Matching Items (11)
Filtering by
- All Subjects: Health Sciences
- All Subjects: Transfer Learning
- Creators: Hooker, Steven
- Creators: Li, Jing
Description
Introduction: Several faith-based or faith-placed programs have focused on the physical dimension of wellness in efforts to improve health by increasing physical activity and improving diet behaviors. However, these programs were not designed to intervene on the mental dimension of wellness which is critical for stress reduction and health behavior change. Purpose: To evaluate the feasibility of a spirituality-based stress reduction and health behavior change intervention using the Spiritual Framework of Coping (SFC) model. Methods: This study was a quasi-experimental one group pretest posttest design. The study was a total of eight weeks conducted at a non-denominational Christian church. Participants were recruited from the church through announcements and flyers. The Optimal Health program met once a week for 1.5 hours with weekly phone calls during an additional four week follow-up period. Feasibility was assessed by the acceptability, demand, implementation, practicality, integration, and limited efficacy of the program. Analysis: Frequencies for demographics were assessed. Statistical analyses of feasibility objectives were assessed by frequencies and distribution of responses to feasibility evaluations. Limited efficacy of pretest and posttest measures were conducted using paired t-test (p <.05). Results: The Optimal Health Program was positively accepted by participants. The demand for the program was shown with average attendance of 78.7%. The program was successfully implemented as shown by meeting session objectives and 88% homework completion. The program was both practical for the intended participants and was successfully integrated within the existing environment. Limited efficacy changes within the program were mostly non-significant. Conclusion: This study tested the feasibility of implementing the Optimal Health program that specifically targeted the structural components of the Spiritual Framework of Coping Model identified to create meaning making and enhance well-being. This program may ultimately be used to help individuals improve and balance the spiritual, mental, and physical dimensions of wellness. However, length of study and limited efficacy measures will need to be reevaluated for program success.
ContributorsWalker, Jenelle R (Author) / Swan, Pamela (Thesis advisor) / Ainsworth, Barbara (Committee member) / Chisum, Jack (Committee member) / Fleury, Julie (Committee member) / Hooker, Steven (Committee member) / Arizona State University (Publisher)
Created2012
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
Background: Heart failure is the leading cause of hospitalization in older adults and has the highest 30-day readmission rate of all diagnoses. An estimated 30 to 60 percent of older adults lose some degree of physical function in the course of an acute hospital stay. Few studies have addressed the role of posture and mobility in contributing to, or improving, physical function in older hospitalized adults. No study to date that we are aware of has addressed this in the older heart failure population.
Purpose: To investigate the predictive value of mobility during a hospital stay and patterns of mobility during the month following discharge on hospital readmission and 30-day changes in functional status in older heart failure patients.
Methods: This was a prospective observational study of 21 older (ages 60+) patients admitted with a primary diagnosis of heart failure. Patients wore two inclinometric accelerometers (rib area and thigh) to record posture and an accelerometer placed at the ankle to record ambulatory activity. Patients wore all sensors continuously during hospitalization and the ankle accelerometer for 30 days after hospital discharge. Function was assessed in all patients the day after hospital discharge and again at 30 days post-discharge.
Results: Five patients (23.8%) were readmitted within the 30 day post-discharge period. None of the hospital or post-discharge mobility measures were associated with readmission after adjustment for covariates. Higher percent lying time in the hospital was associated with slower Timed Up and Go (TUG) time (b = .08, p = .01) and poorer hand grip strength (b = -13.94, p = .02) at 30 days post-discharge. Higher daily stepping activity during the 30 day post-discharge period was marginally associated with improvements in SPPB scores at 30 days (b = <.001, p = .06).
Conclusion: For older heart failure patients, increased time lying while hospitalized is associated with slower walking time and poor hand grip strength 30 days after discharge. Higher daily stepping after discharge may be associated with improvements in physical function at 30 days.
