Abstract Cardiovascular diseases are the leading cause of death in developed countries. Approximately 50% of these events are due to sudden cardiac death (SCD) and often without preceding diagnosis of cardiac disease. Many risk factors for cardiac events have been identified and prevention strategies have improved markedly. The aim of this thesis was to evaluate the usability of the 12-lead electrocardiogram (ECG) to predict cardiac events. The study population consisted of 10,904 middle-aged general population subjects with ECG recordings between the years 1966–1972 with a long follow-up (30±11 years). The first part of the thesis (I) focused on the prevalence and prognostic significance of fragmented QRS complex (fQRS). The prevalence of fQRS was 19.7%. Fragmented QRS complex did not predict mortality in subjects with no history of cardiac disease. Among subjects with underlying cardiac disease and lateral fQRS, the risk of cardiac death was 2.5-fold (P=0.001) and the risk of SCD was almost 3-fold (P=0.004). Other major electrocardiographic abnormalities were assessed in subjects without known cardiac disease for the risk of cardiac death, SCD and hospitalization due to coronary artery disease (II, III). Abnormal ECG was moderately associated with cardiac death after 10 and 30 years of follow-up (hazard ratio 1.7, P=0.009; hazard ratio 1.3, P>0.001, respectively) (II). The risk of hospitalization was not associated with abnormal ECG findings. Abnormal ECG moderately predicted SCD during 10 and 30 years of follow-up (hazard ratio 1.6, P=0.052; hazard ratio 1.3, P=0.007) (III). The risk of SCD was 3-fold when ≥2 ECG abnormalities were present. In conclusion, lateral fQRS in middle-aged subjects with underlying cardiac disease was associated with increased risk of death. Certain abnormal ECG findings associated with the risk of non-arrhythmic cardiac mortality and arrhythmic death. The risk of arrhythmic mortality was substantially elevated when multiple ECG abnormalities were present in middle-aged population. / Tiivistelmä Sydänsairaudet ovat yleisin kuolinsyy kehittyneissä maissa. Noin 50 % näistä kuolemista aiheutuu äkillisestä sydänpysähdyksestä, suuri osa ilman aiempaa tietoa sairaudesta. Useita sydänsairauksien riskitekijöitä on tunnistettu ja ennaltaehkäisy on kehittynyt merkittävästi. Väitöstutkimuksen tavoitteena on tutkia 12-kytkentäisen sydänsähkökäyrän (EKG) käyttökelpoisuutta sydänsairauksien ilmenemisen ennustamisessa. Tutkimusväestöön kuului 10,904 keski-ikäistä suomalaista henkilöä. Aineisto kerättiin vuosina 1966-1972 ja seuranta-aika oli 30 (±11) vuotta. Ensimmäisessä osajulkaisussa (I) tutkimme QRS-kompleksin fragmentaation vallitsevuutta ja sen vaikutusta ennusteeseen väestössä. Fragmentoituneen QRS-kompleksin esiintyvyys oli 19.7 %. Fragmentoitunut QRS-kompleksi ei lisännyt kuolemanriskiä henkilöillä, joilla ei ollut sydänsairautta. Henkilöillä, joilla oli todettu sydänsairaus, lateraalinen fQRS lisäsi sydänperäistä kuolleisuutta 2.5-kertaiseksi (P=0.001) ja rytmihäiriöperäistä kuolleisuutta 3-kertaiseksi (P=0.004). Tutkimme muiden poikkeavien EKG-löydösten ennustearvoa kuolleisuuteen ja sairaalahoidon tarpeeseen sepelvaltimokohtauksen vuoksi (II, III). Poikkeavien EKG-muutosten esiintymiseen liittyi lisääntyneen sydänperäisen kuoleman riski sekä 10 vuoden (riskitiheyssuhde 1.7, P=0.009) että 30 vuoden seurannassa (riskitiheyssuhde 1.3, P>0.001) (II). Poikkeavat EKG-muutokset eivät ennustaneet sairaalahoitojaksoja. Poikkeava EKG ennusti rytmihäiriöperäisen kuoleman riskiä sekä 10 vuoden (riskitiheyssuhde 1.6, P=0.052) että 30 vuoden seurannassa (riskitiheyssuhde 1.3, P=0.007) (III). Äkkikuoleman riski oli 3-kertainen henkilöillä, joilla todettiin ≥ 2 EKG-poikkeavuutta. Tutkimuksen yhteenvetona voidaan todeta, että fQRS lateraalisissa kytkennöissä lisäsi sydänperäisen kuoleman riskiä henkilöillä, joilla on todettu sydänsairaus. Tiettyihin poikkeaviin EKG-muutoksiin liittyi lisääntynyt ei-rytmihäiriöperäisen ja rytmihäiriöperäisen kuoleman riski. Useiden tutkittujen EKG-muutosten ilmentyminen samanaikaisesti lisäsi merkittävästi rytmihäiriöperäisen kuoleman riskiä keski-ikäisessä väestössä.

