The current proposed unmanned aircraft system (UAS) detect and avoid standards require the same safety metrics, even when in close proximity to the ground or structures. This requirement has the potential to hinder low altitude small unmanned aircraft operations, such as local package delivery and utility inspection. One of the main safety metrics for UASs to adhere to is a ``well clear" volume that quantifies the vertical and horizontal separation UASs are required to maintain from manned aircraft. The current volume of 2000 feet horizontal and +/- 250 feet vertical does not provide credit for the safety benefit of being close to an obstacle where manned aircraft do not fly and could prove to be too restricting for low-level flight operations (i.e., under 400 feet above ground level). This thesis suggests using smaller safety metric volumes than the well clear volume to demonstrate that operations at lower altitudes can still be proven to be just as safe as if they were held to the larger well clear volume standard by using obstacle and terrain shielding. The research leverages simulation to analyze different safety metrics and provides an example use case in which the methodology of shielded operations is applied to demonstrate how this methodology can be applied for a safety case. / Master of Science / With the development of small unmanned aircraft system (sUAS) technologies have come many practical and regulatory challenges, especially in low altitude airspaces. At lower altitudes, manned aircraft are likely to be operating at lower velocities and restricting standards require UASs to maneuver against aircraft that may not present a significant risk of collision. The excessive avoidance maneuvering can cause the successful execution of even simple operations such as package delivery or survey operations to become difficult. The strict requirements have the potential to specifically inhibit sUAS beyond visual line-of-sight commercial operations, which are of great interest to the industry. This thesis describes a method for demonstrating an equivalent level of safety of small UAS operations when utilizing avoidance algorithms that leverage obstacle and terrain awareness. The purpose of this research is to demonstrate that by remaining close to obstacles, which pose a hazard to other aircraft, an unmanned aircraft can lower the risk of a mid-air collision and to demonstrate an equivalent level of safety for operations using a reduced safety metrics.

Objective: Previous population-based studies provided conflicting results regarding the association of total serum insulin-like growth factor I (IGF-I) and mortality. The aim of the present study was to assess the relation of IGF-I levels with all-cause mortality in a prospective study. Design: DETECT (Diabetes Cardiovascular Risk-Evaluation: Targets and Essential Data for Commitment of Treatment) is a large, multistage, and nationally representative study of primary care patients in Germany. The study population included 2463 men and 3603 women. Death rates were recorded by the respective primary care physician. Serum total IGF-I levels were determined by chemiluminescence immunoassays and categorized into three groups (low, moderate, and high) according to the sex- and age-specific 10th and 90th percentiles. Results: Adjusted analyses revealed that men with low [hazard ratio (HR) 1.70 (95% confidence interval [CI] 1.05–2.73), p=0.03] and high [HR 1.76 (95% CI 1.09–2.85), p=0.02] IGF-I levels had higher risk of all-cause mortality compared to men with moderate IGF-I levels. The specificity of low IGF-I and high IGF-I levels increased with lower and higher cut-offs, respectively. No such association became apparent in women. Conclusions: The present study revealed a U-shaped relation between IGF-I and all-cause mortality in male primary care patients.

Background: There is a lack of the generic data comparing the influence of different diseases on health-related quality of life (HrQoL) in a representative sample of primary care patients. Methods: Patient data were collected in the DETECT (Diabetes Cardiovascular Risk Evaluation: Targets and Essential Data for Commitment of Treatment) study including 55,000 patients. Results: 3,109 patients (33.3% female) with myocardial infarction (MI), stroke or both were compared to patients with a wide range of other diagnoses. Stroke and MI patients revealed a lower HrQoL as compared to patients with other diagnoses. Stroke was associated with strongest quality of life reduction. Multivariate analysis revealed several different determining factors. Conclusions: The reduction of HrQoL of patients with MI and stroke is primarily determined by the CNS insult. These data provide further evidence that early diagnosis and treatment of cardiovascular risk factors is essential to reduce subsequent stroke.

