京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第21623号 / 医博第4429号 / 新制||医||1033(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 川口 義弥, 教授 坂井 義治, 教授 伊達 洋至 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

I monitored cause-specific mortality and factors influencing mortality risk for white-tailed deer in the Upper Peninsula of Michigan, USA, during two high mortality risk periods: adult female deer during Feb–May, and fawns from birth to 6 months. I observed high rates of predation and starvation for adult female deer during Apr–May, suggesting that late winter represents a survival bottleneck due to nutritional declines. A strong negative relationship existed between snow free days during late winter and mortality risk. Predation was the dominant mortality source for fawns but predation risk decreased with larger birth mass. Black bears and coyotes accounted for most fawn kills at the population level, but wolves and bobcats had greatest per-individual fawn kill rates. My results suggest predation was the dominant mortality source for fawns and adult female deer, but multiple predator species were important and nutritional condition of deer influenced their vulnerability to predation.

Chemotherapy is a method used to treat cancer but it has a number of side-effects. Research conducted by the Department of Chemical Engineering at BYU involves a new method of administering chemotherapy using ultrasound waves and water-soluble capsules. The goal is to reduce the side-effects by localizing the delivery of the medication. As part of this research, a two-factor experiment was conducted on rats to test if the water-soluble capsules and ultrasound waves by themselves have an effect on tumor growth or patient survival. Our project emphasizes the usage of Bayesian Hierarchical Models and Win-BUGS to jointly model the survival data and the longitudinal data—mass. The results of the joint analysis indicate that the use of ultrasound and water-soluble microcapsules have no negative effect on survival. In fact, there appears to be a positive effect on the survival since the rats in the ultrasound-capsule group had higher survival rates than the rats in other treatment groups. From these results, it does appear that the new technology involving ultrasound waves and microcapsules is a promising way to reduce the side-effects of chemotherapy. It is strongly advocated that the formulation of a joint model for any longitudinal and survival data be performed. For future work for the ultrasound-microcapsule data it is recommended that joint modeling of the mass, tumor volume, and survival data be conducted to obtain additional information.

The goal of this research is to expand the field of conformal predictions using Random Survival Forests. The standard Conformal Random Survival Forest can predict with a fixed certainty whether something will survive up until a certain time point. This research is the first to show that there is little practical use in the standard Conformal Random Survival Forest algorithm. It turns out that the confidence guarantees of the conformal prediction framework are violated if the Standard algorithm makes predictions for a user-controlled fixed time point. To solve this challenge, this thesis proposes two algorithms that specialize in conformal predictions for a fixed point in time: a Fixed Time algorithm and a Hybrid algorithm. Both algorithms transform the survival data that is used by the split evaluation metric in the Random Survival Forest algorithm. The algorithms are evaluated and compared along six different set prediction evaluation criteria. The prediction performance of the Hybrid algorithm outperforms the prediction performance of the Fixed Time algorithm in most cases. Furthermore, the Hybrid algorithm is more stable than the Fixed Time algorithm when the predicting job extends to various time points. The hybrid Conformal Random Survival Forest should thus be considered by anyone who wants to make conformal survival predictions at usercontrolled time points. / Målet med denna avhandling är att utöka området för konformitetsprediktion med hjälp av Random Survival Forests. Standardutförandet av Conformal Random Survival Forest kan förutsäga med en viss säkerhet om någonting kommer att överleva fram till en viss tidpunkt. Denna avhandling är den första som visar att det finns liten praktisk användning i standardutförandet av Conformal Random Survival Forest-algoritmen. Det visar sig att konfidensgarantierna för konformitetsprediktionsramverket bryts om standardalgoritmen gör förutsägelser för en användarstyrd fast tidpunkt. För att lösa denna utmaning, föreslår denna avhandling två algoritmer som specialiserar sig i konformitetsprediktion för en bestämd tidpunkt: en fast-tids algoritm och en hybridalgoritm. Båda algoritmerna omvandlar den överlevnadsdata som används av den delade utvärderingsmetoden i Random Survival Forest-algoritmen. Uppskattningsförmågan för hybridalgoritmen överträffar den för fast-tids algoritmen i de flesta fall. Dessutom är hybrid algoritmen stabilare än fast-tids algoritmen när det förutsägelsejobbet sträcker sig till olika tidpunkter. Hybridalgoritmen för Conformal Random Survival Forest bör därför föredras av den som vill göra konformitetsprediktion av överlevnad vid användarstyrda tidpunkter.

