Matching Items (12)
Filtering by
- All Subjects: Health Sciences
- All Subjects: Transfer Learning
- Creators: Buman, Matthew
- Creators: Li, Jing
Description
INTRODUCTION: Exercise performed at moderate to vigorous intensities has been shown to generate a post exercise hypotensive response. Whether this response is observed with very low exercise intensities is unclear. PURPOSE: To compare post physical activity ambulatory blood pressure (ABP) response to a single worksite walking day and a normal sedentary work day in pre-hypertensive adults. METHODS: Participants were 7 pre-hypertensive (127 + 8 mmHg / 83 + 8 mmHg) adults (3 male, 4 female, age = 42 + 12 yr) who participated in a randomized, cross-over study that included a control and a walking treatment. Only those who indicated regularly sitting at least 8 hours/day and no structured physical activity were enrolled. Treatment days were randomly assigned and were performed one week apart. Walking treatment consisted of periodically increasing walk time up to 2.5 hours over the course of an 8 hour work day on a walking workstation (Steelcase Company, Grand Rapids, MI). Walk speed was set at 1 mph. Participants wore an ambulatory blood pressure cuff (Oscar 2, SunTech Medical, Morrisville, NC) for 24-hours on both treatment days. Participants maintained normal daily activities on the control day. ABP data collected from 9:00 am until 10:00 pm of the same day were included in statistical analyses. Linear mixed models were used to detect differences in systolic (SBP) and diastolic blood pressure (DBP) by treatment condition over the whole day and post workday for the time periods between 4 -10 pm when participants were no longer at work. RESULTS:BP was significantly lower in response to the walking treatment compared to the control day (Mean SBP 126 +7 mmHg vs.124 +7 mmHg, p=.043; DBP 80 + 3 mmHg vs. 77 + 3 mmHg, p = 0.001 respectively). Post workday (4:00 to 10:00 pm) SBP decreased 3 mmHg (p=.017) and DBP decreased 4 mmHg (p<.001) following walking. CONCLUSION: Even low intensity exercise such as walking on a walking workstation is effective for significantly reducing acute BP when compared to a normal work day.
ContributorsZeigler, Zachary (Author) / Swan, Pamela (Thesis advisor) / Buman, Matthew (Committee member) / Gaesser, Glenn (Committee member) / Arizona State University (Publisher)
Created2013
Description
Cardiovascular disease (CVD) is the number one cause of death in the United States and type 2 diabetes (T2D) and obesity lead to cardiovascular disease. Obese adults are more susceptible to CVD compared to their non-obese counterparts. Exercise training leads to large reductions in the risk of CVD and T2D. Recent evidence suggests high-intensity interval training (HIT) may yield similar or superior benefits in a shorter amount of time compared to traditional continuous exercise training. The purpose of this study was to compare the effects of HIT to continuous (CONT) exercise training for the improvement of endothelial function, glucose control, and visceral adipose tissue. Seventeen obese men (N=9) and women (N=8) were randomized to eight weeks of either HIT (N=9, age=34 years, BMI=37.6 kg/m2) or CONT (N=8, age=34 years, BMI=34.6 kg/m2) exercise 3 days/week for 8 weeks. Endothelial function was assessed via flow-mediated dilation (FMD), glucose control was assessed via continuous glucose monitoring (CGM), and visceral adipose tissue and body composition was measured with an iDXA. Incremental exercise testing was performed at baseline, 4 weeks, and 8 weeks. There were no changes in weight, fat mass, or visceral adipose tissue measured by the iDXA, but there was a significant reduction in body fat that did not differ by group (46±6.3 to 45.4±6.6%, P=0.025). HIT led to a significantly greater improvement in FMD compared to CONT exercise (HIT: 5.1 to 9.0%; CONT: 5.0 to 2.6%, P=0.006). Average 24-hour glucose was not improved over the whole group and there were no group x time interactions for CGM data (HIT: 103.9 to 98.2 mg/dl; CONT: 99.9 to 100.2 mg/dl, P>0.05). When statistical analysis included only the subjects who started with an average glucose at baseline > 100 mg/dl, there was a significant improvement in glucose control overall, but no group x time interaction (107.8 to 94.2 mg/dl, P=0.027). Eight weeks of HIT led to superior improvements in endothelial function and similar improvements in glucose control in obese subjects at risk for T2D and CVD. HIT was shown to have comparable or superior health benefits in this obese sample with a 36% lower total exercise time commitment.
