28232 research outputs found
Sort by
Alignment Targets, Curve Proportion and Mechanical Loading: Preliminary Analysis of an Ideal Shape Toward Reducing Proximal Junctional Kyphosis.
No full textStudy designRetrospective cohort study.ObjectiveInvestigate risk factors for PJK including theoretical kyphosis, mechanical loading at the UIV and age adjusted offset alignment.Methods373 ASD patients (62.7 yrs ± 9.9; 81%F) with 2-year follow up and UIV of at least L1 and LIV of sacrum were included. Images of patients without PJK, with PJK and with PJF were compared using standard spinopelvic parameters before and after the application of the validated virtual alignment method which corrects for the compensatory mechanisms of PJK. Age-adjusted offset, theoretical thoracic kyphosis and mechanical loading at the UIV were then calculated and compared between groups. A subanalysis was performed based on the location of the UIV (upper thoracic (UT) vs. Lower thoracic (LT)).ResultsAt 2-years 172 (46.1%) had PJK, and 21 (5.6%) developed PJF. As PJK severity increased, the post-operative global alignment became more posterior secondary to increased over-correction of PT, PI-LL, and SVA (all P ConclusionsSpinopelvic over correction, under correction of TK (flattening), and under loading of the UIV (decreased bending moment) were associated with PJK and PJF. These differences are often missed when compensation for PJK is not accounted for in post-operative radiographs
Deep Learning for Applications in Inverse Modeling, Legislator Analysis, and Computer Vision for Security
No full textTo judiciously use machine learning – particularly deep learning – requires identifying how to extract features from data and effectively leveraging those features to make predictions. This dissertation concerns deep learning methods for three applications: inverse modeling, legislator analysis, and computer vision for security. To address inverse problems, we present a new method, the Mixture Manifold Network, which uses multiple neural backward models in a forward-backward architecture. We experimentally demonstrate that the Mixture Manifold Network performs better than computationally fast generative model baselines, while performance approaching that of computationally slow iterative methods. For legislator modeling, we seek to learn representations that capture legislator attitudes that may not be contained in their voting records. We present a model that instead considers their tweeting behavior, and we use reactions to former President Donald Trump on Twitter as an illustrative example. For computer vision, we address two security-related applications using deep convolutional feature extractors. In the first of these, we leverage domain adaptation with deep object detection for threatening items – such as guns, knives, and blunt objects – in X-ray scans of air passenger luggage. In the second, we apply an occlusion-robust classifier to infrared imagery. For each application above, we describe the datasets for the problem, how the presented methods extract features from that data, and how efficacious predictions are produced from each of our proposed models</p
Anything For Views Parenting: Framing Privacy, Ethics, and Norms for Children of Influencers on YouTube
No full textChildren who appear as the main characters or primary consumers of YouTube content have been the focus of emerging academic literature and public debate (Feller & Burroughs, 2022; Ferguson, 2018; Kumar, 2021). Sharenting, or posting information, photos, or videos about one's children on social media, has also been a discussion and concern among researchers, legal scholars, and parents (Kumar, 2021). Sharenting has online and offline consequences. It exposes personal information, such as a child’s name and whereabouts, which may lead to unwanted attention or safety risks (Brosch, 2016; Blum-Ross, 2015). However, there is a significant subsection of YouTube media where children appear as integral supporting characters of an adult’s content that has yet to be meaningfully researched.The normalization of sharenting has coincided with an upsurge of influencers and influencer marketing (Abidin, 2018). The influencer marketing industry was estimated to be worth 16 billion dollars in 2022, projected to increase to 21 billion dollars in 2023 (Geyser, 2023). Influencers who involve their children in content position them, at times, as unintentional microcelebrities or brand assets (Abidin, 2015). When this happens, their appearance in user-generated content contributes to the premise and profitability of their parent’s brand. However, children who consistently contribute to their parent’s brand have no rights to the money their names, images, and likenesses generate. They have no working hours to abide by and no access to representation by a third party acting without a personal stake in their profitability (Geider, 2021). Children are unaware of the long-term consequences of exposure to a digital audience, including potential privacy violations, online harassment, or reputational harm. They may also not fully understand the implications of having a digital identity established for them before they can make decisions for themselves.
