Automatic Lameness Detection in a Milking Robot : Instrumentation, measurement software, algorithms for data analysis and a neural network model

The aim of this thesis is to develop a fully automatic lameness detection system that operates in a milking robot. The instrumentation, measurement software, algorithms for data analysis and a neural network model for lameness detection were developed. Automatic milking has become a common practice in dairy husbandry, and in the year 2006 about 4000 farms worldwide used over 6000 milking robots. There is a worldwide movement with the objective of fully automating every process from feeding to milking. Increase in automation is a consequence of increasing farm sizes, the demand for more efficient production and the growth of labour costs. As the level of automation increases, the time that the cattle keeper uses for monitoring animals often decreases. This has created a need for systems for automatically monitoring the health of farm animals. The popularity of milking robots also offers a new and unique possibility to monitor animals in a single confined space up to four times daily. Lameness is a crucial welfare issue in the modern dairy industry. Limb disorders cause serious welfare, health and economic problems especially in loose housing of cattle. Lameness causes losses in milk production and leads to early culling of animals. These costs could be reduced with early identification and treatment. At present, only a few methods for automatically detecting lameness have been developed, and the most common methods used for lameness detection and assessment are various visual locomotion scoring systems. The problem with locomotion scoring is that it needs experience to be conducted properly, it is labour intensive as an on-farm method and the results are subjective. A four balance system for measuring the leg load distribution of dairy cows during milking in order to detect lameness was developed and set up in the University of Helsinki Research farm Suitia. The leg weights of 73 cows were successfully recorded during almost 10,000 robotic milkings over a period of 5 months. The cows were locomotion scored weekly, and the lame cows were inspected clinically for hoof lesions. Unsuccessful measurements, caused by cows standing outside the balances, were removed from the data with a special algorithm, and the mean leg loads and the number of kicks during milking was calculated. In order to develop an expert system to automatically detect lameness cases, a model was needed. A probabilistic neural network (PNN) classifier model was chosen for the task. The data was divided in two parts and 5,074 measurements from 37 cows were used to train the model. The operation of the model was evaluated for its ability to detect lameness in the validating dataset, which had 4,868 measurements from 36 cows. The model was able to classify 96% of the measurements correctly as sound or lame cows, and 100% of the lameness cases in the validation data were identified. The number of measurements causing false alarms was 1.1%. The developed model has the potential to be used for on-farm decision support and can be used in a real-time lameness monitoring system.

Statistical and Information-Theoretic Methods for Data-Analysis

In this Thesis, we develop theory and methods for computational data analysis. The problems in data analysis are approached from three perspectives: statistical learning theory, the Bayesian framework, and the information-theoretic minimum description length (MDL) principle. Contributions in statistical learning theory address the possibility of generalization to unseen cases, and regression analysis with partially observed data with an application to mobile device positioning. In the second part of the Thesis, we discuss so called Bayesian network classifiers, and show that they are closely related to logistic regression models. In the final part, we apply the MDL principle to tracing the history of old manuscripts, and to noise reduction in digital signals.

Geant4 Hadronic Cascade Models and CMS Data Analysis : Computational Challenges in the LHC era

This work belongs to the field of computational high-energy physics (HEP). The key methods used in this thesis work to meet the challenges raised by the Large Hadron Collider (LHC) era experiments are object-orientation with software engineering, Monte Carlo simulation, the computer technology of clusters, and artificial neural networks. The first aspect discussed is the development of hadronic cascade models, used for the accurate simulation of medium-energy hadron-nucleus reactions, up to 10 GeV. These models are typically needed in hadronic calorimeter studies and in the estimation of radiation backgrounds. Various applications outside HEP include the medical field (such as hadron treatment simulations), space science (satellite shielding), and nuclear physics (spallation studies). Validation results are presented for several significant improvements released in Geant4 simulation tool, and the significance of the new models for computing in the Large Hadron Collider era is estimated. In particular, we estimate the ability of the Bertini cascade to simulate Compact Muon Solenoid (CMS) hadron calorimeter HCAL. LHC test beam activity has a tightly coupled cycle of simulation-to-data analysis. Typically, a Geant4 computer experiment is used to understand test beam measurements. Thus an another aspect of this thesis is a description of studies related to developing new CMS H2 test beam data analysis tools and performing data analysis on the basis of CMS Monte Carlo events. These events have been simulated in detail using Geant4 physics models, full CMS detector description, and event reconstruction. Using the ROOT data analysis framework we have developed an offline ANN-based approach to tag b-jets associated with heavy neutral Higgs particles, and we show that this kind of NN methodology can be successfully used to separate the Higgs signal from the background in the CMS experiment.

