Biomedical Event Extraction with Machine Learning

Biomedical natural language processing (BioNLP) is a subfield of natural language processing, an area of computational linguistics concerned with developing programs that work with natural language: written texts and speech. Biomedical relation extraction concerns the detection of semantic relations such as protein-protein interactions (PPI) from scientific texts. The aim is to enhance information retrieval by detecting relations between concepts, not just individual concepts as with a keyword search. In recent years, events have been proposed as a more detailed alternative for simple pairwise PPI relations. Events provide a systematic, structural representation for annotating the content of natural language texts. Events are characterized by annotated trigger words, directed and typed arguments and the ability to nest other events. For example, the sentence “Protein A causes protein B to bind protein C” can be annotated with the nested event structure CAUSE(A, BIND(B, C)). Converted to such formal representations, the information of natural language texts can be used by computational applications. Biomedical event annotations were introduced by the BioInfer and GENIA corpora, and event extraction was popularized by the BioNLP'09 Shared Task on Event Extraction. In this thesis we present a method for automated event extraction, implemented as the Turku Event Extraction System (TEES). A unified graph format is defined for representing event annotations and the problem of extracting complex event structures is decomposed into a number of independent classification tasks. These classification tasks are solved using SVM and RLS classifiers, utilizing rich feature representations built from full dependency parsing. Building on earlier work on pairwise relation extraction and using a generalized graph representation, the resulting TEES system is capable of detecting binary relations as well as complex event structures. We show that this event extraction system has good performance, reaching the first place in the BioNLP'09 Shared Task on Event Extraction. Subsequently, TEES has achieved several first ranks in the BioNLP'11 and BioNLP'13 Shared Tasks, as well as shown competitive performance in the binary relation Drug-Drug Interaction Extraction 2011 and 2013 shared tasks. The Turku Event Extraction System is published as a freely available open-source project, documenting the research in detail as well as making the method available for practical applications. In particular, in this thesis we describe the application of the event extraction method to PubMed-scale text mining, showing how the developed approach not only shows good performance, but is generalizable and applicable to large-scale real-world text mining projects. Finally, we discuss related literature, summarize the contributions of the work and present some thoughts on future directions for biomedical event extraction. This thesis includes and builds on six original research publications. The first of these introduces the analysis of dependency parses that leads to development of TEES. The entries in the three BioNLP Shared Tasks, as well as in the DDIExtraction 2011 task are covered in four publications, and the sixth one demonstrates the application of the system to PubMed-scale text mining.

Rare adverse events associated with oral poliovirus vaccine in Brazil

Oral poliovirus vaccine (OPV) developed by A. Sabin has been effectively used to control poliomyelitis in Brazil, and the last case with the isolation of a wild poliovirus strain occurred in March 1989. Although the vaccine controlled the circulation of wild strains and poliomyelitis cases associated with these strains were not detected during the last eight years, rare cases classified as vaccine-associated paralytic poliomyelitis (VAPP) have been detected. Molecular characterization studies of poliovirus strains isolated from VAPP cases and from healthy contacts have confirmed that the isolates are derived from the Sabin vaccine strains and also detected genomic modifications known or suspected to increase neurovirulence such as mutations and recombination. The molecular characterization of polioviruses isolated during the last eight years from paralysis cases classified as Guillain-Barré (GBS) syndrome and transverse myelitits (TM), and from facial paralysis (FP) cases also confirmed the vaccine origin of the strains and demonstrated mutations known to increase neurovirulence. Analysis of the epidemiologic data of these GBS, TM and FP cases demonstrated that in most of them the last OPV dose was given months or years before the onset of the disease and the isolation of the polioviruses. The temporal association between the isolation of these strains and the GBS, TM and FP suggested that the Sabin vaccine-derived poliovirus strains could also rarely trigger the diseases.