Purpose: To investigate the predictive value of mobility during a hospital stay and patterns of mobility during the month following discharge on hospital readmission and 30-day changes in functional status in older heart failure patients.
Methods: This was a prospective observational study of 21 older (ages 60+) patients admitted with a primary diagnosis of heart failure. Patients wore two inclinometric accelerometers (rib area and thigh) to record posture and an accelerometer placed at the ankle to record ambulatory activity. Patients wore all sensors continuously during hospitalization and the ankle accelerometer for 30 days after hospital discharge. Function was assessed in all patients the day after hospital discharge and again at 30 days post-discharge.
Results: Five patients (23.8%) were readmitted within the 30 day post-discharge period. None of the hospital or post-discharge mobility measures were associated with readmission after adjustment for covariates. Higher percent lying time in the hospital was associated with slower Timed Up and Go (TUG) time (b = .08, p = .01) and poorer hand grip strength (b = -13.94, p = .02) at 30 days post-discharge. Higher daily stepping activity during the 30 day post-discharge period was marginally associated with improvements in SPPB scores at 30 days (b = <.001, p = .06).
Conclusion: For older heart failure patients, increased time lying while hospitalized is associated with slower walking time and poor hand grip strength 30 days after discharge. Higher daily stepping after discharge may be associated with improvements in physical function at 30 days.
ContributorsFloegel, Theresa A (Author) / Buman, Matthew P (Thesis advisor) / Hooker, Steven (Committee member) / Dickinson, Jared (Committee member) / DerAnanian, Cheryl (Committee member) / McCarthy, Marianne (Committee member) / Arizona State University (Publisher)
Created2015
Description
Background: Hispanic women are at high risk for Type 2 Diabetes (T2D), in part due to their high prevalence of obesity, which may influence the development of insulin resistance and disease onset. Unhealthy eating contributes to T2D risk. Dietary patterns are the combination of total foods and beverages among individual’s over time, but there is limited information regarding its role on T2D risk factors among Hispanic women. Objective: To identify a posteriori dietary patterns and their associations with diabetes risk factors (age, BMI, abdominal obesity, elevated fasting blood glucose, and hemoglobin A1c) among overweight/obese Hispanic women. Design: Cross-sectional dietary data were collected among 191 women with or at risk for T2D using the Southwestern Food Frequency Questionnaire capturing the prior three months of intake. Dietary patterns were derived using exploratory factor analysis. Regression scores were used to explore associations between dietary patterns and diabetes risk factors. Results: The patterns derived were: 1) “sugar and fat-laden”, with high loads of sweets, drinks, pastries, and fats; 2) “plant foods and fish”, with high loads of vegetables, fruits, fish, and beans; 3) “soups and starchy dishes”, with high loads of soups, starchy foods, and mixed dishes; 4) “meats and snacks”, with high loads of red meat, salty snacks, and condiments; 5) “beans and grains”, with high loads of beans and seeds, whole-wheat and refined grain foods, fish, and alcohol; and 6) “eggs and dairy”, with high loads of eggs, dairy, and fats. The “sugar and fat-laden” and “meats and snacks” patterns were negatively associated with age (r= -0.230, p= 0.001 and r= -0.298, p<0.001, respectively). Scores for “plant foods and fish” were associated with fasting blood glucose (r= 0.152, p= 0.037). There were no other statistically significant relationships between the dietary patterns and risk factors for T2D. Conclusions: A variety of patterns with healthy and unhealthy traits among Hispanic women were observed. Being younger may play an important role in adhering to a dietary pattern rich in sugary and high-fat foods and highlights the importance of assessing dietary patterns among young women to early identify dietary traits detrimental for their health.