Background: The prognostic significance of chest pain is well established in patients with coronary artery disease, but still ill defined in primary prevention. Therefore, the aim of our analysis was to assess the prognostic value of different forms of chest pain in a large cohort of primary care subjects under the conditions of contemporary modalities of care in primary prevention, including measurement of serum levels of the biomarker NT-pro-BNP. Design: We carried out a post-hoc analysis of the prospective DETECT cohort study. Methods: In a total of 5570 unselected subjects, free of coronary artery disease, within the 55,518 participants of the cross-sectional DETECT study, we assessed chest pain history by a comprehensive questionnaire and measured serum NT-pro-BNP levels. Three types of chest pain, which were any chest pain, exertional chest pain and classical angina, were defined. Major adverse cardiovascular events (MACEs = cardiovascular death, myocardial infarction, coronary revascularization procedures) were assessed during a 5-year follow-up period. Results: During follow-up, 109 subjects experienced a MACE. All types of reported chest pain were associated with an approximately three-fold increased risk for the occurrence of incident MACEs, even after adjusting for cardiovascular risk factors. Any form of reported chest pain had a similar predictive value for MACEs as a one-time measurement of NT-pro-BNP. However, adding a single measurement of NT-pro-BNP and the information on chest pain resulted in reclassification of approximately 40% of subjects, when compared with risk prediction based on established cardiovascular risk factors. Conclusions: In primary prevention, self-reported chest pain and a single measurement of NT-pro-BNP substantially improve cardiovascular risk prediction and allow for risk reclassification of approximately 40% of the subjects compared with assessing classical cardiovascular risk factors alone.

No description available.

Introduction : La cardiopathie arythmogène du ventricule droit (CAVD) est une pathologie d’origine génétique se traduisant par un remplacement cicatriciel qui affecte de façon prédominante le ventricule droit (VD). Le diagnostic est complexe car il repose sur un ensemble de critères cliniques plutôt que sur un seul test diagnostic. L’atteinte du VD se traduit de façon prédominante par des arythmies ventriculaires qui peuvent parfois conduire à la complication la plus redoutée de cette affection : la mort subite. La prédiction et la prévention de celle-ci sont des enjeux cruciaux de la prise en charge de cette maladie. Objectifs : Ce travail vise à améliorer la prise en charge des patients atteints de CAVD de deux façons distinctes. Premièrement, en tentant de faciliter le diagnostic par la validation des critères diagnostiques en vigueur. Deuxièmement, en améliorant la stratification du risque d’arythmie ventriculaires soutenues et plus spécifiquement celui de la mort subite et des arythmies potentiellement mortelles (tachycardie ventriculaire > 250 bpm, fibrillation ventriculaire) en créant des modèles de prédiction du risque permettant de déterminer le risque individuel de chaque patient. Résultats : Article 1 - Un total de 407 patients consécutifs référés pour une résonnance magnétique cardiaque pour suspicion de CAVD ont été inclus. De ceux-ci, 66 (16%) ont reçu un diagnostic définitif selon le critère de référence établi pour cette étude: le consensus d’un panel d’experts. Globalement, les critères performent bien avec une sensibilité et spécificité à 92%. Cependant, certains critères tels l’ECG haute amplitude (SAECG) et certains critères reliés à l’histoire familiale ne sont pas discriminants. Le retrait de ces critères pourrait réduire le nombre de faux positifs sans pour autant augmenter le nombre de faux négatifs (net reclassification improvement de 4,3%, p=0,019). De plus, la combinaison des critères électrocardiographiques et de la présence d’arythmies ventriculaires a une sensibilité de 100%, ce qui peut faciliter dans certains cas le dépistage en limitant la nécessité de recourir à l’imagerie. Pour les articles 2 et 3, une base de données incluant des patients avec un diagnostic définitif de CAVD a été assemblée à partir de bases de données provenant de six pays (Canada, États-Unis, Pays-Bas, Suède, Norvège, Suisse). Article 2 - Un total de 528 patients sans histoire antérieure d’arythmies ventriculaires soutenues a été inclus pour développer un modèle de prédiction de risque. De ceux-ci, 146 (27,7%) ont subi un événement arythmique durant un suivi médian de 4,8 ans. Des huit prédicteurs initialement identifiés (âge inférieur au diagnostic, sexe masculin, syncope cardiaque récente, nombre de dérivations avec des inversions des ondes T, fardeau d’extrasystoles ventriculaires (ESV) en 24h, tachycardie ventriculaire non-soutenue et fractions d’éjection des ventricules gauche et droit), sept ont été retenus dans le modèle, excluant