ContributorsSawyer, Brandon J (Author) / Gaesser, Glenn A (Thesis advisor) / Shaibi, Gabriel (Committee member) / Lee, Chong (Committee member) / Swan, Pamela (Committee member) / Buman, Matthew (Committee member) / Arizona State University (Publisher)
Created2013
Description
Discriminative learning when training and test data belong to different distributions is a challenging and complex task. Often times we have very few or no labeled data from the test or target distribution, but we may have plenty of labeled data from one or multiple related sources with different distributions. Due to its capability of migrating knowledge from related domains, transfer learning has shown to be effective for cross-domain learning problems. In this dissertation, I carry out research along this direction with a particular focus on designing efficient and effective algorithms for BioImaging and Bilingual applications. Specifically, I propose deep transfer learning algorithms which combine transfer learning and deep learning to improve image annotation performance. Firstly, I propose to generate the deep features for the Drosophila embryo images via pretrained deep models and build linear classifiers on top of the deep features. Secondly, I propose to fine-tune the pretrained model with a small amount of labeled images. The time complexity and performance of deep transfer learning methodologies are investigated. Promising results have demonstrated the knowledge transfer ability of proposed deep transfer algorithms. Moreover, I propose a novel Robust Principal Component Analysis (RPCA) approach to process the noisy images in advance. In addition, I also present a two-stage re-weighting framework for general domain adaptation problems. The distribution of source domain is mapped towards the target domain in the first stage, and an adaptive learning model is proposed in the second stage to incorporate label information from the target domain if it is available. Then the proposed model is applied to tackle cross lingual spam detection problem at LinkedIn’s website. Our experimental results on real data demonstrate the efficiency and effectiveness of the proposed algorithms.
ContributorsSun, Qian (Author) / Ye, Jieping (Committee member) / Xue, Guoliang (Committee member) / Liu, Huan (Committee member) / Li, Jing (Committee member) / Arizona State University (Publisher)
Created2015
Description
Transfer learning is a sub-field of statistical modeling and machine learning. It refers to methods that integrate the knowledge of other domains (called source domains) and the data of the target domain in a mathematically rigorous and intelligent way, to develop a better model for the target domain than a model using the data of the target domain alone. While transfer learning is a promising approach in various application domains, my dissertation research focuses on the particular application in health care, including telemonitoring of Parkinson’s Disease (PD) and radiomics for glioblastoma.
The first topic is a Mixed Effects Transfer Learning (METL) model that can flexibly incorporate mixed effects and a general-form covariance matrix to better account for similarity and heterogeneity across subjects. I further develop computationally efficient procedures to handle unknown parameters and large covariance structures. Domain relations, such as domain similarity and domain covariance structure, are automatically quantified in the estimation steps. I demonstrate METL in an application of smartphone-based telemonitoring of PD.
The second topic focuses on an MRI-based transfer learning algorithm for non-invasive surgical guidance of glioblastoma patients. Limited biopsy samples per patient create a challenge to build a patient-specific model for glioblastoma. A transfer learning framework helps to leverage other patient’s knowledge for building a better predictive model. When modeling a target patient, not every patient’s information is helpful. Deciding the subset of other patients from which to transfer information to the modeling of the target patient is an important task to build an accurate predictive model. I define the subset of “transferrable” patients as those who have a positive rCBV-cell density correlation, because a positive correlation is confirmed by imaging theory and the its respective literature.