Accelerator Mass Spectrometry and Bayesian Data Analysis

Accelerator mass spectrometry (AMS) is an ultrasensitive technique for measuring the concentration of a single isotope. The electric and magnetic fields of an electrostatic accelerator system are used to filter out other isotopes from the ion beam. The high velocity means that molecules can be destroyed and removed from the measurement background. As a result, concentrations down to one atom in 10^16 atoms are measurable. This thesis describes the construction of the new AMS system in the Accelerator Laboratory of the University of Helsinki. The system is described in detail along with the relevant ion optics. System performance and some of the 14C measurements done with the system are described. In a second part of the thesis, a novel statistical model for the analysis of AMS data is presented. Bayesian methods are used in order to make the best use of the available information. In the new model, instrumental drift is modelled with a continuous first-order autoregressive process. This enables rigorous normalization to standards measured at different times. The Poisson statistical nature of a 14C measurement is also taken into account properly, so that uncertainty estimates are much more stable. It is shown that, overall, the new model improves both the accuracy and the precision of AMS measurements. In particular, the results can be improved for samples with very low 14C concentrations or measured only a few times.

Residual noise covariance for Planck low-resolution data analysis

Aims: Develop and validate tools to estimate residual noise covariance in Planck frequency maps. Quantify signal error effects and compare different techniques to produce low-resolution maps. Methods: We derive analytical estimates of covariance of the residual noise contained in low-resolution maps produced using a number of map-making approaches. We test these analytical predictions using Monte Carlo simulations and their impact on angular power spectrum estimation. We use simulations to quantify the level of signal errors incurred in different resolution downgrading schemes considered in this work. Results: We find an excellent agreement between the optimal residual noise covariance matrices and Monte Carlo noise maps. For destriping map-makers, the extent of agreement is dictated by the knee frequency of the correlated noise component and the chosen baseline offset length. The significance of signal striping is shown to be insignificant when properly dealt with. In map resolution downgrading, we find that a carefully selected window function is required to reduce aliasing to the sub-percent level at multipoles, ell > 2Nside, where Nside is the HEALPix resolution parameter. We show that sufficient characterization of the residual noise is unavoidable if one is to draw reliable contraints on large scale anisotropy. Conclusions: We have described how to compute the low-resolution maps, with a controlled sky signal level, and a reliable estimate of covariance of the residual noise. We have also presented a method to smooth the residual noise covariance matrices to describe the noise correlations in smoothed, bandwidth limited maps.

How to deal with sparse macroseismic data

This study discusses the scope of historical earthquake analysis in low-seismicity regions. Examples of non-damaging earthquake reports are given from the Eastern Baltic (Fennoscandian) Shield in north-eastern Europe from the 16th to the 19th centuries. The information available for past earthquakes in the region is typically sparse and cannot be increased through a careful search of the archives. This study applies recommended rigorous methodologies of historical seismology developed using ample data to the sparse reports from the Eastern Baltic Shield. Attention is paid to the context of reporting, the identity and role of the authors, the circumstances of the reporting, and the opportunity to verify the available information by collating the sources. We evaluate the reliability of oral earthquake recollections and develop criteria for cases when a historical earthquake is attested to by a single source. We propose parametric earthquake scenarios as a way to deal with sparse macroseismic reports and as an improvement to existing databases.

Interaction in the Multicultural Classroom : Towards Culturally Sensitive Home Economics Education

