with a much lower concentration than fluoxetine, which would support LSD’s much more potent antidepressant and plasticity-inducing effects when compared to fluoxetine’s. Together, these results suggest that psychedelics that bind to TrkB, like LSD, are more potent than classical antidepressants in inducing TrkB-BDNF signalling. Overall, this study provides further evidence that TrkB is a critical mediator of psychedelics’ actions on neurotrophic signalling preceding their plasticity-enhancing and antidepressant effects and sheds more light on the common and differential mechanisms used by psychedelics and conventional antidepressants to produce their therapeutic effects.

Sharing data can lead to scientific discoveries, but it can hurt privacy of the people in the data. In this thesis we use deep generative models Generative adversarial network and Variational autoencoder to generate synthetic data, which could be shared instead of the original data. These models are also modified to satisfy the definition of Differential privacy (DP), which is a mathematically rigorous definition of privacy. First we give some essential definitions for DP and proofs for some of them. Then we discuss data sharing and potential privacy risks related to it as well as methods for mitigating these risks. Then we introduce deep generative models and their DP-versions used for creating synthetic data and finally we measure the quality of synthetic data using several continuous or categorical valued data sets.

Differential privacy is a mathematically defined concept of data privacy that is based on the idea that a person should not face any additional harm by opting to give their data to a data collector. Data release mechanisms that satisfy the definition are said to be differentially private and they guarantee the privacy of the data on a specified privacy level by utilising carefully designed randomness that sufficiently masks the participation of each individual in the data set. The introduced randomness decreases the accuracy of the data analysis, but this effect can be diminished by clever algorithmic design. Robust private linear regression algorithm is a differentially private mechanism originally introduced by A. Honkela, M. Das, O. Dikmen, and S. Kaski in 2016. The algorithm is based on projecting the studied data inside known bounds and applying differentially private Laplace mechanism to perturb the sufficient statistics of the Bayesian linear regression model that is then fitted to the data using the privatised statistics. In this thesis, the idea, definitions and the most important theorems and properties of differential privacy are presented and discussed. The robust private linear regression algorithm is then presented in detail, including improvements that are related to determining and handling the parameters of the mechanism and were developed during my work as a research assistant in the Probabilistic Inference and Computational Biology research group (Department of Computer Science at University of Helsinki and Helsinki Institute for Information Technology) in 2016-2017. The performance of the algorithm is evaluated experimentally on both synthetic and real-life data. The latter data are from the Genomics of Drug Sensitivity in Cancer (GDSC) project and consist of the gene expression data of 985 cancer cell lines and their responses to 265 different anti-cancer drugs. The studied algorithm is applied to the GDSC data with the goal of predicting which cancer cell lines are sensitive to each drug and which are not. The application of a differentially private mechanism to the gene expression data is justifiable because genomic data are identifying and carry highly sensitive information about e.g. an individual's phenotype, health, and risk of various diseases. The results presented in the thesis show the studied algorithm works as planned and is able to benefit from having more data: in the sense of prediction accuracy, it approaches the non-private version of the same algorithm as the size of the available data set increases. It also reaches considerably better accuracy than the three compared algorithms that are based on different differentially private mechanisms: private linear regression with no projection, output perturbed linear regression, and functional mechanism linear regression.

In this work, calibration methods, reference compounds and sample introduction system used by various researchers for different ion mobility spectrometer (IMS) techniques have been discussed. Reduced mobility values of positive and negative reference compounds along with the reason for selecting them by different researchers have been described in the literature. Differential mobility spectrometer’s (DMS) performance was evaluated by measuring calibration curves and performing repeatability tests using two test compounds: 2,6-di-tertbutylpyridine (2,6-DtBP) and hexylamine. 2,6-DtBP is often used as a positive reference compound since the mobility of its ions are independent of temperature and humidity of the drift gas. It produces a single mobility peak and due to its high proton affinity, the instrument is sensitive for the analysis of the compound. Hexylamine on the other hand produces proton bound dimer at high concentrations. Both 2,6-DtBP and hexylamine’s peak intensity and peak area showed a linear correlation when plotted against concentration. However, the linearity was only followed up to certain concentration and above this concentration the peak intensity did not follow a linear relationship. Repeatability and calibration plots were studied and compared for both 2,6-DtBP and hexylamine, where both the compounds were diluted with nitrogen gas and nitrogen gas mixed with purified air. The repeatability was good for both the compounds when they were diluted only with nitrogen