While existing literature demonstrates that social media platforms, laws, and policies do not adequately regulate or protect the children of influencers, there has been no effort first to define the child of an influencer and, second, to identify at what point that regulation becomes necessary. In other words, when do influencer parents go beyond mere sharenting? This research project examines the complex interplay between the potential long-term impacts of children's involvement in influencer content and the gaps in regulations related to children’s work on social media.
I aim to analyze the regulatory gray area children of influencers inhabit on YouTube and to identify salient features of influencer content which place children at disproportionate risk of undesirable exposure online. The present study scopes the value children provide to user-generated monetized content. It constructs a typology to describe the unique privacy and psychological risks they are exposed to when their parents' income involves their presence. It outlines common arguments influencer parents use to justify their children's use in content production and discusses the impossibility of informed consent for children in this context.</p
Point-of-care diagnostics for invasive aspergillosis: nearing the finish line.
No full textIntroductionThe spectrum of disease caused by Aspergillus spp. is dependent on the immune system of the host, with invasive aspergillosis (IA) its most severe manifestation. Early and reliable diagnosis of Aspergillus disease is important to decrease associated morbidity and mortality from IA.Areas coveredThe following review searched Pub Med for literature published since 2007 and will give an update on the current point-of-care diagnostic strategies for the diagnosis of IA, discuss needed areas of improvement for these tests, and future directions.Expert opinionSeveral new diagnostic tests for IA - including point-of-care tests - are now available to complement conventional galactomannan (GM) testing. In particular, the Aspergillus-specific Lateral Flow Device (LFD) test and the sōna Aspergillus GM Lateral Flow Assay (LFA) are promising for the diagnosis of IA in patients with hematologic malignancy, although further evaluation in the non-hematology setting is needed. In addition, a true point-of-care test, particularly for easily obtained specimens like serum or urine that can be done at the bedside or in the Clinic in a matter of minutes is needed, such as the lateral flow dipstick test, which is under current evaluation. Lastly, improved diagnostic algorithms to diagnose IA in non-neutropenic patients is needed
Complications of surgical intervention in adult lumbar scoliosis
No full textIf nonoperative measures are unsuccessful in managing the pain and disability of adult spinal deformities, surgical correction may provide the potential for significant improvement in a patient’s quality of life. However, these procedures have a relatively high risk of complications. Identifying patients that may benefit from surgical intervention requires a thorough understanding of potential complications and managing the risks of any individual patient. Complications do not necessarily result in poor outcomes, and good outcomes are not always complication free. Higher risk patients potentially have more to gain, even if they experience complications. With the rapidly expanding senior population and expanded capabilities to manage high-risk patients, it is helpful to consider the lessons provided by ever expanding databases of outcome measures to refine the surgical decision-making process
Febrile Neutropenia: Improving Care Through an Oncology Acute Care Clinic
No full textBACKGROUND: Patients with cancer are at risk for oncologic emergencies, including febrile neutropenia (FN). Timely treatment of FN can prevent complications. Providing this care in the outpatient setting has been shown to be safe and effective. OBJECTIVES: This project implemented and evaluated a new process using an outpatient acute care clinic (ACC) to manage FN in patients with hematologic cancer. The aims were to reduce the time from fever identification to antibiotic administration, decrease emergency department (ED) visit rates, and evaluate patient satisfaction. METHODS: Using a pre-/postimplementation design, an interprofessional team was educated about a new process of caring for patients with hematologic cancer and FN at an outpatient ACC using a comprehensive algorithm. FINDINGS: 31 patients participated in the project (15 pre-and 16 postimplementation). Time to antibiotic administration decreased from 144.88 minutes to 63.69 minutes. Participant visits to the ED decreased by 2.33 times per month on average. Overall, patients were satisfied with the ACC. These findings support using a dedicated outpatient ACC for patients with FN receiving hematology care
Chiral Cation Doping for Modulating Structural Symmetry of 2D Perovskites.
