2020 MU dissertations - Access restricted to MUhttps://hdl.handle.net/10355/779442024-03-29T11:07:44Z2024-03-29T11:07:44Z3D polymeric scaffolds towards biomedical applicationsZhang, Chenghttps://hdl.handle.net/10355/780162022-12-06T20:34:01Z2020-01-01T00:00:00Z3D polymeric scaffolds towards biomedical applications
Zhang, Cheng
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI--COLUMBIA AT REQUEST OF AUTHOR.] How can research on mechanical engineering and materials science contribute to human health? The fabrication of biomedical scaffolds could be a good entry point. Scaffolds are broadly applied in biomedical fields with multiple functions, such as repair, replacement, and stimulation and monitoring when they are integrated with electronic/optoelectronic devices. Besides biocompatible, the scaffolds should be soft and in form of three-dimensional (3D) structures in order to mechanically and geometrically match the natural tissues and organs. Polymers are the most promising candidate materials for the scaffold fabrication. Compared to metals and ceramics, substantial polymers have biocompatibility and all of them have low Young's modulus and high processability. Benefiting from the high processability, a variety of approaches can be used to shape polymeric scaffolds with 3D architectures. The major three approaches are flexibility, stress induced assembly, and printing. However, none of them is flawless: (1) For flexibility, the scaffolds that integrated with electronic devices have large thickness which exponentially lower the flexibility. (2) For stress-induced assembly, the assembly operation requires complicated actuation equipment and the assembled scaffolds are usually tethered on cumbersome elastomeric substrates. (3) For printing, few of scaffolds fabricated by emerging 4D printing technologies are responsive to biocompatible stimuli. This dissertation aims at addressing these three problems. First, a new device structure, i.e., lateral electrode, is proposed to reduce the thickness and then improve the flexibility of the scaffolds with electronics, which is validated by fabricating flexible photodetectors on polyimide substrates. The photodetectors have excellent flexibility and can be bent to 3D structures. Second, a new stress-induced assembly strategy, i.e., responsive buckling, is developed in which the elastomeric substrates are replaced with deft responsive polymeric substrates. Various 3D polymeric scaffolds either with or without electronic devices are assembled when the substrates are exposed to external stimuli without manual intervention. This strategy is first verified by an acetone responsive organogel and then developed toward biomedical applications by using a body temperature responsive hydrogel. Third, a new shape memory polymer, i.e., poly (glycerol dodecanoate) acrylate (PGDA), whose transition temperature is in the range of 20-37 [degrees]C, is exploited for 4D printing of scaffolds. Because of the propriate transition temperature, the shape memory process of the scaffolds can be completed by using room temperature and body temperature as stimuli, which are harmless for human body. Moreover, a variety of delicate 3D structures including an artery-like tube are printed.
2020-01-01T00:00:00ZAccelerating materials processing via machine learning : towards autonomous manufacturingXie, Yunchaohttps://hdl.handle.net/10355/864622022-12-06T20:33:56Z2020-01-01T00:00:00ZAccelerating materials processing via machine learning : towards autonomous manufacturing
Xie, Yunchao
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI-COLUMBIA AT REQUEST OF AUTHOR.] "Given the advantages of ML, this research thesis focuses on developing appropriate methodologies to solve the problems that inherit in the three phases of the traditional research paradigm: chemical space designing, characterization data analysis, manual experimental conduction. The goal is to accelerate the materials development with the aid of ML towards autonomous manufacturing. Correspondingly, the research work consists of three parts: (i) guiding materials synthesis by ML; (ii) automating XRD spectra analysis by convolutional neural networks (CNNs); (iii) accelerating materials processing via combination of ML and a robotic platform. The part I corresponds Chapter 2, the part II corresponds to Chapter 3, and part III includes Chapter 4 and Chapter 5. In summary, Chapter 2 describes a well-trained ML model to predict the crystallinity propensity of metal-organic nanocapsules (MONCs). Chapter 3 reports a convolutional neural networks (CNNs) for rapid identification of X-ray diffraction (XRD) patterns of metal-organic frameworks (MOFs). Chapter 4 shows development of efficient materials synthesis methodologies, i.e., microwave-/Joule-heating for accelerating the synthesis of MOFs using laser-induced graphene (LIG) microreactors. Chapter 5 illustrates a robotic synthesis platform that is powered by ML algorithms for optimizing the crystallinity of MOFs. Chapter 6 concludes the research work and presents the possible research direction in future."--From Introduction.