Background: There is a lack of the generic data comparing the influence of different diseases on health-related quality of life (HrQoL) in a representative sample of primary care patients. Methods: Patient data were collected in the DETECT (Diabetes Cardiovascular Risk Evaluation: Targets and Essential Data for Commitment of Treatment) study including 55,000 patients. Results: 3,109 patients (33.3% female) with myocardial infarction (MI), stroke or both were compared to patients with a wide range of other diagnoses. Stroke and MI patients revealed a lower HrQoL as compared to patients with other diagnoses. Stroke was associated with strongest quality of life reduction. Multivariate analysis revealed several different determining factors. Conclusions: The reduction of HrQoL of patients with MI and stroke is primarily determined by the CNS insult. These data provide further evidence that early diagnosis and treatment of cardiovascular risk factors is essential to reduce subsequent stroke.

Master of Science / Department of Electrical and Computer Engineering / Balasubramaniam Natarajan / In this thesis, we present an automated system for detection and classification of cracks, based on the new standard proposed by `American Association of State Highway and Transportation Officials (AASHTO)'. The AASHTO standard is a draft standard, that attempts to overcome the limitations of current crack quantifying and classification methods. In the current standard, the crack classification relies heavily on the judgment of the expert. Thus the results are susceptible to human error. The effect of human error is especially severe when the amount of data collected is large. This lead to inconsistencies even if a single standard is being followed. The new AASHTO guidelines attempt to develop a method for consistent measurement of pavement condition. Gray scale images of the road are captured by an image capture vehicle and stored on a database. Through steps of thresholding, line detect and scanning, the gray scale image is converted to binary image, with 'zeros' representing cracked pixels. PCA analysis, followed by closing and filtering operation, are carried out on the gray scale image to identify cracked sub-images. The output from the filtering operation, is then replaced with its binary counterpart. In the final step the crack parameters are calculated. The region around the crack is divided into blocks of 32x32 to approximate and calculate the crack parameters with ease. The width of the crack is approximated by the average width of crack in each block. The orientation of the crack is calculated from the angle between direction of travel and the line joining the ends of the crack. Length of the crack is the displacement between the ends of the crack, and the position of the crack is calculated from the midpoint of the line joining the end points.

Power system fault detection is conventionally achieved using current and potential measurements. An alternate and unconventional form of protective relaying is feasible using rigid bus conductor motion as the means of detection. The research presented focuses on the detection of power system faults using visual displacement of conductor spans. Substation rigid bus conductor motion is modeled using dual spring-mass systems for accurate representation of conductor response to electromagnetic forces generated during system faults. Bundled rigid conductors have advantages including detection independent of system load currents and improved ability to detect polyphase and single phase faults. The dynamic motion of the conductors during the fault is optically monitored with a laser detection system. Timeovercurrent characteristics are derived for the application of fault detection. The response time of the conductor detector system is slower than conventional relays due to the natural frequencies of the conductor span limiting the speed of its displacement. This response time makes the fault detection system using conductor displacement an ideal candidate for a backup relay in power system protection schemes.

Cybersecurity threats have surged in the past decades. Experts agree that conventional security measures will soon not be enough to stop the propagation of more sophisticated and harmful cyberattacks. Recently, there has been a growing interest in mastering the complexity of cybersecurity by adopting methods borrowed from Artificial Intelligence (AI) in order to support automation. Moreover, entire security frameworks, such as DETECT (Decision Triggering Event Composer and Tracker), are designed aimed to the automatic and early detection of threats against systems, by using model analysis and recognising sequences of events and other tropes, inherent to attack patterns. In this project, I concentrate on cybersecurity threat assessment by the translation of Attack Trees (AT) into probabilistic detection models based on Bayesian Networks (BN). I also show how these models can be integrated and dynamically updated as a detection engine in the existing DETECT framework for automated threat detection, hence enabling both offline and online threat assessment. Integration in DETECT is important to allow real-time model execution and evaluation for quantitative threat assessment. Finally, I apply my methodology to some real-world case studies, evaluate models with sample data, perform data sensitivity analyses, then present and discuss the results.