No full textCation mixing in two-dimensional (2D) hybrid organic-inorganic perovskite (HOIP) structures represents an important degree of freedom for modifying organic templating effects and tailoring inorganic structures. However, the limited number of known cation-mixed 2D HOIP systems generally employ a 1:1 cation ratio for stabilizing the 2D perovskite structure. Here, we demonstrate a chiral-chiral mixed-cation system wherein a controlled small amount (S)-(-)-2-methylbutylammonium) can be doped into (S-BrMBA)2PbI4 (S-BrMBA = (S)-(-)-4-bromo-α-methylbenzylammonium), modulating the structural symmetry from a higher symmetry (C2) to the lowest symmetry state (P1). This structural change occurs when the concentration of S-2-MeBA, measured by solution nuclear magnetic resonance, exceeds a critical level─specifically, for 1.4 ± 0.6%, the structure remains as C2, whereas 3.9 ± 1.4% substitution induces the structure change to P1 (this structure is stable to ∼7% substitution). Atomic occupancy analysis suggests that one specific S-BrMBA cation site is preferentially substituted by S-2-MeBA in the unit cell. Density functional theory calculations indicate that the spin splitting along different k-paths can be modulated by cation doping. A true circular dichroism band at the exciton energy of the 3.9% doping phase shows polarity inversion and a ∼45 meV blue shift of the Cotton-effect-type line-shape relative to (S-BrMBA)2PbI4. A trend toward suppressed melting temperature with higher doping concentration is also noted. The chiral cation doping system and the associated doping-concentration-induced structural transition provide a material design strategy for modulating and enhancing those emergent properties that are sensitive to different types of symmetry breaking
Walking Backwards: How the Re-Storying of Collective Identity Unlocks the Potential for Churches to Make Significant Changes to their Congregational Practices
No full textHow do churches change? In the life of congregations, collective identity informs congregational practice which, in turn, informs collective identity, forming a reinforcing loop that artificially prevents congregations from making significant changes to their congregational practice. To change the practices would be to change the identity, and to change the identity would be to change the practices.This thesis explores the interaction of collective identity, congregational practice, and change. After a review of pertinent scholarship concerning organizational and congregational change, this study provides an in-depth analysis of three churches that have made significant changes to their congregational practice in the last decade. Employing a multiple case study methodology, the actions of these congregations are compared to one another and to existing change literature.
In the end, these three congregations demonstrate how the effective use of engaging with their histories to re-story their present collective identities allowed them to meet these new changes in a way that fits with their identities. Rather than preventing them from making significant changes, the reinforcing loop of collective identity and congregational practice propelled them.</p
Explainable Artificial Intelligence Techniques in Medical Imaging Analysis
No full textArtificial intelligence (AI), including classic machine learning (ML) and deep learning (DL), has recently made an impact on advanced medical image analysis. Classic ML learns the data representation by manual image feature engineering, namely radiomics, based on experts' domain knowledge. DL directly learns the image feature through hierarchical data modeling directly from the input data. Both classic ML and DL models have emerged as promising AI tools for medical image analysis. Despite promising academic research in which algorithms are beginning to outperform humans, clinical radiography analysis still has limited AI involvement. One issue of current AI development (for both classic ML and DL) is the lack of model explainability, i.e., the extent to which the internal mechanics of an AI model can be explained in human terms from a clinical perspective. The unexplainable issues include, but are not limited to, model confidence ('Can we trust the results with some clues?'), data utilization ('Do we need this as a part of the model?'), and model generalization ('How do I know if it works?'). Without such model explainability, AI models remain a black box in implementation, which leads to a lack of accountability and confidence in clinic application. We hypothesize that the current medical domain knowledge, both in theory and in practice, can be incorporated into AI designs to provide explainability. Therefore, the objective of this dissertation is to explore potential techniques to enhance AI model explainability. Specifically, three novel AI models were developed:
• The first model aimed to explore a radiomic filtering model to quantify and visualize radiomic features associated with pulmonary ventilation from lung computed tomography (CT). In this model, lung volume was segmented on 46 CT images, and a 3D sliding window kernel was implemented across the lung volume to capture the spatial-encoded image information. Fifty-three radiomic features were extracted within the kernel, resulting in a 4th-order tensor object. As such, each voxel coordinate of the original lung was represented as a 53-dimensional feature vector, such that radiomic features could be viewed as feature maps within the lungs. To test the technique as a potential pulmonary ventilation biomarker, the radiomic feature maps were compared to paired functional images (Galligas-positron emission tomography, PET or DTPA-single photon emission computed tomography, SPECT) based on Spearman correlation (?) analysis. From the results, the radiomic feature map Gray Level Run Length Matrix (GLRLM)-based Run-Length Non-Uniformity and Gray Level Co-occurrence Matrix (GLCOM)-based Sum Average are found to be highly correlated with functional imaging. The achieved ? (median [range]) for the two features are 0.46 [0.05, 0.67] and 0.45 [0.21, 0.65] across 46 patients and 2 functional imaging modalities, respectively. The results provide evidence that local regions of sparsely encoded heterogeneous lung parenchyma on CT are associated with diminished radiotracer uptake and measured lung ventilation defects on PET/SPECT imaging. Collectively, these findings demonstrate the potential of radiomic filtering to provide a visual explanation of lung CT radiomic features associated with lung ventilation. The developed technique may serve as a complementary tool to the current lung quantification techniques and provide hypothesis-generating data for future studies.
• The second model aimed to explore a neural ordinary differential equation (ODE)-based segmentation model to observe deep neural network (DNN) behavior in multi-parametric magnetic resonance imaging (MRI)-based glioma segmentation. In this model, by hypothesizing that deep feature extraction can be modeled as a spatiotemporally continuous process, we implemented a novel DL model, neural ODE, in which deep feature extraction was governed by an ODE parameterized by a neural network. The dynamics of 1) MR images after interactions with the DNN and 2) segmentation formation can thus be visualized after solving the ODE. An accumulative contribution curve (ACC) was designed to quantitatively evaluate each MR image’s utilization by the DNN toward the final segmentation results. The proposed neural ODE model was demonstrated using 369 glioma patients with a 4-modality multi-parametric MRI protocol: T1, contrast-enhanced T1 (T1-Ce), T2, and fluid-attenuated inversion recovery (FLAIR). Three neural ODE models were trained to segment enhancing tumor (ET), tumor core (TC), and whole tumor (WT), respectively. The key MR modalities with significant utilization by DNNs were identified based on ACC analysis. Segmentation results by DNNs using only the key MR modalities were compared to the ones using all 4 MR modalities in terms of Dice coefficient, accuracy, sensitivity, and specificity. From the results, all neural ODE models successfully illustrated image dynamics as expected. ACC analysis identified T1-Ce as the only key modality in ET and TC segmentations, while both FLAIR and T2 were key modalities in WT segmentation. Compared to the U-Net results using all 4 MR modalities, the Dice coefficient of ET (0.784→0.775), TC (0.760→0.758), and WT (0.841→0.837). Collectively, the neural ODE model offers a new tool for optimizing the DL model inputs with enhanced explainability in data utilization. The presented methodology can be generalized to other medical image-related DL applications.