The last topic is a Privacy-Preserving Positive Transfer Learning (P3TL) model. Although negative transfer has been recognized as an important issue by the transfer learning research community, there is a lack of theoretical studies in evaluating the risk of negative transfer for a transfer learning method and identifying what causes the negative transfer. My work addresses this issue. Driven by the theoretical insights, I extend Bayesian Parameter Transfer (BPT) to a new method, i.e., P3TL. The unique features of P3TL include intelligent selection of patients to transfer in order to avoid negative transfer and maintain patient privacy. These features make P3TL an excellent model for telemonitoring of PD using an At-Home Testing Device.
The first topic is a Mixed Effects Transfer Learning (METL) model that can flexibly incorporate mixed effects and a general-form covariance matrix to better account for similarity and heterogeneity across subjects. I further develop computationally efficient procedures to handle unknown parameters and large covariance structures. Domain relations, such as domain similarity and domain covariance structure, are automatically quantified in the estimation steps. I demonstrate METL in an application of smartphone-based telemonitoring of PD.
The second topic focuses on an MRI-based transfer learning algorithm for non-invasive surgical guidance of glioblastoma patients. Limited biopsy samples per patient create a challenge to build a patient-specific model for glioblastoma. A transfer learning framework helps to leverage other patient’s knowledge for building a better predictive model. When modeling a target patient, not every patient’s information is helpful. Deciding the subset of other patients from which to transfer information to the modeling of the target patient is an important task to build an accurate predictive model. I define the subset of “transferrable” patients as those who have a positive rCBV-cell density correlation, because a positive correlation is confirmed by imaging theory and the its respective literature.
The last topic is a Privacy-Preserving Positive Transfer Learning (P3TL) model. Although negative transfer has been recognized as an important issue by the transfer learning research community, there is a lack of theoretical studies in evaluating the risk of negative transfer for a transfer learning method and identifying what causes the negative transfer. My work addresses this issue. Driven by the theoretical insights, I extend Bayesian Parameter Transfer (BPT) to a new method, i.e., P3TL. The unique features of P3TL include intelligent selection of patients to transfer in order to avoid negative transfer and maintain patient privacy. These features make P3TL an excellent model for telemonitoring of PD using an At-Home Testing Device.
ContributorsYoon, Hyunsoo (Author) / Li, Jing (Thesis advisor) / Wu, Teresa (Committee member) / Yan, Hao (Committee member) / Hu, Leland S. (Committee member) / Arizona State University (Publisher)
Created2018
Description
Background and purpose: Regular physical activity (PA) provides benefits for cognitive health and helps to improve or maintain quality of life among older adults. Objective PA measures have been increasingly used to overcome limitations of self-report measures. The purpose of this study was to investigate the association of objectively measured PA and sedentary time with cognitive function among older adults.
Methods: Participants were recruited from the parent REasons for Geographic and Racial Differences in Stroke (REGARDS) Study. ActicalTM accelerometers provided estimates of PA variables, including moderate-to-vigorous PA (MVPA), high light PA (HLPA), low light PA (LLPA) and sedentary time, for 4-7 consecutive days. Prevalence and incidence of cognitive impairment were defined by the Six-Item Screener. Letter fluency, animal fluency, word list learning and Montreal Cognitive Assessment (orientation and recall) were conducted to assess executive function and memory.
Results: Of the 7,339 participants who provided accelerometer wear data > 4 days (70.1 ± 8.6 yr, 54.2% women, 31.7% African American), 320 participants exhibited impaired cognition. In cross-sectional analysis, participants in the highest MVPA% quartile had 39% lower odds of cognitive impairment than those in the lowest quartile (OR: 0.61, 95% C.I.: 0.39-0.95) after full adjustment. Further analysis shows most quartiles of MVPA% and HLPA% were significantly associated with executive function and memory (P<0.01). During 2.7 ± 0.5 years of follow-up, 3,385 participants were included in the longitudinal analysis, with 157 incident cases of cognitive impairment. After adjustments, participants in the highest MVPA% quartile had 51% lower hazards of cognitive impairment (HR: 0.49, 95% C.I.: 0.28-0.86). Additionally, MVPA% was inversely associated with change in memory z-scores (P<0.01), while the highest quartile of HLPA% was inversely associated with change in executive function and memory z-scores (P<0.01).