Continuous growth in the number of immigrant students has changed the Finnish school environment. The resulting multicultural school environment is new for both teachers and students. In order to develop multicultural learning environments, there is a need to understand immigrant students everyday lives in school. In this study, home economics is seen as a fruitful school subject area for understanding these immigrant students lives as they cope with school and home cultures that may be very different from each other. Home economics includes a great deal of knowledge and skills that immigrant students need during their everyday activities outside of school. -- The main aim of the study is to clarify the characteristics of multicultural home economics classroom practices and the multicultural contacts and interaction that take place between the students and the teacher. The study includes four parts. The first part, an ethnographical prestudy, aims to understand the challenges of multicultural schoolwork with the aid of ethnographical fieldwork done in one multicultural school. The second part outlines the theoretical frames of the study and focuses on the sociocultural approach. The third part of the study presents an analysis of videodata collected in a multicultural home economics classroom. The teacher s and students interaction in the home economics classroom is analyzed through the concepts of the sociocultural approach and the cultural-historical activity theory. Firstly, this is done by analyzing the focusedness of the teacher s and the students actions as well as the questions presented and apparent disturbances during classroom interaction. Secondly, the immigrant students everyday experiences and cultural background are examined as they appear during discussions in the home economics lessons. Thirdly, the teacher s tool-use and actions as a human mediator are clarified during interaction in the classroom. The fourth part presents the results, according to which a practice-based approach in the multicultural classroom situation is a prerequisite for the teacher s and the students shared object during classroom interaction. Also, the practice-based approach facilitates students understanding during teaching and learning situations. Practice in this study is understood as collaborative teaching and learning situations that include 1) guided activating learning, 2) establishing connections with students everyday lives and 3) multiple tool-use. Guided activating learning in the classroom is defined as situations that occur and assignments that are done with a knowledgeable adult or peer and include action. The teacher s demonstrations during the practical part of the lessons seemed to be fruitful in the teaching and learning situations in the multicultural classroom. Establishing connections with students everyday lives motivated students to follow the lesson and supported understanding of meaning. Furthermore, if multiple tools (both psychological and material) were used, the students managed better with new and sometimes difficult concepts and different working habits, and accomplished the practical work more smoothly . The teacher s tool-use and role as a mediator of meaning are also highlighted in the data analysis. Hopefully, this study can provide a seedbed for situations in which knowledge produced together, as well as horizontally oriented tool-use, can make school-learned knowledge more relevant to immigrant students everyday lives, and help students to better cope with both classroom work and outside activities. KEY WORDS: home economics education, multicultural education, sociocultural perspective, classroom interaction, videoanalysis

Mining Cell Transition Data

Cell transition data is obtained from a cellular phone that switches its current serving cell tower. The data consists of a sequence of transition events, which are pairs of cell identifiers and transition times. The focus of this thesis is applying data mining methods to such data, developing new algorithms, and extracting knowledge that will be a solid foundation on which to build location-aware applications. In addition to a thorough exploration of the features of the data, the tools and methods developed in this thesis provide solutions to three distinct research problems. First, we develop clustering algorithms that produce a reliable mapping between cell transitions and physical locations observed by users of mobile devices. The main clustering algorithm operates in online fashion, and we consider also a number of offline clustering methods for comparison. Second, we define the concept of significant locations, known as bases, and give an online algorithm for determining them. Finally, we consider the task of predicting the movement of the user, based on historical data. We develop a prediction algorithm that considers paths of movement in their entirety, instead of just the most recent movement history. All of the presented methods are evaluated with a significant body of real cell transition data, collected from about one hundred different individuals. The algorithms developed in this thesis are designed to be implemented on a mobile device, and require no extra hardware sensors or network infrastructure. By not relying on external services and keeping the user information as much as possible on the user s own personal device, we avoid privacy issues and let the users control the disclosure of their location information.

Elämän tiede : Tutkimus lääketieteellisestä teknologiasta, vanhemmuudesta ja perimän hallinnasta

Doctoral dissertation work in sociology examines how human heredity became a scientific, political and a personal issue in the 20th century Finland. The study focuses on the institutionalisation of rationales and technologies concerning heredity, in the context of Finnish medicine and health care. The analysis concentrates specifically on the introduction and development of prenatal screening within maternity care. The data comprises of medical articles, policy documents and committee reports, as well as popular guidebooks and health magazines. The study commences with an analysis on the early 20th century discussions on racial hygiene. It ends with an analysis on the choices given to pregnant mothers and families at present. Freedom to choose, considered by geneticists and many others as a guarantee of the ethicality of medical applications, is presented in this study as a historically, politically and scientifically constructed issue. New medical testing methods have generated new possibilities of governing life itself. However, they have also created new ethical problems. Leaning on recent historical data, the study illustrates how medical risk rationales on heredity have been asserted by the medical profession into Finnish health care. It also depicts medical professions ambivalence between maintaining the patients autonomy and utilizing for example prenatal testing according to health policy interests. Personalized risk is discussed as a result of the empirical analysis. It is indicated that increasing risk awareness amongst the public, as well as offering choices, have had unintended consequences. According to doctors, present day parents often want to control risks more than what is considered justified or acceptable. People s hopes to anticipate the health and normality of their future children have exceeded the limits offered by medicine. Individualization of the government of heredity is closely linked to a process that is termed as depolitization. The concept refers to disembedding of medical genetics from its social contexts. Prenatal screening is regarded to be based on individual choice facilitated by neutral medical knowledge. However, prenatal screening within maternity care also has its basis in health policy aims and economical calculations. Methodological basis of the study lies in Michel Foucault s writings on the history of thought, as well as in science and technology studies.