A comparative analysis of preprocessing techniques of cardiac event series for the study of heart rhythm variability using simulated signals

In the present study, using noise-free simulated signals, we performed a comparative examination of several preprocessing techniques that are used to transform the cardiac event series in a regularly sampled time series, appropriate for spectral analysis of heart rhythm variability (HRV). First, a group of noise-free simulated point event series, which represents a time series of heartbeats, was generated by an integral pulse frequency modulation model. In order to evaluate the performance of the preprocessing methods, the differences between the spectra of the preprocessed simulated signals and the true spectrum (spectrum of the model input modulating signals) were surveyed by visual analysis and by contrasting merit indices. It is desired that estimated spectra match the true spectrum as close as possible, showing a minimum of harmonic components and other artifacts. The merit indices proposed to quantify these mismatches were the leakage rate, defined as a measure of leakage components (located outside some narrow windows centered at frequencies of model input modulating signals) with respect to the whole spectral components, and the numbers of leakage components with amplitudes greater than 1%, 5% and 10% of the total spectral components. Our data, obtained from a noise-free simulation, indicate that the utilization of heart rate values instead of heart period values in the derivation of signals representative of heart rhythm results in more accurate spectra. Furthermore, our data support the efficiency of the widely used preprocessing technique based on the convolution of inverse interval function values with a rectangular window, and suggest the preprocessing technique based on a cubic polynomial interpolation of inverse interval function values and succeeding spectral analysis as another efficient and fast method for the analysis of HRV signals

Adverse pregnancy outcome in rats following exposure to a Salacia reticulata (Celastraceae) root extract

The root extract of Salacia reticulata Wight (family: Celastraceae) is used in Sri Lanka by traditional practitioners as a herbal therapy for glycemic control even during pregnancy. It is recognized that some clinically used antidiabetic drugs have harmful effects on pregnancy but the effects of the S. reticulata root extract on reproductive outcome is unknown and deserves examination. We determined the effects of the S. reticulata root extract on the reproductive outcome of Wistar rats (250-260 g) when administered orally (10 g/kg) during early (days 1-7) and mid- (days 7-14) pregnancy. The root extract significantly (P<0.05) enhanced post-implantation losses (control vs treatment: early pregnancy, 4.7 ± 2.4 vs 49.3 ± 13%; mid-pregnancy, 4.7 ± 2.4 vs 41.7 ± 16.1%). Gestational length was unaltered but the pups born had a low birth weight (P<0.05) (early pregnancy, 6.8 ± 0.1 vs 5.3 ± 0.1 g; mid-pregnancy, 6.8 ± 0.1 vs 5.0 ± 0.1 g) and low birth index (P<0.05) (early pregnancy, 95.2 ± 2.4 vs 50.7 ± 12.9%; mid-pregnancy, 95.2 ± 2.4 vs 58.3 ± 16.1%), fetal survival ratio (P<0.05) (early pregnancy, 95.2 ± 2.4 vs 50.7 ± 12.9; mid-pregnancy, 95.2 ± 2.4 vs 58.3 ± 16.1), and viability index (P<0.05) (early pregnancy, 94.9 ± 2.6 vs 49.5 ± 12.5%; mid-pregnancy, 94.9 ± 2.6 vs 57.1 ± 16.1%). However, the root extract was non-teratogenic. We conclude that the S. reticulata root extract can be hazardous to successful pregnancy in women and should not be used in pregnancy complicated by diabetes.

Minor segmental wall motion abnormalities detected in patients with Chagas' disease have adverse prognostic implications