Survival analysis is an essential statistics and machine learning field in various critical applications like medical research and predictive maintenance. In these domains understanding models' predictions is paramount. While machine learning techniques are increasingly applied to enhance the predictive performance of survival models, they simultaneously sacrifice transparency and explainability.  Survival models, in contrast to regular machine learning models, predict functions rather than point estimates like regression and classification models. This creates a challenge regarding explaining such models using the known off-the-shelf machine learning explanation techniques, like Shapley Values, Counterfactual examples, and others.    Censoring is also a major issue in survival analysis where the target time variable is not fully observed for all subjects. Moreover, in predictive maintenance settings, recorded events do not always map to actual failures, where some components could be replaced because it is considered faulty or about to fail in the future based on an expert's opinion. Censoring and noisy labels create problems in terms of modeling and evaluation that require to be addressed during the development and evaluation of the survival models. Considering the challenges in survival modeling and the differences from regular machine learning models, this thesis aims to bridge this gap by facilitating the use of machine learning explanation methods to produce plausible and actionable explanations for survival models. It also aims to enhance survival modeling and evaluation revealing a better insight into the differences among the compared survival models. In this thesis, we propose two methods for explaining survival models which rely on discovering survival patterns in the model's predictions that group the studied subjects into significantly different survival groups. Each pattern reflects a specific survival behavior common to all the subjects in their respective group. We utilize these patterns to explain the predictions of the studied model in two ways. In the first, we employ a classification proxy model that can capture the relationship between the descriptive features of subjects and the learned survival patterns. Explaining such a proxy model using Shapley Values provides insights into the feature attribution of belonging to a specific survival pattern. In the second method, we addressed the "what if?" question by generating plausible and actionable counterfactual examples that would change the predicted pattern of the studied subject. Such counterfactual examples provide insights into actionable changes required to enhance the survivability of subjects. We also propose a variational-inference-based generative model for estimating the time-to-event distribution. The model relies on a regression-based loss function with the ability to handle censored cases. It also relies on sampling for estimating the conditional probability of event times. Moreover, we propose a decomposition of the C-index into a weighted harmonic average of two quantities, the concordance among the observed events and the concordance between observed and censored cases. These two quantities, weighted by a factor representing the balance between the two, can reveal differences between survival models previously unseen using only the total Concordance index. This can give insight into the performances of different models and their relation to the characteristics of the studied data. Finally, as part of enhancing survival modeling, we propose an algorithm that can correct erroneous event labels in predictive maintenance time-to-event data. we adopt an expectation-maximization-like approach utilizing a genetic algorithm to find better labels that would maximize the survival model's performance. Over iteration, the algorithm builds confidence about events' assignments which improves the search in the following iterations until convergence. We performed experiments on real and synthetic data showing that our proposed methods enhance the performance in survival modeling and can reveal the underlying factors contributing to the explainability of survival models' behavior and performance.