"Elämää yhdellä siivellä lentäen" : Suomessa asuvien Chilen pakolaisten ja heidän jälkeläistensä historiatietoisuus

The starting point of this study was to find out how the historical consciousness manifest in conceptions and experiences of Chilean refugees and their descendants. The previous research of historical consciousness has shown that powerful experiences such as the revolution and being a refugee may have an effect on historical consciousness. The purpose of this study is to solve how those experiences in the past have influenced Chilean refugees and their descendant s interpretations of the present and expectations for the future. The research material was collected by interviewing four Chilean refugees that escaped to Finland in years 1973 1976 and four young adults who represent the second generation. All second generation interviewees were born in Finland and their other parent or both parents were Chilean refugees. The two groups were not in a family relation to each other. The empirical part of the research was made by qualitative methods. The research material was collected by the method of focused interview and it was analysed by the qualitative data analysis software Atlas.ti 6.0. Content analysis was the main research tool. The previous theory of historical consciousness and the study questions was used to create the seven categories that manifest historical consciousness. The seven categories were biographical memory, collective memory, experiences of living between two cultures, idea of man, the essence of history and the reason for living, value conceptions and expectations of the future. Content analysis was based on those categories. Subcategories were based on the research material and were created during the analysis. The results of this study were made up of categories. The study revealed that experiences of revolution and of being a refugee has a significant role in the historical consciousness of the Chilean refugees. It became evident in their biographical memory being separated in three parts, in their values and in the belief of possibility of an individual to govern her own life. The second generation was also exposed to their parent s experiences in the past. The collective trauma in their parent s past has been part of their life indirectly and has affected the way they think of themselves, their concepts and their place in the present world. The active and regular retrospection in Finland by Chilean adults and special Gabriela Mistral club activities has played a big part in the construction of their historical consciousness.

Data fusion and matching by maximizing statistical dependencies

The core aim of machine learning is to make a computer program learn from the experience. Learning from data is usually defined as a task of learning regularities or patterns in data in order to extract useful information, or to learn the underlying concept. An important sub-field of machine learning is called multi-view learning where the task is to learn from multiple data sets or views describing the same underlying concept. A typical example of such scenario would be to study a biological concept using several biological measurements like gene expression, protein expression and metabolic profiles, or to classify web pages based on their content and the contents of their hyperlinks. In this thesis, novel problem formulations and methods for multi-view learning are presented. The contributions include a linear data fusion approach during exploratory data analysis, a new measure to evaluate different kinds of representations for textual data, and an extension of multi-view learning for novel scenarios where the correspondence of samples in the different views or data sets is not known in advance. In order to infer the one-to-one correspondence of samples between two views, a novel concept of multi-view matching is proposed. The matching algorithm is completely data-driven and is demonstrated in several applications such as matching of metabolites between humans and mice, and matching of sentences between documents in two languages.

Comparing type counts

This work is a case study of applying nonparametric statistical methods to corpus data. We show how to use ideas from permutation testing to answer linguistic questions related to morphological productivity and type richness. In particular, we study the use of the suffixes -ity and -ness in the 17th-century part of the Corpus of Early English Correspondence within the framework of historical sociolinguistics. Our hypothesis is that the productivity of -ity, as measured by type counts, is significantly low in letters written by women. To test such hypotheses, and to facilitate exploratory data analysis, we take the approach of computing accumulation curves for types and hapax legomena. We have developed an open source computer program which uses Monte Carlo sampling to compute the upper and lower bounds of these curves for one or more levels of statistical significance. By comparing the type accumulation from women’s letters with the bounds, we are able to confirm our hypothesis.