Recent data from our laboratory have shown that patients with the indeterminate form of Chagas' disease can have impairment of left ventricular contractility, as evaluated by the slope of the left ventricle end-systolic pressure-dimension relationship. We also showed that Chagas' disease patients with minimal baseline wall motion abnormalities detected by two-dimensional echocardiography have more intense contractility impairment when compared to patients with the indeterminate form of the disease without this abnormality. The prognostic implications of these findings have not been established. We evaluated 59 patients (37-76 years, mean = 55 years) with different clinical forms of Chagas' disease, who had normal left ventricular global systolic function at baseline (57.6 ± 6.9%) and who had at least one additional echo during clinical follow-up (0.4-17.6; mean 4.6 years). Group 1 consisted of 14 patients with minor baseline left ventricle wall motion abnormalities and group 2 consisted of 45 patients without these abnormalities. During follow-up, global left ventricle systolic function deterioration was observed in 10 group 1 patients (71.4%) and in only 10 group 2 patients (22.2%; P < 0.005). Age and duration of follow-up were not independent determinants of left ventricular function deterioration in these patients. The present data indicate that mild segmental left ventricular wall motion abnormalities are associated with worsening of systolic function in Chagas' disease patients who have normal baseline global systolic performance.

High incidence of adverse reactions to initial antiretroviral therapy in Brazil

A concurrent prospective study was conducted from 2001 to 2003 to assess factors associated with adverse reactions among individuals initiating antiretroviral therapy at two public referral HIV/AIDS centers in Belo Horizonte, MG, Brazil. Adverse reactions were obtained from medical charts reviewed up to 12 months after the first antiretroviral prescription. Cox proportional hazard model was used to perform univariate and multivariate analyses. Relative hazards (RH) were estimated with 95% confidence intervals (CI). Among 397 charts reviewed, 377 (95.0%) had precise information on adverse reactions and initial antiretroviral treatment. Most patients received triple combination regimens including nucleoside reverse transcriptase inhibitors, non-nucleoside reverse transcriptase inhibitors and protease inhibitors. At least one adverse reaction was recorded on 34.5% (N = 130) of the medical charts (0.17 adverse reactions/100 person-day), while nausea (14.5%) and vomiting (13.1%) were the most common ones. Variables independently associated with adverse reactions were: regimens with nevirapine (RH = 1.78; 95% CI = 1.07-2.96), indinavir or indinavir/ritonavir combinations (RH = 2.05; 95% CI = 1.15-3.64), female patients (RH = 1.93; 95% CI = 1.31-2.83), 5 or more outpatient visits (RH = 1.94; 95% CI = 1.25-3.01), non-adherence to antiretroviral therapy (RH = 2.38; 95% CI = 1.62-3.51), and a CD4+ count of 200 to 500 cells/mm³ (RH = 2.66; 95% CI = 1.19-5.90). An independent and negative association was also found for alcohol use (RH = 0.55; 95% CI = 0.33-0.90). Adverse reactions were substantial among participants initiating antiretroviral therapy. Specially elaborated protocols in HIV/AIDS referral centers may improve the diagnosis, management and prevention of adverse reactions, thus contributing to improving adherence to antiretroviral therapy among HIV-infected patients.

Statistical analysis of event-related potential elicited by verb-complement merge in Brazilian Portuguese

An interesting fact about language cognition is that stimulation involving incongruence in the merge operation between verb and complement has often been related to a negative event-related potential (ERP) of augmented amplitude and latency of ca. 400 ms - the N400. Using an automatic ERP latency and amplitude estimator to facilitate the recognition of waves with a low signal-to-noise ratio, the objective of the present study was to study the N400 statistically in 24 volunteers. Stimulation consisted of 80 experimental sentences (40 congruous and 40 incongruous), generated in Brazilian Portuguese, involving two distinct local verb-argument combinations (nominal object and pronominal object series). For each volunteer, the EEG was simultaneously acquired at 20 derivations, topographically localized according to the 10-20 International System. A computerized routine for automatic N400-peak marking (based on the ascendant zero-cross of the first waveform derivative) was applied to the estimated individual ERP waveform for congruous and incongruous sentences in both series for all ERP topographic derivations. Peak-to-peak N400 amplitude was significantly augmented (P < 0.05; one-sided Wilcoxon signed-rank test) due to incongruence in derivations F3, T3, C3, Cz, T5, P3, Pz, and P4 for nominal object series and in P3, Pz and P4 for pronominal object series. The results also indicated high inter-individual variability in ERP waveforms, suggesting that the usual procedure of grand averaging might not be considered a generally adequate approach. Hence, signal processing statistical techniques should be applied in neurolinguistic ERP studies allowing waveform analysis with low signal-to-noise ratio.