seulement la fraction d’éjection du ventricule gauche (FEVG). Le modèle peut distinguer adéquatement entre les patients avec et sans événement (C-index de 0,77) avec un optimisme minimal (courbe de calibration de 0,93). L’utilisation de cet algorithme permettrait de réduire l’utilisation de défibrillateurs implantables de 20% par rapport à l’algorithme du consensus le plus largement utilisé. Article 3 - Une cohorte de 864 patients incluant à la fois ceux avec et sans histoire antérieure d’arythmie ventriculaire soutenue a été assemblée. Durant un suivi médian de 5,75 ans, 93 patients ont eu un épisode d’arythmie rapide selon la définition préalablement établie. Des huit facteurs de risque cités ci-haut, seulement quatre ont été retenus dans le modèle : l’âge plus jeune au diagnostic, sexe masculin, fardeau d’ESV en 24h et nombre de dérivations avec des inversions des ondes T. Fait à noter, les événements antérieurs ne se sont pas avérés prédicteurs d’arythmies potentiellement mortelles subséquentes. Le modèle peut distinguer adéquatement entre les patients avec et sans événement (C-index de 0,74) et présente un optimisme minimal avec une courbe de calibration de 0,95. Conclusion : Bien que les critères diagnostiques en vigueur pour la CAVD aient une performance adéquate, ceux-ci peuvent être simplifiés et améliorés par le retrait de certains de ces critères. L’absence de critères électrocardiographiques combinés et d’arythmies ventriculaires peut exclure une CAVD, ce qui peut en simplifier le dépistage. Chez les patients atteints de CAVD, la prédiction du risque et la sélection des patients pour l’implantation d’un défibrillateur peuvent être facilités grâce à deux modèles complémentaires de prédiction du risque permettant de prédire les événements arythmiques soutenus dans le premier et plus spécifiquement les arythmies ventriculaires potentiellement mortelles dans le deuxième. Ces outils sont particulièrement utiles dans une approche de prise de décision partagée. / Introduction: Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a genetic pathology resulting in a fibro-fatty replacement predominantly affecting the right ventricle. The diagnosis is complex and is based on a set of clinical criteria. Involvement of the right ventricle predominantly results in ventricular arrhythmias which constitutes the most common presentation but can also lead to the most feared consequence: sudden cardiac death. Predicting and preventing this catastrophic outcome are crucial in the management of this disease. Objectives: This work aims to improve the management of patients with ARVC in two distinct ways. First, by attempting to facilitate the diagnosis by validating the currently used diagnostic criteria. Second by improving risk stratification for sustained ventricular arrhythmias and specifically life-threatening ventricular arrhythmias (LTVA defined as ventricular tachycardia > 250 bpm, ventricular fibrillation, and sudden death) by creating risk prediction models to derive individual risk. Results: Manuscript 1: a total of 407 patients referred for cardiac magnetic resonance for suspected ARVC were consecutively enrolled. Of these, 66 (16%) received a definitive diagnosis of ARVC by the determined endpoint: the consensus of an expert panel. Overall, the criteria performed well with a sensitivity and specificity of 92%. However, certain criteria such as the signal averaged electrocardiogram (SAECG) and certain criteria related to family history failed to discriminate. Removing these criteria could reduce false positives without increasing false negatives (net reclassification improvement of 4.3%, P = 0.019). In addition, the electrocardiographic criteria and the presence of arrhythmia had a sensitivity of 100%, which can facilitate screening in some cases by making imaging optional. For manuscripts 2 and 3, a cohort including patients with a definitive diagnosis of ARVC was assembled from databases in 6 countries (Canada, United States, Netherlands, Sweden, Norway, Switzerland). Manuscript 2: a total of 528 patients with no previous history of sustained ventricular arrhythmias were included to develop a risk prediction model. Of these, 146 (27.7%) had an arrhythmic event during a median follow-up of 4.8 years. Of the eight predictors initially identified (younger age at diagnosis, male sex, recent cardiac syncope, the number of leads with T wave inversions on the ECG, burden of extrasystoles in 24 hours, non-sustained ventricular tachycardia and left and right ventricular ejection fraction), seven were retained in the model, excluding only left ventricular ejection fraction. The model adequately distinguished between patients with and without an arrhythmic event (C-index of 0.77) with minimal optimism (calibration slope of 0.93). Using this prediction model would reduce the use of defibrillators by 20% compared with the most commonly used consensus based on a risk factor approach. Manuscript 3: a cohort including