Matching Items (133)
Description
In many classication problems data samples cannot be collected easily, example in drug trials, biological experiments and study on cancer patients. In many situations the data set size is small and there are many outliers. When classifying such data, example cancer vs normal patients the consequences of mis-classication are probably more important than any other data type, because the data point could be a cancer patient or the classication decision could help determine what gene might be over expressed and perhaps a cause of cancer. These mis-classications are typically higher in the presence of outlier data points. The aim of this thesis is to develop a maximum margin classier that is suited to address the lack of robustness of discriminant based classiers (like the Support Vector Machine (SVM)) to noise and outliers. The underlying notion is to adopt and develop a natural loss function that is more robust to outliers and more representative of the true loss function of the data. It is demonstrated experimentally that SVM's are indeed susceptible to outliers and that the new classier developed, here coined as Robust-SVM (RSVM), is superior to all studied classier on the synthetic datasets. It is superior to the SVM in both the synthetic and experimental data from biomedical studies and is competent to a classier derived on similar lines when real life data examples are considered.
ContributorsGupta, Sidharth (Author) / Kim, Seungchan (Thesis advisor) / Welfert, Bruno (Committee member) / Li, Baoxin (Committee member) / Arizona State University (Publisher)
Created2011
Description
Detecting anatomical structures, such as the carina, the pulmonary trunk and the aortic arch, is an important step in designing a CAD system of detection Pulmonary Embolism. The presented CAD system gets rid of the high-level prior defined knowledge to become a system which can easily extend to detect other anatomic structures. The system is based on a machine learning algorithm --- AdaBoost and a general feature --- Haar. This study emphasizes on off-line and on-line AdaBoost learning. And in on-line AdaBoost, the thesis further deals with extremely imbalanced condition. The thesis first reviews several knowledge-based detection methods, which are relied on human being's understanding of the relationship between anatomic structures. Then the thesis introduces a classic off-line AdaBoost learning. The thesis applies different cascading scheme, namely multi-exit cascading scheme. The comparison between the two methods will be provided and discussed. Both of the off-line AdaBoost methods have problems in memory usage and time consuming. Off-line AdaBoost methods need to store all the training samples and the dataset need to be set before training. The dataset cannot be enlarged dynamically. Different training dataset requires retraining the whole process. The retraining is very time consuming and even not realistic. To deal with the shortcomings of off-line learning, the study exploited on-line AdaBoost learning approach. The thesis proposed a novel pool based on-line method with Kalman filters and histogram to better represent the distribution of the samples' weight. Analysis of the performance, the stability and the computational complexity will be provided in the thesis. Furthermore, the original on-line AdaBoost performs badly in imbalanced conditions, which occur frequently in medical image processing. In image dataset, positive samples are limited and negative samples are countless. A novel Self-Adaptive Asymmetric On-line Boosting method is presented. The method utilized a new asymmetric loss criterion with self-adaptability according to the ratio of exposed positive and negative samples and it has an advanced rule to update sample's importance weight taking account of both classification result and sample's label. Compared to traditional on-line AdaBoost Learning method, the new method can achieve far more accuracy in imbalanced conditions.
ContributorsWu, Hong (Author) / Liang, Jianming (Thesis advisor) / Farin, Gerald (Committee member) / Ye, Jieping (Committee member) / Arizona State University (Publisher)
Created2011
Description
Building mathematical models and examining the compatibility of their theoretical predictions with empirical data are important for our understanding of evolution. The rapidly increasing amounts of genomic data on polymorphisms greatly motivate evolutionary biologists to find targets of positive selection. Although intensive mathematical and statistical studies for characterizing signatures of positive selection have been conducted to identify targets of positive selection, relatively little is known about the effects of other evolutionary forces on signatures of positive selection. In this dissertation, I investigate the effects of various evolutionary factors, including purifying selection and population demography, on signatures of positive selection. Specifically, the effects on two highly used methods for detecting positive selection, one by Wright's Fst and its analogues and the other by footprints of genetic hitchhiking, are investigated. In Chapters 2 and 3, the effect of purifying selection on Fst is studied. The results show that purifying selection intensity greatly affects Fst by modulating allele frequencies across populations. The footprints of genetic hitchhiking in a geographically structured population are studied in Chapter 4. The results demonstrate that footprints of genetic hitchhiking are significantly influenced by geographic structure, which may help scientists to infer the origin and spread of the beneficial allele. In Chapter 5, the stochastic dynamics of a hitchhiking allele are studied using the diffusion process of genetic hitchhiking conditioned on the fixation of the beneficial allele. Explicit formulae for the conditioned two-locus diffusion process of genetic hitchhiking are derived and stochastic aspects of genetic hitchhiking are investigated. The results in this dissertation show that it is essential to model the interaction of neutral and selective forces for correct identification of the targets of positive selection.