Conclusion: Higher levels of objectively measured MVPA% were independently associated with lower prevalence and incidence of cognitive impairment, and better memory and executive function in older adults. Higher levels of HLPA% were also independently associated with better memory and executive function. The amount of MVPA associated with lower risk of cognitive impairment (259 min/week) is >70% higher than the minimal amount of MVPA recommended by PA guidelines.
Methods: Participants were recruited from the parent REasons for Geographic and Racial Differences in Stroke (REGARDS) Study. ActicalTM accelerometers provided estimates of PA variables, including moderate-to-vigorous PA (MVPA), high light PA (HLPA), low light PA (LLPA) and sedentary time, for 4-7 consecutive days. Prevalence and incidence of cognitive impairment were defined by the Six-Item Screener. Letter fluency, animal fluency, word list learning and Montreal Cognitive Assessment (orientation and recall) were conducted to assess executive function and memory.
Results: Of the 7,339 participants who provided accelerometer wear data > 4 days (70.1 ± 8.6 yr, 54.2% women, 31.7% African American), 320 participants exhibited impaired cognition. In cross-sectional analysis, participants in the highest MVPA% quartile had 39% lower odds of cognitive impairment than those in the lowest quartile (OR: 0.61, 95% C.I.: 0.39-0.95) after full adjustment. Further analysis shows most quartiles of MVPA% and HLPA% were significantly associated with executive function and memory (P<0.01). During 2.7 ± 0.5 years of follow-up, 3,385 participants were included in the longitudinal analysis, with 157 incident cases of cognitive impairment. After adjustments, participants in the highest MVPA% quartile had 51% lower hazards of cognitive impairment (HR: 0.49, 95% C.I.: 0.28-0.86). Additionally, MVPA% was inversely associated with change in memory z-scores (P<0.01), while the highest quartile of HLPA% was inversely associated with change in executive function and memory z-scores (P<0.01).
Conclusion: Higher levels of objectively measured MVPA% were independently associated with lower prevalence and incidence of cognitive impairment, and better memory and executive function in older adults. Higher levels of HLPA% were also independently associated with better memory and executive function. The amount of MVPA associated with lower risk of cognitive impairment (259 min/week) is >70% higher than the minimal amount of MVPA recommended by PA guidelines.
ContributorsZhu, Wenfei (Author) / Hooker, Steven P (Thesis advisor) / Wadley, Virginia (Committee member) / Ainsworth, Barbara (Committee member) / Der Ananian, Cheryl (Committee member) / Buman, Matthew (Committee member) / Arizona State University (Publisher)
Created2015
Description
Transfer learning refers to statistical machine learning methods that integrate the knowledge of one domain (source domain) and the data of another domain (target domain) in an appropriate way, in order to develop a model for the target domain that is better than a model using the data of the target domain alone. Transfer learning emerged because classic machine learning, when used to model different domains, has to take on one of two mechanical approaches. That is, it will either assume the data distributions of the different domains to be the same and thereby developing one model that fits all, or develop one model for each domain independently. Transfer learning, on the other hand, aims to mitigate the limitations of the two approaches by accounting for both the similarity and specificity of related domains. The objective of my dissertation research is to develop new transfer learning methods and demonstrate the utility of the methods in real-world applications. Specifically, in my methodological development, I focus on two different transfer learning scenarios: spatial transfer learning across different domains and temporal transfer learning along time in the same domain. Furthermore, I apply the proposed spatial transfer learning approach to modeling of degenerate biological systems.Degeneracy is a well-known characteristic, widely-existing in many biological systems, and contributes to the heterogeneity, complexity, and robustness of biological systems. In particular, I study the application of one degenerate biological system which is to use transcription factor (TF) binding sites to predict gene expression across multiple cell lines. Also, I apply the proposed temporal transfer learning approach to change detection of dynamic network data. Change detection is a classic research area in Statistical Process Control (SPC), but change detection in network data has been limited studied. I integrate the temporal transfer learning method called the Network State Space Model (NSSM) and SPC and formulate the problem of change detection from dynamic networks into a covariance monitoring problem. I demonstrate the performance of the NSSM in change detection of dynamic social networks.