both patients with and without a prior history of ventricular arrhythmia of 864 patients was assembled. During a follow-up of 5.75 years, 93 patients had an LTVA as defined above. Of the 8 risk factors cited above, only 4 were retained in the model: younger age at diagnosis, male sex, burden of extrasystoles in 24 hours and number of leads with T-wave inversions. Importantly, previous events are not predictive of these subsequent life-threatening arrhythmias. The model adequately distinguished between patients with and without an event (C-index of 0.74) with minimal optimism (calibration slope of 0.95). Conclusion: Although the current diagnostic criteria for ARVC perform adequately, they can be simplified and improved by removing underperforming individual criteria. The absence of any ECG criteria and ventricular arrhythmias may rule out ARVC, which may simplify screening. In patients with ARVC, risk prediction and patient selection for a defibrillator can be facilitated by two complementary risk prediction models for sustained arrhythmic events or more specifically for LTVA. These tools are particularly useful in a shared decision-making approach for implantable cardioverter defibrillator implantation.

This thesis tackles the problem of predicting the collision risk for vehicles driving in complex traffic scenes for a few seconds into the future. The method is based on previous research using dynamic Bayesian networks to represent the state of the system. Common risk prediction methods are often categorized into three different groups depending on their abstraction level. The most complex of these are interaction-aware models which take driver interactions into account. These models often suffer from high computational complexity which is a key limitation in practical use. The model studied in this work takes interactions between drivers into account by considering driver intentions and the traffic rules in the scene. The state of the traffic scene used in the model contains the physical state of vehicles, the intentions of drivers and the expected behaviour of drivers according to the traffic rules. To allow for real-time risk assessment, an approximate inference of the state given the noisy sensor measurements is done using sequential importance resampling. Two different measures of risk are studied. The first is based on driver intentions not matching the expected maneuver, which in turn could lead to a dangerous situation. The second measure is based on a trajectory prediction step and uses the two measures time to collision (TTC) and time to critical collision probability (TTCCP). The implemented model can be applied in complex traffic scenarios with numerous participants. In this work, we focus on intersection and roundabout scenarios. The model is tested on simulated and real data from these scenarios. %Simulations of these scenarios is used to test the model. In these qualitative tests, the model was able to correctly identify collisions a few seconds before they occur and is also able to avoid false positives by detecting the vehicles that will give way. / Detta arbete behandlar problemet att förutsäga kollisionsrisken för fordon som kör i komplexa trafikscenarier för några sekunder i framtiden. Metoden är baserad på tidigare forskning där dynamiska Bayesianska nätverk används för att representera systemets tillstånd. Vanliga riskprognosmetoder kategoriseras ofta i tre olika grupper beroende på deras abstraktionsnivå. De mest komplexa av dessa är interaktionsmedvetna modeller som tar hänsyn till förarnas interaktioner. Dessa modeller lider ofta av hög beräkningskomplexitet, vilket är en svår begränsning när det kommer till praktisk användning. Modellen som studeras i detta arbete tar hänsyn till interaktioner mellan förare genom att beakta förarnas avsikter och trafikreglerna i scenen. Tillståndet i trafikscenen som används i modellen innehåller fordonets fysiska tillstånd, förarnas avsikter och förarnas förväntade beteende enligt trafikreglerna. För att möjliggöra riskbedömning i realtid görs en approximativ inferens av tillståndet givet den brusiga sensordatan med hjälp av sekventiell vägd simulering. Två olika mått på risk studeras. Det första är baserat på förarnas avsikter, närmare bestämt att ta reda på om de inte överensstämmer med den förväntade manövern, vilket då skulle kunna leda till en farlig situation. Det andra riskmåttet är baserat på ett prediktionssteg som använder sig av time to collision (TTC) och time to critical collision probability (TTCCP). Den implementerade modellen kan tillämpas i komplexa trafikscenarier med många fordon. I detta arbete fokuserar vi på scerarier i korsningar och rondeller. Modellen testas på simulerad och verklig data från dessa scenarier. I dessa kvalitativa tester kunde modellen korrekt identifiera kollisioner några få sekunder innan de inträffade. Den kunde också undvika falsklarm genom att lista ut vilka fordon som kommer att lämna företräde.