ContributorsArias-Gastelum, Mayra (Author) / Vega-Lopez, Sonia (Thesis advisor) / Der Ananian, Cheryl (Committee member) / Whisner, Corrie (Committee member) / Bruening, Meg (Committee member) / Hooker, Steven (Committee member) / Arizona State University (Publisher)
Created2018
Description
Most American children consume less than the recommend amount of fruits and vegetables (F&V), 74% and 84%, respectively. Eating too few F&V in childhood is associated with increased risk of cardiovascular disease, hypertension, respiratory symptoms, and some cancers later in life. Adequate F&V consumption favorably impacts antioxidant status, gut flora, mood, and cognitive functioning. Nutrients such as vitamin C and fiber are only naturally occurring in plant foods. For many children, school lunches are an important source of F&V. This pilot study assessed the feasibility of providing condiments to increase children’s consumption of salad bar F&V in an elementary school cafeteria at lunchtime. The trial site was a single Title 1 elementary school in a large, urban district in the greater Phoenix metropolitan area. Taste tests were conducted on three convenience samples of children in grades 3 – 7, aged 8 – 12 years (n=57) to identify the most popular condiment flavors. The five highest rated flavors were made available daily at a “flavor station” in the school’s lunchroom for three consecutive weeks during the Fall 2018 semester. Descriptive and inferential statistics were used to analyze data. A cost analysis was conducted for capital outlays related to the flavor station. School employee perceptions of F&V and the flavor station were assessed via posttest online surveys. Peanut butter was rated the best tasting condiment by children and was the only condiment that increased in popularity throughout the intervention. Overall, daily F&V consumption increased 17 g per child. There was a linear increase in F&V consumption during the study (r=0.986; P=0.014). As a proportion of the total F&V selected, F&V waste decreased by nearly 3%. The average daily cost of providing the flavor station was $0.09 per student. Sixty-five percent of school staff felt that the flavor station should continue at their school. Peanut butter is an affordable, nutrient-dense food that accommodates the USDA Food and Nutrition Service meal patterns and nutrition standards, and thus, is a viable strategy for increasing F&V consumption and decreasing F&V waste. The results herein inform the development of future interventions to improve the palatability of F&V for children.
ContributorsScholtz, Cameron (Author) / Johnston, Carol (Thesis advisor) / Alexon, Christy (Committee member) / Hooker, Steven (Committee member) / Schwake, David (Committee member) / Swan, Pamela (Committee member) / Wharton, Christopher (Christopher Mack), 1977- (Committee member) / Arizona State University (Publisher)
Created2019
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
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
The purpose of this phenomenological study was to explore the cultural, social, environmental, and gender factors that may influence physical activity (PA) in older Mexican American (MA) men living in Tucson, Arizona. The Mexican origin population is the fastest growing Hispanic subgroup in our nation, increasing from 20.6 million in the year 2000 to 31.8 million in 2010. Arizona has the sixth largest Hispanic population in the United States and the Mexican origin population accounts for 91% of Arizona's Hispanics. Despite the fast growing Mexican population, there are a limited number of studies that examine MAs and PA. There are even fewer interventions created to foster PA among older (≥65 years old) MA men. Fourteen individual interviews were conducted with older MA men living in Tucson, Arizona. Data was collected, organized, and analyzed according to the methodologies of Clark Moustakas and the Social Ecology Model for Health Promotion framework. Six themes emerged which reflected the older MA male's perception of health, masculinity, and physical activity: a) Retirement promotes self-care behaviors, b) Women, health care providers, and the Internet are important in promoting health, c) Aging changes physical activity, d) I take care of myself, e) Physical activity is a personal choice and lifestyle, and f) I learn and make adjustments as needed. Themes were used to create textural and structural descriptions of their experiences. Descriptions were formed into the essence of the phenomenon. The results of this study increase our understanding of health, masculinity, and physical activity in older MA men. This research will inform the development of an evidence-based PA intervention to promote cardiovascular (CV) health in older MA men that may be implemented in a variety of community-based settings.
ContributorsDowling, Evangeline M (Author) / Hooker, Steven (Thesis advisor) / Grando, Victoria (Committee member) / Der Ananian, Cheryl (Committee member) / Larkey, Linda (Committee member) / Arizona State University (Publisher)
Created2015
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
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