Sponsor Satisfaction in a Regional Charity Sport Event

The purpose of this Master’s thesis is to study sponsor satisfaction in charity sport events. Lack of research in regional charity sport events, emergence of corporate social responsibility and increasing popularity of charity sport events have created a research gap to be further explored. Theoretical part of the thesis focuses in development of sponsorships, charity sport event sponsorships and sponsorship as a marketing tool. Concept of satisfaction is discussed by implementing marketing theories to weight options on measuring sponsor satisfaction as a part of sponsorship evaluation process. Empirical analysis of the thesis was conducted in a regional charity sport event – Maailman Pisin Salibandyottelu. Evidences were collected in qualitative research method through semi-structured theme interviews. Altogether 12 major and minor sponsors were selected for the primary source of data. The data was analyzed by comparing sponsors’ expectations and experiences, and by displaying sponsors’ perceived satisfaction. The results indicated that sponsors were involved by partly altruistic and partly selfish motives as suggested by previous research. Respondents expressed very few, mainly non-financial expectations, yet were hoping to gain positive image association via event exposure. Negative experiences appear to have relatively small impact in overall satisfaction. Exceeding or fulfilling expectations appears to increase perceived satisfaction which was mainly driven by contribution towards the goodwill, perceived success of the event (successful record attempt, visibility (on- and off-line) and event execution.

Highly sensitive C-reactive protein and male gender are independently related to the severity of coronary disease in patients with metabolic syndrome and an acute coronary event

Patients with metabolic syndrome are at high-risk for development of atherosclerosis and cardiovascular events. The objective of this study was to examine the major determinants of coronary disease severity, including those coronary risk factors associated with metabolic syndrome, during the early period after an acute coronary episode. We tested the hypothesis that inflammatory markers, especially highly sensitive C-reactive protein (hsCRP), are related to coronary atherosclerosis, in addition to traditional coronary risk factors. Subjects of both genders aged 30 to 75 years (N = 116) were prospectively included if they had suffered a recent acute coronary syndrome (acute myocardial infarction or unstable angina pectoris requiring hospitalization) and if they had metabolic syndrome diagnosed according to the National Cholesterol Education Program/Adult Treatment Panel III. Patients were submitted to a coronary angiography and the burden of atherosclerosis was estimated by the Gensini score. The severity of coronary disease was correlated (Spearman’s or Pearson’s coefficient) with gender (r = 0.291, P = 0.008), age (r = 0.218, P = 0.048), hsCRP (r = 0.256, P = 0.020), ApoB/ApoA ratio (r = 0.233, P = 0.041), and carotid intima-media thickness (r = 0.236, P = 0.041). After multiple linear regression, only male gender (P = 0.046) and hsCRP (P = 0.012) remained independently associated with the Gensini score. In this high-risk population, male gender and high levels of hsCRP, two variables that can be easily obtained, were associated with more extensive coronary disease, identifying patients with the highest potential of developing new coronary events.

A high-carbohydrate diet enhances the adverse effect of the S2 allele of APOC3 SstI polymorphism on the TG/HDL-C ratio only in young Chinese females