No / To build an early warning score (EWS) based exclusively on routinely undertaken laboratory tests that might provide early discrimination of in-hospital death and could be easily implemented on paper. Using a database of combined haematology and biochemistry results for 86,472 discharged adult patients for whom the admission specialty was Medicine, we used decision tree (DT) analysis to generate a laboratory decision tree early warning score (LDT-EWS) for each gender. LDT-EWS was developed for a single set (n=3496) (Q1) and validated in 22 other discrete sets each of three months long (Q2, Q3...Q23) (total n=82,976; range of n=3428 to 4093) by testing its ability to discriminate in-hospital death using the area under the receiver-operating characteristic (AUROC) curve. The data generated slightly different models for male and female patients. The ranges of AUROC values (95% CI) for LDT-EWS with in-hospital death as the outcome for the validation sets Q2-Q23 were: 0.755 (0.727-0.783) (Q16) to 0.801 (0.776-0.826) [all patients combined, n=82,976]; 0.744 (0.704-0.784, Q16) to 0.824 (0.792-0.856, Q2) [39,591 males]; and 0.742 (0.707-0.777, Q10) to 0.826 (0.796-0.856, Q12) [43,385 females]. CONCLUSIONS: This study provides evidence that the results of commonly measured laboratory tests collected soon after hospital admission can be represented in a simple, paper-based EWS (LDT-EWS) to discriminate in-hospital mortality. We hypothesise that, with appropriate modification, it might be possible to extend the use of LDT-EWS throughout the patient's hospital stay.

The worldwide increasing number of persons affected by largely preventable diseases like diabetes demands better prevention and treatment. Insulin is required for effective utilisation of circulating nutrients. Impaired responsiveness to insulin (insulin resistance, IR) is a hallmark of type 2 diabetes and independently raises the risk of heart attack and stroke. The pathophysiology of IR is incompletely understood. High-throughput measurement of large numbers of circulating biomarkers may provide new insights beyond established risk factors. The aims of this thesis were to (i) use proteomics, metabolomics and genomics methods in large community samples to identify biomarkers of IR; (ii) assess biomarkers for risk prediction and insights into aetiology and consequences of IR; and (iii) use Mendelian randomisation analysis to assess causality. In Study I, analysis of 80 circulating proteins in 70-to-77-year-old Swedes identified cathepsin D as a biomarker for IR and highlighted a tentative causal effect of IR on raised plasma tissue plasminogen activator levels. In Study II, nontargeted fasting plasma metabolomics was used to discover 52 metabolites associated with glycaemic traits in non-diabetic 70-year-old men. Replication in independent samples of several thousand persons provided evidence for a causal effect of IR on reduced plasma oleic acid and palmitoleic acid levels. In Study III, nontargeted metabolomics in plasma samples obtained at three time points during an oral glucose challenge in 70-year-old men identified associations between a physiologic measure of IR and concentration changes in medium-chain acylcarnitines, monounsaturated fatty acids, bile acids and lysophosphatidylethanolamines. Study IV provided evidence in two large longitudinal cohorts for causal effects of type 2 diabetes and impaired insulin secretion on raised coronary artery disease risk. In conclusion, the Studies in this thesis provide new insights into the pathophysiology and adverse health consequences of IR and illustrate the value of combining traditional epidemiologic designs with recent molecular techniques and bioinformatics methods. The results provide limited evidence for the role of circulating proteins and small molecules in IR and require replication in separate studies and validation in experimental designs.