ContributorsMaruki, Takahiro (Author) / Kim, Yuseob (Thesis advisor) / Taylor, Jesse E (Thesis advisor) / Greenwood, Priscilla E (Committee member) / Hedrick, Philip W (Committee member) / Rosenberg, Michael S. (Committee member) / Arizona State University (Publisher)
Created2011
Description
Genomic and proteomic sequences, which are in the form of deoxyribonucleic acid (DNA) and amino acids respectively, play a vital role in the structure, function and diversity of every living cell. As a result, various genomic and proteomic sequence processing methods have been proposed from diverse disciplines, including biology, chemistry, physics, computer science and electrical engineering. In particular, signal processing techniques were applied to the problems of sequence querying and alignment, that compare and classify regions of similarity in the sequences based on their composition. However, although current approaches obtain results that can be attributed to key biological properties, they require pre-processing and lack robustness to sequence repetitions. In addition, these approaches do not provide much support for efficiently querying sub-sequences, a process that is essential for tracking localized database matches. In this work, a query-based alignment method for biological sequences that maps sequences to time-domain waveforms before processing the waveforms for alignment in the time-frequency plane is first proposed. The mapping uses waveforms, such as time-domain Gaussian functions, with unique sequence representations in the time-frequency plane. The proposed alignment method employs a robust querying algorithm that utilizes a time-frequency signal expansion whose basis function is matched to the basic waveform in the mapped sequences. The resulting WAVEQuery approach is demonstrated for both DNA and protein sequences using the matching pursuit decomposition as the signal basis expansion. The alignment localization of WAVEQuery is specifically evaluated over repetitive database segments, and operable in real-time without pre-processing. It is demonstrated that WAVEQuery significantly outperforms the biological sequence alignment method BLAST for queries with repetitive segments for DNA sequences. A generalized version of the WAVEQuery approach with the metaplectic transform is also described for protein sequence structure prediction. For protein alignment, it is often necessary to not only compare the one-dimensional (1-D) primary sequence structure but also the secondary and tertiary three-dimensional (3-D) space structures. This is done after considering the conformations in the 3-D space due to the degrees of freedom of these structures. As a result, a novel directionality based 3-D waveform mapping for the 3-D protein structures is also proposed and it is used to compare protein structures using a matched filter approach. By incorporating a 3-D time axis, a highly-localized Gaussian-windowed chirp waveform is defined, and the amino acid information is mapped to the chirp parameters that are then directly used to obtain directionality in the 3-D space. This mapping is unique in that additional characteristic protein information such as hydrophobicity, that relates the sequence with the structure, can be added as another representation parameter. The additional parameter helps tracking similarities over local segments of the structure, this enabling classification of distantly related proteins which have partial structural similarities. This approach is successfully tested for pairwise alignments over full length structures, alignments over multiple structures to form a phylogenetic trees, and also alignments over local segments. Also, basic classification over protein structural classes using directional descriptors for the protein structure is performed.
ContributorsRavichandran, Lakshminarayan (Author) / Papandreou-Suppappola, Antonia (Thesis advisor) / Spanias, Andreas S (Thesis advisor) / Chakrabarti, Chaitali (Committee member) / Tepedelenlioğlu, Cihan (Committee member) / Lacroix, Zoé (Committee member) / Arizona State University (Publisher)
Created2011
Description
We propose a novel solution to prevent cancer by developing a prophylactic cancer. Several sources of antigens for cancer vaccines have been published. Among these, antigens that contain a frame-shift (FS) peptide or viral peptide are quite attractive for a variety of reasons. FS sequences, from either mistake in RNA processing or in genomic DNA, may lead to generation of neo-peptides that are foreign to the immune system. Viral peptides presumably would originate from exogenous but integrated viral nucleic acid sequences. Both are non-self, therefore lessen concerns about development of autoimmunity. I have developed a bioinformatical approach to identify these aberrant transcripts in the cancer transcriptome. Their suitability for use in a vaccine is evaluated by establishing their frequencies and predicting possible epitopes along with their population coverage according to the prevalence of major histocompatibility complex (MHC) types. Viral transcripts and transcripts with FS mutations from gene fusion, insertion/deletion at coding microsatellite DNA, and alternative splicing were identified in NCBI Expressed Sequence Tag (EST) database. 48 FS chimeric transcripts were validated in 50 breast cell lines and 68 primary breast tumor samples with their frequencies from 4% to 98% by RT-PCR and sequencing confirmation. These 48 FS peptides, if translated and presented, could be used to protect more than 90% of the population in Northern America based on the prediction of epitopes derived from them. Furthermore, we synthesized 150 peptides that correspond to FS and viral peptides that we predicted would exist in tumor patients and we tested over 200 different cancer patient sera. We found a number of serological reactive peptide sequences in cancer patients that had little to no reactivity in healthy controls; strong support for the strength of our bioinformatic approach. This study describes a process used to identify aberrant transcripts that lead to a new source of antigens that can be tested and used in a prophylactic cancer vaccine. The vast amount of transcriptome data of various cancers from the Cancer Genome Atlas (TCGA) project will enhance our ability to further select better cancer antigen candidates.