Os objetivos deste estudo foram desenvolver e testar in vivo e in vitro um método de detecção de lesões de cárie utilizando um espectrômetro portátil (EP); analisar o desempenho do EP e do equipamento comercial DIAGNOdent (Dd); correlacioná-los entre si, assim como com o padrão ouro, a área de secção e a profundidade das lesões. Os 66 sítios oclusais de pré-molares foram analisados in vivo com o Dd e, em seguida, a fluorescência induzida por um laser diodo (lexc ~ 657nm) foi coletada por fibra óptica, conduzida ao EP e analisada sob a forma de espectros, os quais foram normalizados e para os quais foi calculada a razão da área sob a curva dos espectros cariado e sadio (RACE). Os experimentos foram conduzidos in vitro nos mesmos sítios. O padrão ouro foi obtido por meio de microscopia de luz polarizada. Utilizou-se o índice de correlação linear de Pearson para comparar o Dd e o EP entre si e com a profundidade e área das lesões. A área sob a curva ROC, a sensibilidade, a especificidade e a acurácia foram calculadas e comparadas com o teste de McNemar. O Dd e a RACE apresentaram correlação significante com o padrão ouro (p < 0,01 para Dd e p < 0,05 para RACE) e entre si (r = 0,83 in vivo e r = 0,87 in vitro). Os equipamentos apresentaram baixa correlação, porém significante, com a profundidade das lesões tanto in vivo quanto in vitro (r = ~ 0,43). A área de secção transversal da lesão não influenciou nas medidas obtidas com o Dd e o EP. O Dd apresentou maior sensibilidade (0,76) do que o EP (0,60) in vivo (p < 0,05), embora isto não tenha aumentado o seu desempenho. In vitro, o EP apresentou maior sensibilidade (0,88) do que o Dd (0,79), mas essa diferença não foi significante. Os outros parâmetros também não apresentaram diferenças estatisticamente significantes (p < 0,05). O EP apresentou correlação positiva com o Dd, igual correlação com a profundidade da lesão e maior capacidade para detectar o tecido cariado in vitro, em relação ao Dd, o que sugere que com ponta convergente e angulada e software dedicado, o método será promissor para utilização em clínicas odontológicas em um futuro próximo. / The aims of this study were to develop and test a method to detect caries lesions in vivo and in vitro, using a portable spectrometer (PS); to analyze the performance of PS as well as the commercial device DIAGNOdent (Dd); correlate them with the gold standard, their tranversal section areas and lesions depth and between themselves. 66 occlusal pre-molars sites were examined in vivo with Dd. Sequentially, fluorescence (lexc ~ 657nm) was collected by an optical fiber, conducted to PS and then analyzed as spectra, which were normalized and had calculated the Ratios of their Areas Under the Curves (RAUC) of carious and sound tissues. Experiments were conducted in vitro in the same sites. Gold Standard was obtained by polarized light microscopy. Pearson correlation was used to compare the devices with transversal section area, lesions depth and between themselves. The area under ROC curve, sensitivity, specificity as well as accuracy were calculated and verified with McNemar test. Dd and RAUC showed statistically significant correlation with gold standard (p < 0.01 for Dd and p < 0.05 for RAUC) and between themselves (r = 0,83 in vivo and r = 0,87 in vitro). Although it was significant, the devices showed low correlation with depth of lesions in vivo and in vitro (r = ~ 0.43). The transversal section area of the lesion had no influence on readings in both devices. Dd showed higher sensitivity (0.76) than PS (0.60) in vivo (p < 0.05), though this fact was not able to improve its performance. In turn, PS showed higher sensitivity (0.88) than Dd (0.79) in vitro, but this difference was not significantly. The other parameters did not show statistically significant differences (p < 0.05) between methods. PS showed positive correlation with Dd, equal correlation with lesions depth and higher ability of detecting the disease in vitro, what suggests that if accompanied with a conic and an angulated probe and a dedicated software, the PS method could be useful in clinics in the near future.