Saltmarshes provide a diverse range of ecosystem services and are protected under a number of international designations. Nevertheless they are generally declining in extent in the United Kingdom and North West Europe. The drivers of this decline are complex and poorly understood. When considering mitigation and management for future ecosystem service provision it will be important to understand why, where, and to what extent decline is likely to occur. Few studies have attempted to forecast saltmarsh morphodynamics at a system level over decadal time scales. There is no synthesis of existing knowledge available for specific site predictions nor is there a formalised framework for individual site assessment and management. This project evaluates the extent to which machine learning model approaches (boosted regression trees, neural networks and Bayesian networks) can facilitate synthesis of information and prediction of decadal-scale morphological tendencies of saltmarshes. Importantly, data-driven predictions are independent of the assumptions underlying physically-based models, and therefore offer an additional opportunity to crossvalidate between two paradigms. Marsh margins and interiors are both considered but are treated separately since they are regarded as being sensitive to different process suites. The study therefore identifies factors likely to control morphological trajectories and develops geospatial methodologies to derive proxy measures relating to controls or processes. These metrics are developed at a high spatial density in the order of tens of metres allowing for the resolution of fine-scale behavioural differences. Conventional statistical approaches, as have been previously adopted, are applied to the dataset to assess consistency with previous findings, with some agreement being found. The data are subsequently used to train and compare three types of machine learning model. Boosted regression trees outperform the other two methods in this context. The resulting models are able to explain more than 95% of the variance in marginal changes and 91% for internal dynamics. Models are selected based on validation performance and are then queried with realistic future scenarios which represent altered input conditions that may arise as a consequence of future environmental change. Responses to these scenarios are evaluated, suggesting system sensitivity to all scenarios tested and offering a high degree of spatial detail in responses. While mechanistic interpretation of some responses is challenging, process-based justifications are offered for many of the observed behaviours, providing confidence that the results are realistic. The work demonstrates a potentially powerful alternative (and complement) to current morphodynamic models that can be applied over large areas with relative ease, compared to numerical implementations. Powerful analyses with broad scope are now available to the field of coastal geomorphology through the combination of spatial data streams and machine learning. Such methods are shown to be of great potential value in support of applied management and monitoring interventions.

Abstract Critical Infrastructures (CIs) are an integral part of our society and economy. Services like electricity supply or telecommunication services are expected to be available at all times and a service failure may have catastrophic consequences for society or economy. Current CI protection strategies are from a time when CIs or CI sectors could be operated more or less self-sufficient and interconnections among CIs or CI sectors, which may lead to cascading service failures to other CIs or CI sectors, where not as omnipresent as today. In this PhD thesis, a cross-sector CI model for on-line risk monitoring of CI services, called CI security model, is presented. The model allows to monitor a CI service risk and to notify services that depend on it of possible risks in order to reduce and mitigate possible cascading failures. The model estimates CI service risk by observing the CI service state as measured by base measurements (e.g. sensor or software states) within the CI service components and by observing the experienced service risk of CI services it depends on (CI service dependencies). CI service risk is estimated in a probabilistic way using a Bayesian network based approach. Furthermore, the model allows CI service risk prediction in the short-term, mid-term and long-term future, given a current CI service risk and it allows to model interdependencies (a CI service risk that loops back to the originating service via dependencies), a special case that is difficult to model using Bayesian networks. The representation of a CI as a CI security model requires analysis. In