Vår studie undersöker hur huvudkaraktärerna i filmerna Adrift och Arctic gestaltas utifrån könsstereotyper och om de faller in i en genusordning. Med utgångspunkt i hur könsroller representeras i media är studiens syfte att undersöka hur en kvinnlig protagonist porträtteras jämfört med en manlig i en överlevnadsfilm där karaktärerna är utelämnade i svåra överlevnadssituationer. Hur skildras kvinnor och män i förhållande till överlevnadssituationer där de måste rädda sitt eget och någon annans liv? Genom en multimodal kritisk diskursanalys har vi undersökt filmerna Adrift och Arctic för att uppnå studiens syfte. Adrift är baserad på en sann händelse om hur en kvinna och en man kämpar för sin överlevnad till havs. Arctic är berättelsen om en man som är strandsatt på Arktis efter en flygkrasch. Resultaten visar på att huvudkaraktärerna i filmerna både följer och bryter med könsstereotyper och könsnormer. Utifrån situation och relation har vi sett hur karaktärerna avviker eller följer könsstereotyper. I båda filmerna upprätthålls en genusordning oavsett vilka egenskaper karaktärerna uppvisar. / Our study examines how the main characters in the films Adrift and Arctic are portrayed based on gender stereotypes and whether they conform to gender norms. Drawing on representations of gender roles in media, the study investigates how a female protagonist is portrayed compared to a male in survival films where characters are faced with challenging situations. How are women and men depicted in relation to survival situations where they must save their own and someone else’s life? Through a multimodal critical discourse analysis, we have examined the films Adrift and Arctic to achieve the study’s objective. Adrift is based on a true story of a woman and a man fighting for survival at sea. Arctic tells the story of a man stranded in the Arctic after a plane crash. The results indicate that the main characters in the films both conform to and break gender stereotypes and norms. Based on the situation and relationship, we have observed how the characters deviate from or adhere to gender stereotypes. In both films, a gender hierarchy is maintained regardless of the characters’ attributes.

No description available.

The abundance of Rio Grande wild turkeys (Meleagris gallopavo intermedia) in the southeastern Edwards Plateau of Texas has declined since the late 1970s. Because knowledge of reproductive rates is important to understanding the dynamics of a population, radio-tagged hens were monitored during the 2005–2007 reproductive seasons to evaluate and compare reproductive parameters from areas with both declining and stable population trends. During January–March of 2005–2007, turkey hens were captured and radiotagged on 4 study areas; 2 within a region of stable turkey populations, and 2 within a region of declining populations. Monitoring occurred from January–July each season to determine nest- site locations. Nesting attempts, nest fate, clutch size, initiation date, and nest age were recorded. Nests were monitored ≥3 times weekly in order to estimate production parameters and daily nest survival. Poults were captured by hand and fitted with a 1.2 glue-on transmitter and monitored daily to estimate daily survival. Estimates show production was greater in stable regions than declining regions of the Edwards Plateau. Eighty-four percent of hens attempted to nest in the stable region and 67% attempted in the declining region. Eighteen of 102 nests were successful (≥1 egg hatched), in the stable region and 7 of 60 nests were successful in the declining region. Nest-survival analysis showed an influence of temporal variation within years, yet no differences in nest survival were detected between stable and declining regions. Poult survival also showed no difference between regions. The 2 overall objectives of this study were to determine if nesting parameters and nest survival were limiting factors in Rio Grande wild turkey abundance in the Edwards Plateau. Regional differences in production suggest the cause of the decline in the southeastern portion of the Edwards Plateau could be associated with lower reproductive output and consequently, success. Regional differences in nest survival were not detected, thus not likely to cause differences in turkey abundance between regions.