• The third model aimed to explore a multi-feature-combined (MFC) model to quantify the role of radiomic features, DL image features, and their combination in predicting local failure from pre-treatment CT images of early-stage non-small cell lung cancer (NSCLC) patients after either lung surgery or stereotactic body radiation therapy (SBRT). The MFC model comprised three key steps. (1) Extraction of 92 handcrafted radiomic features from the gross tumor volume (GTV) segmented on pre-treatment CT images. (2) Extraction of 512 deep features from pre-trained DL U-Net encoder structure. Specifically, the 512 latent activation values from the last fully connected layers were studied. (3) The extracted 92 handcrafted radiomic features, 512 deep features, along with 4 patient demographic information (i.e., gender, age, tumor volume, and Charlson comorbidity index), were concatenated as a multi-dimensional input to three classifiers: logistic regression (LR), supporting vector machine (SVM), and random forest (RF) to predict the local failure. Two NSCLC patient cohorts from our institution were investigated: (1) the surgery cohort includes 83 patients who underwent segmentectomy or wedge resection (with 7 local failures), and (2) the SBRT cohort includes 84 patients who received lung SBRT (with 9 local failures). The MFC model was developed and evaluated independently for both patient cohorts. For each cohort, the MFC model was also compared against (1) the R model: LR/SVM/RF prediction models using only radiomic features, (2) the PI model: LR/SVM/RF prediction models using only patient demographic information, and (3) the DL model: DL design that directly predicts the local failure based on the U-Net encoder. All models were tested based on two validation methods: leave-one-out cross-validation (LOOCV) and 100-fold Monte Carlo cross-validation (MCCV) with a 70%-30% train-test ratio. ROC with AUC analysis was adopted as the main evaluator to measure the prediction performance. The student’s t-test was performed to identify the statistically significant differences when applicable. In LOOCV, the AUC range of the proposed MFC model (for three classifiers) was 0.811-0.956 for the surgery patient cohort and 0.913-0.981 for the SBRT cohort, which was higher than the other studied models: the AUC range was 0.356-0.480 (surgery) and 0.295-0.347 (SBRT) for the PI models, 0.388-0.655 (surgery) and 0.648-0.747 (SBRT) for the R models, and 0.816 (surgery) and 0.842 (SBRT) for the DL models. Similar results can be observed in the 100-fold MCCV: the MFC model again showed the highest AUC results (surgery: 0.831-0.841, SBRT: 0.860-0.947), which were significantly higher than the PI models (surgery: 0.464-0.564, SBRT: 0.457-0.519), R models (surgery: 0.546-0.653, SBRT: 0.559-0.667), and DL models (surgery: 0.690, SBRT: 0.773). Collectively, the developed MFC model improves the ability to predict the occurrence of local failure for both surgery and SBRT patient cohorts with enhanced explainability in the role of different feature sources. It may hold the potential to assist clinicians to optimize treatment procedures in the future.
In summary, the three developed models provide substantial contributions to enhance the explainability of current classic ML and DL models. The concepts and techniques developed in this dissertation, as well as understandings and inspirations from the key results, provide valuable knowledge for the future development of AI techniques toward wide clinical trust and acceptance.</p
Asthma, Airflow Obstruction, and Eosinophilic Airway Inflammation Prevalence in Western Kenya: A Population-Based Cross-Sectional Study.
No full textObjectives: Determine the prevalence of airway disease (e.g., asthma, airflow obstruction, and eosinophilic airway inflammation) in Kenya, as well as related correlates of airway disease and health-related quality of life. Methods: A three-stage, cluster-randomized cross-sectional study in Uasin Gishu County, Kenya was conducted. Individuals 12 years and older completed questionnaires (including St. George's Respiratory Questionnaire for COPD, SGRQ-C), spirometry, and fractional exhaled nitric oxide (FeNO) testing. Prevalence ratios with 95% confidence intervals (CIs) were calculated. Multivariable models were used to assess correlates of airflow obstruction and high FeNO. Results: Three hundred ninety-two participants completed questionnaires, 369 completed FeNO testing, and 305 completed spirometry. Mean age was 37.5 years; 64% were women. The prevalence of asthma, airflow obstruction on spirometry, and eosinophilic airway inflammation was 21.7%, 12.3% and 15.7% respectively in the population. Women had significantly higher SGRQ-C scores compared to men (15.0 vs. 7.7). Wheezing or whistling in the last year and SGRQ-C scores were strongly associated with FeNO levels >50 ppb after adjusting for age, gender, BMI, and tobacco use. Conclusion: Airway disease is a significant health problem in Kenya affecting a young population who lack a significant tobacco use history