ContributorsZou, Na (Author) / Li, Jing (Thesis advisor) / Baydogan, Mustafa (Committee member) / Borror, Connie (Committee member) / Montgomery, Douglas C. (Committee member) / Wu, Teresa (Committee member) / Arizona State University (Publisher)
Created2015
Description
Physical activity, sedentary behaviors, and sleep are often associated with cardiometabolic biomarkers commonly found in metabolic syndrome. These relationships are well studied, and yet there are still questions on how each activity may affect cardiometabolic biomarkers. The objective of this study was to examine data from the BeWell24 studies to evaluate the relationship between objectively measured physical activity and sedentary behaviors and cardiometabolic biomarkers in middle age adults, while also determining if sleep quality and duration mediates this relationship. A group of inactive participants (N = 29, age = 52.1 ± 8.1 years, 38% female) with increased risk for cardiometabolic disease were recruited to participate in BeWell24, a trial testing the impact of a lifestyle-based, multicomponent smartphone application targeting sleep, sedentary, and more active behaviors. During baseline, interim (4 weeks), and posttest visits (8 weeks), biomarker measurements were collected for weight (kg), waist circumference (cm), glucose (mg/dl), insulin (uU/ml), lipids (mg/dl), diastolic and systolic blood pressures (mm Hg), and C reactive protein (mg/L). Participants wore validated wrist and thigh sensors for one week intervals at each time point to measure sedentary behavior, physical activity, and sleep outcomes. Long bouts of sitting time (>30 min) significantly affected triglycerides (beta = .15 (±.07), p<.03); however, no significant mediation effects for sleep quality or duration were present. No other direct effects were observed between physical activity measurements and cardiometabolic biomarkers. The findings of this study suggest that reductions in long bouts of sitting time may support reductions in triglycerides, yet these effects were not mediated by sleep-related improvements.
ContributorsLanich, Boyd (Author) / Buman, Matthew (Thesis advisor) / Ainsworth, Barbara (Committee member) / Huberty, Jennifer (Committee member) / Arizona State University (Publisher)
Created2017
Description
Hepatocellular carcinoma (HCC) is a malignant tumor and seventh most common cancer in human. Every year there is a significant rise in the number of patients suffering from HCC. Most clinical research has focused on HCC early detection so that there are high chances of patient's survival. Emerging advancements in functional and structural imaging techniques have provided the ability to detect microscopic changes in tumor micro environment and micro structure. The prime focus of this thesis is to validate the applicability of advanced imaging modality, Magnetic Resonance Elastography (MRE), for HCC diagnosis. The research was carried out on three HCC patient's data and three sets of experiments were conducted. The main focus was on quantitative aspect of MRE in conjunction with Texture Analysis, an advanced imaging processing pipeline and multi-variate analysis machine learning method for accurate HCC diagnosis. We analyzed the techniques to handle unbalanced data and evaluate the efficacy of sampling techniques. Along with this we studied different machine learning algorithms and developed models using them. Performance metrics such as Prediction Accuracy, Sensitivity and Specificity have been used for evaluation for the final developed model. We were able to identify the significant features in the dataset and also the selected classifier was robust in predicting the response class variable with high accuracy.