this PhD thesis, a CI analysis method based on the PROTOS-MATINE dependency analysis methodology is presented in order to analyse CIs and represent them as CI services, CI service dependencies and base measurements. Additional research presented in this PhD thesis is related to a study of assurance indicators able to perform an on-line evaluation of the correctness of risk estimates within a CI service, as well as for risk estimates received from dependencies. A tool that supports all steps of establishing a CI security model was implemented during this PhD research. The research on the CI security model and the assurance indicators was validated based on a case study and the initial results suggest its applicability to CI environments. / Tiivistelmä Tässä väitöskirjassa esitellään läpileikkausmalli kriittisten infrastruktuurien jatkuvaan käytön riskimallinnukseen. Tämän mallin avulla voidaan tiedottaa toisistaan riippuvaisia palveluita mahdollisista vaaroista, ja siten pysäyttää tai hidastaa toisiinsa vaikuttavat ja kumuloituvat vikaantumiset. Malli analysoi kriittisen infrastruktuurin palveluriskiä tutkimalla kriittisen infrastruktuuripalvelun tilan, joka on mitattu perusmittauksella (esimerkiksi anturi- tai ohjelmistotiloina) kriittisen infrastruktuurin palvelukomponenttien välillä ja tarkkailemalla koetun kriittisen infrastruktuurin palveluriskiä, joista palvelut riippuvat (kriittisen infrastruktuurin palveluriippuvuudet). Kriittisen infrastruktuurin palveluriski arvioidaan todennäköisyyden avulla käyttämällä Bayes-verkkoja. Lisäksi malli mahdollistaa tulevien riskien ennustamisen lyhyellä, keskipitkällä ja pitkällä aikavälillä, ja mahdollistaa niiden keskinäisten riippuvuuksien mallintamisen, joka on yleensä vaikea esittää Bayes-verkoissa. Kriittisen infrastruktuurin esittäminen kriittisen infrastruktuurin tietoturvamallina edellyttää analyysiä. Tässä väitöskirjassa esitellään kriittisen infrastruktuurin analyysimenetelmä, joka perustuu PROTOS-MATINE -riippuvuusanalyysimetodologiaan. Kriittiset infrastruktuurit esitetään kriittisen infrastruktuurin palveluina, palvelujen keskinäisinä riippuvuuksina ja perusmittauksina. Lisäksi tutkitaan varmuusindikaattoreita, joilla voidaan tutkia suoraan toiminnassa olevan kriittisen infrastruktuuripalvelun riskianalyysin oikeellisuutta, kuin myös riskiarvioita riippuvuuksista. Tutkimuksessa laadittiin työkalu, joka tukee kriittisen infrastruktuurin tietoturvamallin toteuttamisen kaikkia vaiheita. Kriittisen infrastruktuurin tietoturvamalli ja varmuusindikaattorien oikeellisuus vahvistettiin konseptitutkimuksella, ja alustavat tulokset osoittavat menetelmän toimivuuden. / Kurzfassung In dieser Doktorarbeit wird ein Sektorübergreifendes Modell für die kontinuierliche Risikoabschätzung von kritische Infrastrukturen im laufenden Betrieb vorgestellt. Das Modell erlaubt es, Dienstleistungen, die in Abhängigkeit einer anderen Dienstleistung stehen, über mögliche Gefahren zu informieren und damit die Gefahr des Übergriffs von Risiken in andere Teile zu stoppen oder zu minimieren. Mit dem Modell können Gefahren in einer Dienstleistung anhand der Überwachung von kontinuierlichen Messungen (zum Beispiel Sensoren oder Softwarestatus) sowie der Überwachung von Gefahren in Dienstleistungen, die eine Abhängigkeit darstellen, analysiert werden. Die Abschätzung von Gefahren erfolgt probabilistisch mittels eines Bayessches Netzwerks. Zusätzlich erlaubt dieses Modell die Voraussage von zukünftigen Risiken in der kurzfristigen, mittelfristigen und langfristigen Zukunft und es erlaubt die Modellierung von gegenseitigen Abhängigkeiten, die im Allgemeinen schwer mit Bayesschen Netzwerken darzustellen sind. Um eine kritische Infrastruktur als ein solches Modell darzustellen, muss eine Analyse der kritischen Infrastruktur durchgeführt werden. In dieser Doktorarbeit wird diese Analyse durch die PROTOS-MATINE Methode zur Analyse von Abhängigkeiten unterstützt. Zusätzlich zu dem vorgestellten Modell wird in dieser Doktorarbeit eine Studie über Indikatoren, die das Vertrauen in die Genauigkeit einer Risikoabschätzung evaluieren können, vorgestellt. Die Studie beschäftigt sich sowohl mit der Evaluierung von Risikoabschätzungen innerhalb von Dienstleistungen als auch mit der Evaluierung von Risikoabschätzungen, die von Dienstleistungen erhalten wurden, die eine Abhängigkeiten darstellen. Eine Software, die alle Aspekte der Erstellung des vorgestellten Modells unterstützt, wurde entwickelt. Sowohl das präsentierte Modell zur Abschätzung von Risiken in kritischen Infrastrukturen als auch die Indikatoren zur Uberprüfung der Risikoabschätzungen wurden anhand einer Machbarkeitsstudie validiert. Erste Ergebnisse suggerieren die Anwendbarkeit dieser Konzepte auf kritische Infrastrukturen.