Both genetic background and diet have profound effects on plasma lipid profiles. We hypothesized that a high-carbohydrate (high-CHO) diet may affect the ratios of serum lipids and apolipoproteins (apo) differently in subjects with different genotypes of the SstI polymorphism in the apoCIII gene (APOC3). Fifty-six healthy university students (27 males and 29 females, 22.89 ± 1.80 years) were given a washout diet of 54% carbohydrate for 7 days, followed by a high-CHO diet of 70% carbohydrate for 6 days without total energy restriction. Serum triglyceride (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), apoB100, apoAI, and the APOC3 SstI polymorphism were analyzed. The ratios of serum lipids and apoB100/apoAI were calculated. At baseline, the TG/HDL-C ratio was significantly higher in females, but not in males, with the S2 allele. The differences in the TG/HDL-C ratio between genotypes remained the same after the washout and the high-CHO diet in females. When compared with those before the high-CHO diet, the TC/HDL-C (male S2 carriers: 3.13 ± 1.00 vs 2.36 ± 0.65, P = 0.000; male subjects with the S1S1 genotype: 2.97 ± 0.74 vs 2.09 ± 0.55, P = 0.000; female S2 carriers: 2.68 ± 0.36 vs 2.24 ± 0.37, P = 0.004; female subjects with the S1S1 genotype: 2.69 ± 0.41 vs 2.09 ± 0.31, P = 0.000) and LDL-C/HDL-C (male S2 carriers: 1.44 ± 0.71 vs 1.06 ± 0.26, P = 0.012; male subjects with the S1S1 genotype: 1.35 ± 0.61 vs 1.01 ± 0.29, P = 0.005; female S2 carriers: 1.18 ± 0.33 vs 1.00 ± 0.18, P = 0.049; female subjects with the S1S1 genotype: 1.18 ± 0.35 vs 1.04 ± 0.19, P = 0.026) ratios were significantly decreased after the high-CHO diet regardless of gender and of genotype of the APOC3 SstI polymorphism. However, in female S2 carriers, the TG/HDL-C (1.38 ± 0.46 vs 1.63 ± 0.70, P = 0.039) ratio was significantly increased after the high-CHO diet. In conclusion, the high-CHO diet has favorable effects on the TC/HDL-C and LDL-C/HDL-C ratios regardless of gender and of genotype of the APOC3 SstI polymorphism. Somehow, it enhanced the adverse effect of the S2 allele on the TG/HDL-C ratio only in females.

Markers of insulin resistance and sedentary lifestyle are predictors of preeclampsia in women with adverse obstetric results

Some thrombophilias and severe preeclampsia may increase the risk for preterm deliveries and fetal death due to placental insufficiency. Our objective was to evaluate clinical and laboratory data as predictors of preeclampsia in a population of mothers with 3rd trimester fetal losses or preterm deliveries. In a longitudinal retrospective study, 54 consecutive women (age range: 16 to 39 years) with normotensive pregnancies were compared to 79 consecutive women with preeclampsia (age range: 16 to 43 years). Weight accrual rate (WAR) was arbitrarily defined as weight gain from age 18 years to the beginning of pregnancy divided by elapsed years. Independent predictors of preeclampsia were past history of oligomenorrhea, WAR >0.8 kg/years, pre-pregnancy or 1st trimester triglyceridemia >150 mg/dL, and elevated acanthosis nigricans in the neck. In a multivariate logistic regression model, two or more predictors conferred an odds ratio of 15 (95%CI [5.9-37]; P < 0.001) to develop preeclampsia (85% specificity, 73% sensitivity, c-statistic of 81 ± 4%; P < 0.0001). Clinical markers related to insulin resistance and sedentary lifestyles are strong independent predictors of preeclampsia in mothers with 3rd trimester fetal losses or preterm deliveries due to placental insufficiency. Women at risk for preeclampsia in this particular population might benefit from measures focused on overcoming insulin resistance.

Attentional bias modulation by reappraisal in patients with generalized anxiety disorder: an event-related potential study

Affective states influence subsequent attention allocation. We evaluated emotional negativity bias modulation by reappraisal in patients with generalized anxiety disorder (GAD) relative to normal controls. Event-related potential (ERP) recordings were obtained, and changes in P200 and P300 amplitudes in response to negative or neutral words were noted after decreasing negative emotion or establishing a neutral condition. We found that in GAD patients only, the mean P200 amplitude after negative word presentation was much higher than after the presentation of neutral words. In normal controls, after downregulation of negative emotion, the mean P300 amplitude in response to negative words was much lower than after neutral words, and this was significant in both the left and right regions. In GAD patients, the negative bias remained prominent and was not affected by reappraisal at the early stage. Reappraisal was observed to have a lateralized effect at the late stage.