ContributorsLee, HoJoon (Author) / Johnston, Stephen A. (Thesis advisor) / Kumar, Sudhir (Committee member) / Miller, Laurence (Committee member) / Stafford, Phillip (Committee member) / Sykes, Kathryn (Committee member) / Arizona State University (Publisher)
Created2012
Description
Major Depressive Disorder (MDD) and Posttraumatic Stress Disorder (PTSD) are highly prevalent illnesses that can result in profound impairment. While many patients with these disorders present in primary care, research suggests that physicians under-detect and suboptimally manage MDD and PTSD in their patients. The development of more effective training interventions to aid primary care providers in diagnosing mental health disorders is of the utmost importance. This research focuses on evaluating computer-based training tools (Avatars) for training family physicians to better diagnose MDD and PTSD. Three interventions are compared: a "choice" avatar simulation training program, a "fixed" avatar simulation training program, and a text-based training program for training physicians to improve their diagnostic interviewing skills in detecting and diagnosing MDD and PTSD. Two one-way ANCOVAs were used to analyze the differences between the groups on diagnostic accuracy while controlling for mental health experience. In order to assess specifically how prior mental health experience affected diagnostic accuracy the covariate of prior mental health experience was then used as an independent variable and simple main effects and pairwise comparisons were evaluated. Results indicated that for the MDD case both avatar treatment groups significantly outperformed the text-based treatment in diagnostic accuracy regardless of prior mental health experience. For the PTSD case those receiving the fixed avatar simulation training more accurately diagnosed PTSD than the text-based training group and the choice-avatar training group regardless of prior mental health experience. Confidence ratings indicated that the majority of participants were very confident with their diagnoses for both cases. Discussion focused on the utility of avatar technology in medical education. The findings in this study indicate that avatar technology aided the participants in diagnosing MDD and PTSD better than traditional text-based methods employed to train PCPs to diagnose. Regardless of experience level the fixed avatar group outperformed the other groups for both cases. Avatar technology used in diagnostic training can be user-friendly and cost-effective. It can also have a world-wide reach. Additional educational benefit could be provided by using automated text analysis to provide physicians with feedback based on the extent to which their case diagnostic summaries cover relevant content. In conclusion, avatar technology can offer robust training that could be potentially transferred to real environment performance.
ContributorsSatter, Rachel (Author) / Kinnier, Richard (Thesis advisor) / Mackenzie, James (Committee member) / Claiborn, Charles (Committee member) / Arizona State University (Publisher)
Created2012
Description
The living world we inhabit and observe is extraordinarily complex. From the perspective of a person analyzing data about the living world, complexity is most commonly encountered in two forms: 1) in the sheer size of the datasets that must be analyzed and the physical number of mathematical computations necessary to obtain an answer and 2) in the underlying structure of the data, which does not conform to classical normal theory statistical assumptions and includes clustering and unobserved latent constructs. Until recently, the methods and tools necessary to effectively address the complexity of biomedical data were not ordinarily available. The utility of four methods--High Performance Computing, Monte Carlo Simulations, Multi-Level Modeling and Structural Equation Modeling--designed to help make sense of complex biomedical data are presented here.