ContributorsBansal, Gaurav (Author) / Wu, Teresa (Thesis advisor) / Mitchell, Ross (Thesis advisor) / Li, Jing (Committee member) / Arizona State University (Publisher)
Created2013
Description
High fiber diets have been associated with improved cardiometabolic health with specific efforts to lower circulating levels of low-density lipoprotein (LDL cholesterol). Whole grain and grain-based foods are major contributors of dietary fiber in the American diet, of which wheat has been extensively studied. Corn, however, has not been well studied for its cholesterol-lowering properties. Further, the mechanisms by which grains improve cardiometabolic health require further exploration with regard to the human microbiome. The objective of this single-blind randomized controlled, crossover trial was to assess the impact of three different corn flours (whole grain, refined, and bran-enhanced refined flour mixture) on serum LDL cholesterol and the gut microbiota diversity and composition. Twenty-three participants were recruited, between the ages of 18-70 with hypercholesterolemia (Male = 10, Female = 13, LDL >120 mg/dL) who were not taking any cholesterol-lowering medications. Participants consumed each flour mixture for 4 weeks prepared as muffins and pita breads. At the beginning and end of each 4-week period serum for cholesterol assessment, anthropometrics, and stool samples were obtained. Serum cholesterol was assessed using a clinical analyzer. Stool samples were processed, and microbial DNA extracted and sequenced based on the 16S rRNA gene. A generalized linear model demonstrated a significant treatment effect (p=0.016) on LDL cholesterol and explained a majority of the variance (R-squared= 0.89). Post hoc tests revealed bran-enhanced refined flour had a significant effect on cholesterol in comparison to whole grain flour (p=0.001). No statistically significant differences were observed for gut microbial community composition (Jaccard and weighted Unifrac) after corn consumption. However, relative abundance analysis (LEfSE) identified Mycobacterium celatum (p=0.048 FDR=0.975) as a potential marker of post-corn consumption with this microbe being differentially less abundant following bran-enhanced flour treatment. These data suggest that corn flour consumption may be beneficial for individuals with hypercholesterolemia but the role of gut microbiota in this relationship requires further exploration, especially given the small sample size. Further research and analysis of a fully powered cohort is needed to more accurately describe the associations and potential mechanisms of corn-derived dietary fiber on circulating LDL cholesterol and the gut microbiota.
ContributorsWilson, Shannon L (Author) / Whisner, Corrie M (Thesis advisor) / Sears, Dorothy (Committee member) / Buman, Matthew (Committee member) / Dickinson, Jared (Committee member) / Zhu, Qiyun (Committee member) / Arizona State University (Publisher)
Created2022
Description
Deep learning is a sub-field of machine learning in which models are developed to imitate the workings of the human brain in processing data and creating patterns for decision making. This dissertation is focused on developing deep learning models for medical imaging analysis of different modalities for different tasks including detection, segmentation and classification. Imaging modalities including digital mammography (DM), magnetic resonance imaging (MRI), positron emission tomography (PET) and computed tomography (CT) are studied in the dissertation for various medical applications. The first phase of the research is to develop a novel shallow-deep convolutional neural network (SD-CNN) model for improved breast cancer diagnosis. This model takes one type of medical image as input and synthesizes different modalities for additional feature sources; both original image and synthetic image are used for feature generation. This proposed architecture is validated in the application of breast cancer diagnosis and proved to be outperforming the competing models. Motivated by the success from the first phase, the second phase focuses on improving medical imaging synthesis performance with advanced deep learning architecture. A new architecture named deep residual inception encoder-decoder network (RIED-Net) is proposed. RIED-Net has the advantages of preserving pixel-level information and cross-modality feature transferring. The applicability of RIED-Net is validated in breast cancer diagnosis and Alzheimer’s disease (AD) staging. Recognizing medical imaging research often has multiples inter-related tasks, namely, detection, segmentation and classification, my third phase of the research is to develop a multi-task deep learning model. Specifically, a feature transfer enabled multi-task deep learning model (FT-MTL-Net) is proposed to transfer high-resolution features from segmentation task to low-resolution feature-based classification task. The application of FT-MTL-Net on breast cancer detection, segmentation and classification using DM images is studied. As a continuing effort on exploring the transfer learning in deep models for medical application, the last phase is to develop a deep learning model for both feature transfer and knowledge from pre-training age prediction task to new domain of Mild cognitive impairment (MCI) to AD conversion prediction task. It is validated in the application of predicting MCI patients’ conversion to AD with 3D MRI images.
ContributorsGao, Fei (Author) / Wu, Teresa (Thesis advisor) / Li, Jing (Committee member) / Yan, Hao (Committee member) / Patel, Bhavika (Committee member) / Arizona State University (Publisher)
Created2019