Master of Science / Department of Biology / David A. Haukos / Abstract–Duckling Survival Lesser scaup (scaup) populations have been experiencing continent-wide decline since the 1980s. It is important to have complete understanding of the critical factors influencing population change (e.g., duckling survival, nesting success, and health) to advance our understanding of population dynamics and improving species conservation. Duckling survival is a primary driver of scaup demography. I conducted a capture-mark-recapture study using Cormack-Jolly-Seber models in Program MARK to compute apparent daily survival and recapture probabilities for 3256 individually marked ducklings with 620 recaptures during 2010 to 2013. The most parsimonious survival model based on a priori hypotheses found that Julian hatch date squared was the most significant predictor of survival and consistent through all four years. Mass at hatch also was significant as a quadratic effect. Duckling survival to 30 days ranged from 29.0 to 80.0. During this study, stabilizing selection played a significant role in duckling survival, which indicates that there was trade-offs for selection of an optimal timing of hatch on survival and a cost associated with hatching too early or too late and being too heavy or to light. / Abstract–Nest-Site Selection There is a hierarchical process of behavioral and environmental processes that influence habitat selection, which inherently influences the survival and fitness of that individual and contributes to population growth. I investigated nest fate, spatial attributes, and all relationships between high and low-water levels with habitat attributes (distance to upland, distance to open water, nearest neighbor distance) of located nests using general linear models in SAS, t-tests in R, and Hot Spot Analysis in ArcGIS of 481 nests over eight years. In low-water years, successful nests (X̅ = 1153 m) were located 22.0% farther from upland than unsuccessful nests (X̅ = 944 m), but support for a similar relationship was lacking in high-water years. Successful nests were located 21.0% and 23.0% (i.e., 49 and 50 m) closer to conspecific nests than unsuccessful nests in low and high-water levels, respectively. In both high and low-water level years, clusters of nests initiated later in the season coincided with Hot Spots for nest fate (i.e., high-quality habitat patches, clusters of successful nests), whereas areas that tended to be selected first, evidenced by clusters of nests initiated earlier, tended to overlap with clusters of Cold Spots for nest fate. The core Hot Spot for nest fate was in the same spot in both water level conditions and located in flooded emergent vegetation in the heart of Lower Red Rock Lake furthest from any upland habitat. Three out of six Cold Spots for nest fate both in high and low-water years were located in emergent vegetation on the perimeter of Lower Red Rock Lake adjacent to uplands. Density-dependence seems to be a factor affecting late-nesting scaup females that are apparently cuing in on the reproductive performance of conspecifics when determining where to nest. Therefore, management actions focused on survival and reproductive success of scaup should consider managing water levels and habitat for later nesting scaup to increase adult survival and ultimately recruitment of ducklings. / Abstract – Blood Parasite Prevalence Blood parasites, per se, do not lead to direct mortality, but instead reduce the health of individual birds, which may ultimately lead to decreased reproductive success. Evidence has shown that presence of blood parasites can reduce fitness, body condition, and reproductive success of waterfowl. For many avian species, the cost of reproduction is manifested as a negative relationship between female breeding effort and breeding season survival, with trade-offs occurring when these adaptive choices become detrimental to future reproductive performance. Blood was drawn for parasite load determination from 112 individual adult scaup captured from 2011to 2012 via spotlighting and drive-trapping. Parasite prevalence was determined through blood assays that were created using a two-slide wedge technique. Relationships among seasonal heterophile:lymphocyte ratio (a proxy for health), body mass at time of capture (throughout pre-, during, and post-breeding periods), breeding status (females only), and Julian date of capture (date of capture) of capture with parasite prevalence were analyzed using linear (lm) regression models in R 2.15.2. The blood parasite infection rate was 5.0%, with prevalence differing by gender with 33.3% of males positive for blood parasites compared to 1.0% of females. The presence of blood parasites did not affect health, fitness, or breeding status of scaup. A quadratic relationship was found with body mass and date of capture, indicating that body mass increased from pre-breeding period to the breeding period and decreased significantly at the end of the summer during molt. A negative relationship between the heterophile:lymphocyte ratio of female scaup and date of capture (i.e., the health of scaup females was greatest during the pre-breeding period after which it consistently decreased until the molting period). A strong negative correlation between heterophile:lymphocyte ratio and body mass was found in both genders, which indicated that scaup in poor body condition were also in poor health at the end of the breeding season.

Healthcare providers encounter a wide range of hazards on the job, including exposure to infectious diseases. Protecting them from occupational infectious disease is very important. Healthcare workers use personal protective equipment (PPE) as a measure to decrease the risk of getting infected during patient care. For high-risk diseases like Ebola, Tyvek suits are coverall suits that protect the body and reduce the risk of body fluid exposure. However, a person removing a contaminated suit may also be exposed to virus. Previous studies have shown that enveloped viruses can survive on different types of surfaces, so the objective of this study is to determine the inactivation of bacteriophage Φ6, a surrogate for enveloped human virus, on the surface of Tyvek suits at two different relative humidity levels, 40% and 60% at 22°C. The results showed the inactivation rate of virus was higher at 60% RH than 40% RH. There was ~3log10 (99.9%) reduction of virus inactivation after 6 hours at 40% but ~3log10 (99.9%) inactivation took 9 hours at 60%. This suggests that enveloped viruses can survive on the surface of Tyvek suits for more than 6 hours, and should be considered a potential risk for contamination when they are taken off after use.