The Impact of Event Marketing on Brand Awareness in eSport Industry

The purpose of this qualitative research is to study what is the impact of event marketing on brand awareness in the context of electronic sport industry. Based on the research questions, the theoretical framework will be developed. This research will analyze earlier theories, and also searching more fresh literature to explain the current phenomenon in the eSport industry. In the empirical part, there were total of five case companies interviewed. The context of this research is eSport, which has its own chapter. The theoretical part of the thesis focuses on event marketing and brand awareness. In this research, event marketing is analyzed from the event organizers perspective. In some occasions, event exhibitors’ perspective is also analyzed. In brand awareness, the focus is how to create a brand recognizable, recalled and from there top of mind in consumers’ minds. The results of this research revealed that many companies’ struggles on getting their brand recognizable. Some of the case companies lacks a strategy and don’t exactly know the core values of their customers. However some of the case companies were opposite. One reason behind this is that some of them has experience on the field and the companies have resources that covers them. Also the current strong brand has clearly a positive affect on their business.

Cholesterol, cardiovascular disease and statin use in older persons

Older age increases the risk of developing a chronic atherosclerotic cardiovascular disease (CVD), such as coronary heart disease. Complications of CVDs, myocardial infarction or stroke often lead to loss of functional capacity or premature death. Dyslipidemia, high serum levels of total or low-density lipoprotein cholesterol (LDL-c) and low levels of high-density lipoprotein cholesterol (HDL-c), is among the most important modifiable risk factors for CVDs; it can be treated with lifestyle modifications, and with lipid-lowering drugs, primarily statins. In older persons, however, the association of cholesterol levels with cardiovascular and all-cause mortality has been inconsistent in previous studies. Furthermore, the beneficial effects of statins in older persons without previous CVD are still somewhat unclear, and older persons are more prone to adverse effects from statins. This thesis presents a prospective cohort study (TUVA), exploring associations of cholesterol levels with mortality and the changes in cholesterol levels of a 70-year-old population in long-term follow-up. Further, prevalence of CVDs, risk factors and preventive medication use in the TUVA cohort is compared with respective prevalences in another age-matched cohort (UTUVA) 20 years later in order to examine the changes in cardiovascular risk over time. Additionally, to evaluate statin use patterns among older persons, an observational register study was conducted covering the total Finnish population aged 70 and older during 2000-2008. Based on individual-level data retrieved from national health registries, the population was classified into low, moderate and high risk groups according to estimated CVD risk. The prevalence, incidence and persistence of statin use among the risk groups was then evaluated based upon yearly statin purchases tracked from the Prescription Register. The prospective cohort study demonstrated that low total cholesterol, LDL-c and HDL-c were associated with higher mortality in a cohort of home-dwelling 70-year-olds. However, after adjusting for traditional cardiovascular risk factors and cancer this association disappeared. Further, low total cholesterol seemed to be protective, whereas low HDL-c strongly predicted increased risk of CVD death. Cholesterol levels of those elderly who remained available for follow-up and were still home-dwelling at the age of 85 seemed to improve with advancing age. Compared to the TUVA cohort, the later born UTUVA cohort had less CVDs and their risk factors were better controlled, which was reflected in the higher use of preventive medications such as statins and antihypertensives. The register studies confirmed that statin use has increased significantly during 2000-2008 among older persons, especially among the oldest (80+) age groups and among those at high risk for cardiovascular events. Two-thirds of new statin users persisted with their use during the four years of follow-up; the most discontinuations were made during the first year of use. In conclusion, statins are commonly used among older age groups in Finland. Most of the older statin users had a high cardiovascular event risk, indicating that the treatment is well directed towards those who are likely to benefit from it the most. No age-limits should be put on the screening and treatment of dyslipidemia in older persons, but the benefits and adverse effects of statin treatment should be carefully weighed based on an individual assessment of the person’s general health status and functional capacity. Physicians should pay more attention to medication adherence, especially when prescribing preventive medications.