Juoksuhaudoista yhteiskuntaan : Vuosien 1965-1975 sotilaallista maanpuolustusta ja puolustusvoimia koskevan joukkotiedotuksen diskursiivinen tarkastelu

The Finnish society developed rapidly in the 1960´s and 1970´s. This was result of international trends. Development of education, urbanization and wide organization of society increased discontent towards prevailing social structure and towards the power elite. Development of technology created possibility to present radical perspectives in mass media. This caused widely spread discussions dividing opinions. The purpose of this thesis was to complement research on national defence and the Finnish Defence Forces especially between years 1965 and 1975. The task of research was to clarify how changes in society and how the significance of this change was interpreted in public discussion about national defence and development of the Defence Forces. The most essential points for this thesis turned out to be discourses structured from public discussion. Main research material consisted of approximately 35000 news, editorials, articles and opinions presented in mass media supplemented by literature, committee reports and other archival sources. Frame of reference for this thesis is based on relativistic worldview. According to this, social reality is relative and there is no single truth. Environment has significant influence on the issue how knowledge and truth are formed. Data analysis was based on critical discourse. The key objective was to clarify the effects of broad changes in society using discursive methods. One essential goal was to form order of discourse using linguistic analysis and also connect discourses to wider sociocultural custom. On this thesis I came to the conclusion that on the review period there were five significant ensembles of discourse. They consisted of several discussions focused on different themes. The discourse of official security policy aimed to define national defence and the position of the Defence Forces as parts of foreign policy. Foreign policy is often perceived as the most significant part of security policy. Historical memory, geographical position of Finland and also the state contracts, changes in international warfare, tasks of the Defence Forces and increasing critic of national defence and the difference in thinking between generations formed the discourse of security policy. In the discourse of the liability to military service, the issue was about individual responsibility to society and national defence. Resisters and unarmed defence demands, encouraged by international examples were the themes. The discourse pointed out how mass media is used to influence and forced the Defence Forces to develop the practices in public information. The discourses of democracy and politics were closer to internal development of the Defence Forces to integrate more into society. The discourse of democracy focused in changing power relationships of the Defence Forces that were known as authoritarian. Issues like conscript and personnel union activity had lot of similarities to general social development. The discourse of politics presented how the Defence Forces were pushed towards parliamentary decision making. The personnel was granted the same rights as other population. Themes related to the discourse on the will to national defence were development of mental national defence, increasing education on national defence and creation of more open public information culture. According to discourses presented above I can state, that the position of the Defence Forces in society was changed between years 1965-1975. This change was advanced by the Defence Forces reformed attitude towards mass media and public information in general. Active participation in public information important became important instead of only answering topics. This positive development created an atmosphere, that was easier for the public to understand and create own pictures of the armed forces. Due to this, I can describe that the defenders and supporters of the armed forces were stuck in their trenches, until discussions presented in discourses and themes developed the Defence Forces to be better fitting part of society. Key words; society, national defence, Defence Forces, discourse, mass media, security policy, liability to military service, conscription, democracy

Endemic malaria: an 'indoor' disease in northern Europe

Background: Endemic northern malaria reached 68°N latitude in Europe during the 19th century, where the summer mean temperature only irregularly exceeded 16°C, the lower limit needed for sporogony of Plasmodium vivax. Because of the available historical material and little use of quinine, Finland was suitable for an analysis of endemic malaria and temperature. Methods: Annual malaria death frequencies during 1800–1870 extracted from parish records were analysed against long-term temperature records in Finland, Russia and Sweden. Supporting data from 1750–1799 were used in the interpretation of the results. The life cycle and behaviour of the anopheline mosquitoes were interpreted according to the literature. Results: Malaria frequencies correlated strongly with the mean temperature of June and July of the preceding summer, corresponding to larval development of the vector. Hatching of imagoes peaks in the middle of August, when the temperature most years is too low for the sporogony of Plasmodium. After mating some of the females hibernate in human dwellings. If the female gets gametocytes from infective humans, the development of Plasmodium can only continue indoors, in heated buildings. Conclusion: Northern malaria existed in a cold climate by means of summer dormancy of hypnozoites in humans and indoor transmission of sporozoites throughout the winter by semiactive hibernating mosquitoes. Variable climatic conditions did not affect this relationship. The epidemics, however, were regulated by the population size of the mosquitoes which, in turn, ultimately was controlled by the temperatures of the preceding summer.