Os objetivos deste estudo foram desenvolver e testar in vivo e in vitro um método de detecção de lesões de cárie utilizando um espectrômetro portátil (EP); analisar o desempenho do EP e do equipamento comercial DIAGNOdent (Dd); correlacioná-los entre si, assim como com o padrão ouro, a área de secção e a profundidade das lesões. Os 66 sítios oclusais de pré-molares foram analisados in vivo com o Dd e, em seguida, a fluorescência induzida por um laser diodo (lexc ~ 657nm) foi coletada por fibra óptica, conduzida ao EP e analisada sob a forma de espectros, os quais foram normalizados e para os quais foi calculada a razão da área sob a curva dos espectros cariado e sadio (RACE). Os experimentos foram conduzidos in vitro nos mesmos sítios. O padrão ouro foi obtido por meio de microscopia de luz polarizada. Utilizou-se o índice de correlação linear de Pearson para comparar o Dd e o EP entre si e com a profundidade e área das lesões. A área sob a curva ROC, a sensibilidade, a especificidade e a acurácia foram calculadas e comparadas com o teste de McNemar. O Dd e a RACE apresentaram correlação significante com o padrão ouro (p < 0,01 para Dd e p < 0,05 para RACE) e entre si (r = 0,83 in vivo e r = 0,87 in vitro). Os equipamentos apresentaram baixa correlação, porém significante, com a profundidade das lesões tanto in vivo quanto in vitro (r = ~ 0,43). A área de secção transversal da lesão não influenciou nas medidas obtidas com o Dd e o EP. O Dd apresentou maior sensibilidade (0,76) do que o EP (0,60) in vivo (p < 0,05), embora isto não tenha aumentado o seu desempenho. In vitro, o EP apresentou maior sensibilidade (0,88) do que o Dd (0,79), mas essa diferença não foi significante. Os outros parâmetros também não apresentaram diferenças estatisticamente significantes (p < 0,05). O EP apresentou correlação positiva com o Dd, igual correlação com a profundidade da lesão e maior capacidade para detectar o tecido cariado in vitro, em relação ao Dd, o que sugere que com ponta convergente e angulada e software dedicado, o método será promissor para utilização em clínicas odontológicas em um futuro próximo. / The aims of this study were to develop and test a method to detect caries lesions in vivo and in vitro, using a portable spectrometer (PS); to analyze the performance of PS as well as the commercial device DIAGNOdent (Dd); correlate them with the gold standard, their tranversal section areas and lesions depth and between themselves. 66 occlusal pre-molars sites were examined in vivo with Dd. Sequentially, fluorescence (lexc ~ 657nm) was collected by an optical fiber, conducted to PS and then analyzed as spectra, which were normalized and had calculated the Ratios of their Areas Under the Curves (RAUC) of carious and sound tissues. Experiments were conducted in vitro in the same sites. Gold Standard was obtained by polarized light microscopy. Pearson correlation was used to compare the devices with transversal section area, lesions depth and between themselves. The area under ROC curve, sensitivity, specificity as well as accuracy were calculated and verified with McNemar test. Dd and RAUC showed statistically significant correlation with gold standard (p < 0.01 for Dd and p < 0.05 for RAUC) and between themselves (r = 0,83 in vivo and r = 0,87 in vitro). Although it was significant, the devices showed low correlation with depth of lesions in vivo and in vitro (r = ~ 0.43). The transversal section area of the lesion had no influence on readings in both devices. Dd showed higher sensitivity (0.76) than PS (0.60) in vivo (p < 0.05), though this fact was not able to improve its performance. In turn, PS showed higher sensitivity (0.88) than Dd (0.79) in vitro, but this difference was not significantly. The other parameters did not show statistically significant differences (p < 0.05) between methods. PS showed positive correlation with Dd, equal correlation with lesions depth and higher ability of detecting the disease in vitro, what suggests that if accompanied with a conic and an angulated probe and a dedicated software, the PS method could be useful in clinics in the near future.

Detection and tracking of signals used in sonar applications in noisy environment is the focus of this thesis. We have concentrated on the low Signal-to-Noise Ratio (SNR) case where the conventional detection methods are not applicable. Furthermore, it is assumed that the duty cycle is relatively low. In the problem that is of concern the carrier frequency, pulse repetition interval (PRI) and the existence of the signal are not known. The unknown character of PRI makes the problem challenging since it means that the signal exists at some unknown intervals. A recursive, Bayesian track-before-detect (TBD) filter using particle filter based methods is proposed to solve the concerned problem. The data used by the particle filter is the magnitude of a complex spectrum in complex Gaussian noise. The existence variable is added in the design of the filter to determine the existence of the signal. The evolution of the signal state is modeled by a linear stochastic process. The filter estimates the signal state including the carrier frequency and PRI. Simulations are done under different scenarios where the carrier frequency, PRI and the existence of the signal varies. The results demonstrate that the algorithm presented in this thesis can detect signals which cannot be detected by conventional methods. Besides detection, the tracking performance of the filter is satisfying.