ContributorsBrown, Justin Reed (Author) / Dinu, Valentin (Thesis advisor) / Johnson, William (Committee member) / Petitti, Diana (Committee member) / Arizona State University (Publisher)
Created2012
Description
Threshold logic has been studied by at least two independent group of researchers. One group of researchers studied threshold logic with the intention of building threshold logic circuits. The earliest research to this end was done in the 1960's. The major work at that time focused on studying mathematical properties of threshold logic as no efficient circuit implementations of threshold logic were available. Recently many post-CMOS (Complimentary Metal Oxide Semiconductor) technologies that implement threshold logic have been proposed along with efficient CMOS implementations. This has renewed the effort to develop efficient threshold logic design automation techniques. This work contributes to this ongoing effort. Another group studying threshold logic did so, because the building block of neural networks - the Perceptron, is identical to the threshold element implementing a threshold function. Neural networks are used for various purposes as data classifiers. This work contributes tangentially to this field by proposing new methods and techniques to study and analyze functions implemented by a Perceptron After completion of the Human Genome Project, it has become evident that most biological phenomenon is not caused by the action of single genes, but due to the complex interaction involving a system of genes. In recent times, the `systems approach' for the study of gene systems is gaining popularity. Many different theories from mathematics and computer science has been used for this purpose. Among the systems approaches, the Boolean logic gene model has emerged as the current most popular discrete gene model. This work proposes a new gene model based on threshold logic functions (which are a subset of Boolean logic functions). The biological relevance and utility of this model is argued illustrated by using it to model different in-vivo as well as in-silico gene systems.
ContributorsLinge Gowda, Tejaswi (Author) / Vrudhula, Sarma (Thesis advisor) / Shrivastava, Aviral (Committee member) / Chatha, Karamvir (Committee member) / Kim, Seungchan (Committee member) / Arizona State University (Publisher)
Created2012
Description
Immunosignaturing is a technology that allows the humoral immune response to be observed through the binding of antibodies to random sequence peptides. The immunosignaturing microarray is based on complex mixtures of antibodies binding to arrays of random sequence peptides in a multiplexed fashion. There are computational and statistical challenges to the analysis of immunosignaturing data. The overall aim of my dissertation is to develop novel computational and statistical methods for immunosignaturing data to access its potential for diagnostics and drug discovery. Firstly, I discovered that a classification algorithm Naive Bayes which leverages the biological independence of the probes on our array in such a way as to gather more information outperforms other classification algorithms due to speed and accuracy. Secondly, using this classifier, I then tested the specificity and sensitivity of immunosignaturing platform for its ability to resolve four different diseases (pancreatic cancer, pancreatitis, type 2 diabetes and panIN) that target the same organ (pancreas). These diseases were separated with >90% specificity from controls and from each other. Thirdly, I observed that the immunosignature of type 2 diabetes and cardiovascular complications are unique, consistent, and reproducible and can be separated by 100% accuracy from controls. But when these two complications arise in the same person, the resultant immunosignature is quite different in that of individuals with only one disease. I developed a method to trace back from informative random peptides in disease signatures to the potential antigen(s). Hence, I built a decipher system to trace random peptides in type 1 diabetes immunosignature to known antigens. Immunosignaturing, unlike the ELISA, has the ability to not only detect the presence of response but also absence of response during a disease. I observed, not only higher but also lower peptides intensities can be mapped to antigens in type 1 diabetes. To study immunosignaturing potential for population diagnostics, I studied effect of age, gender and geographical location on immunosignaturing data. For its potential to be a health monitoring technology, I proposed a single metric Coefficient of Variation that has shown potential to change significantly when a person enters a disease state.
ContributorsKukreja, Muskan (Author) / Johnston, Stephen Albert (Thesis advisor) / Stafford, Phillip (Committee member) / Dinu, Valentin (Committee member) / Arizona State University (Publisher)
Created2012
Description
Rapid advance in sensor and information technology has resulted in both spatially and temporally data-rich environment, which creates a pressing need for us to develop novel statistical methods and the associated computational tools to extract intelligent knowledge and informative patterns from these massive datasets. The statistical challenges for addressing these massive datasets lay in their complex structures, such as high-dimensionality, hierarchy, multi-modality, heterogeneity and data uncertainty. Besides the statistical challenges, the associated computational approaches are also considered essential in achieving efficiency, effectiveness, as well as the numerical stability in practice. On the other hand, some recent developments in statistics and machine learning, such as sparse learning, transfer learning, and some traditional methodologies which still hold potential, such as multi-level models, all shed lights on addressing these complex datasets in a statistically powerful and computationally efficient way. In this dissertation, we identify four kinds of general complex datasets, including "high-dimensional datasets", "hierarchically-structured datasets", "multimodality datasets" and "data uncertainties", which are ubiquitous in many domains, such as biology, medicine, neuroscience, health care delivery, manufacturing, etc. We depict the development of novel statistical models to analyze complex datasets which fall under these four categories, and we show how these models can be applied to some real-world applications, such as Alzheimer's disease research, nursing care process, and manufacturing.
ContributorsHuang, Shuai (Author) / Li, Jing (Thesis advisor) / Askin, Ronald (Committee member) / Ye, Jieping (Committee member) / Runger, George C. (Committee member) / Arizona State University (Publisher)
Created2012