2020-01-01T00:00:00ZAccurate and robust animal species classification in the wildAhmed, Ahmed Qasimhttps://hdl.handle.net/10355/842612023-12-14T19:22:25Z2020-01-01T00:00:00ZAccurate and robust animal species classification in the wild
Ahmed, Ahmed Qasim
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI--COLUMBIA AT REQUEST OF AUTHOR.] Wildlife monitoring with camera-traps allows us to collect data at large scales in space and time to study the impact of climate changes, land-use, and human actions on wildlife population dynamics, and biodiversity. With the increase of camera trap used per study and the large number of images generated by camera traps, processing and managing the generated images has become a big challenge. Unlike many other image processing and vision analysis tasks, detecting, segmenting, and classifying animal species from the camera-trap images is very challenging since natural scenes in the wild are often highly cluttered due to heavy vegetation and dynamic background. In this dissertation, we focus on animal object detection and species classification in camera-trap images collected in highly cluttered natural scenes. Using a deep neural network (DNN) model trained for animal- background image classification, we analyze the input camera-trap images to generate its multi-level visual representation. We detect semantic regions of interest for animals from this representation using k-mean clustering and graph cut in the DNN feature domain. These animal regions are then classified into animal species using multi-class deep neural network model. According the experimental results, our method achieves 99.75% accuracy for classifying animals and background and 90.89 percent accuracy for classifying 26 animal species on the Snapshot Serengeti dataset, outperforming existing image classification methods. We develop a robust learning method for animal classification from camera-trap images collected in highly cluttered natural scenes and annotated with noisy labels. We proposed two different network structures with and without clean samples to handle noisy labels. We use k-means clustering to divide the training samples into groups with different characteristics, which are then used to train different networks. These networks with enhanced diversity are then used to jointly predict or correct sample labels using max voting. We evaluate the performance of the proposed method on two public available camera-trap image datasets: Snapshot Serengeti and Panama-Netherlands datasets. Our experimental results demonstrate that our method outperforms the state-of-the-art methods from the literature and achieved improved accuracy on animal species classification from camera-trap images with high levels of label noise. We also develop a new approach for learning a deep neural network for image classification with noise labels using ensemble diversified learning. We first partition the training set into multiple subsets with diversified image characteristics. For each subset, we train a deep neural network image classifier. These networks are then used to encode the input image into different feature vectors, providing diversified observations of the input image. The encoded features are then fused together and further analyzed by a decision network to produce the final classification output. We also study image classification on noise labels with and without the access to clean samples. Our extensive experimental results on the CIFAR-10 and MNIST datasets demonstrate that our proposed method outperforms existing methods by a large margin.
2020-01-01T00:00:00ZAdolescent sibling caregiving and responsibilty and competence in adulthood : retrospective reports among Latina/o young adultsKline, Gabrielle C.https://hdl.handle.net/10355/837262023-12-14T19:22:29Z2020-01-01T00:00:00ZAdolescent sibling caregiving and responsibilty and competence in adulthood : retrospective reports among Latina/o young adults
Kline, Gabrielle C.
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI--COLUMBIA AT REQUEST OF AUTHOR.] Using attachment theory, cultural ecological framework, and an adapted cultural transmission model, I examined associations between familism values, sibling caregiving and success during young adulthood, specifically personal accountability and behavioral and emotional control, among 350 Latina/o/x young adults. I found that familism values were positively associated with sibling caregiving and personal accountability. Sibling caregiving was positively associated with personal accountability and negatively associated with behavioral and emotional control. I found two significant in direct effects. First, familism values were positively associated with sibling caregiving, which, in turn was positively associated with personal accountability. Second, familism values were positively associated with sibling caregiving, which, in turn was negatively associated with behavioral and emotional control. I conducted multigroup path analysis based on gender and found no gender differences. Findings highlight how sibling caregiving can be linked to responsibility during young adulthood.
2020-01-01T00:00:00Z