Physical Sciences and Mathematics Commons^™

Automatic Hemorrhage Segmentation In Brain Ct Scans Using Curriculum-Based Semi-Supervised Learning, Solayman H. Emon, Tzu-Liang (Bill) Tseng, Michael Pokojovy, Peter Mccaffrey, Scott Moen, Md Fashiar Rahman

Mathematics & Statistics Faculty Publications

One of the major neuropathological consequences of traumatic brain injury (TBI) is intracranial hemorrhage (ICH), which requires swift diagnosis to avert perilous outcomes. We present a new automatic hemorrhage segmentation technique via curriculum-based semi-supervised learning. It employs a pre-trained lightweight encoder-decoder framework (MobileNetV2) on labeled and unlabeled data. The model integrates consistency regularization for improved generalization, offering steady predictions from original and augmented versions of unlabeled data. The training procedure employs curriculum learning to progressively train the model at diverse complexity levels. We utilize the PhysioNet dataset to train and evaluate the proposed approach. The performance results surpass those of …

Infusing Machine Learning And Computational Linguistics Into Clinical Notes, Funke V. Alabi, Onyeka Omose, Omotomilola Jegede

Mathematics & Statistics Faculty Publications

Entering free-form text notes into Electronic Health Records (EHR) systems takes a lot of time from clinicians. A large portion of this paper work is viewed as a burden, which cuts into the amount of time doctors spend with patients and increases the risk of burnout. We will see how machine learning and computational linguistics can be infused in the processing of taking clinical notes. We are presenting a new language modeling task that predicts the content of notes conditioned on historical data from a patient's medical record, such as patient demographics, lab results, medications, and previous notes, with the …

Application Of Mixture Models For Doubly Inflated Count Data, Monika Arora, N. Rao Chaganty

Mathematics & Statistics Faculty Publications

In health and social science and other fields where count data analysis is important, zero-inflated models have been employed when the frequency of zero count is high (inflated). Due to multiple reasons, there are scenarios in which an additional count value of k > 0 occurs with high frequency. The zero- and k-inflated Poisson distribution model (ZkIP) is more appropriate for such situations. The ZkIP model is a mixture distribution with three components: degenerate distributions at 0 and k count and a Poisson distribution. In this article, we propose an alternative and computationally fast expectation–maximization (EM) algorithm to obtain the parameter …

Modeling The Spread Of Covid-19 In Spatio-Temporal Context, S.H. Sathish Indika, Norou Diawara, Hueiwang Anna Jeng, Bridget D. Giles, Dilini S.K. Gamage

Mathematics & Statistics Faculty Publications

This study aims to use data provided by the Virginia Department of Public Health to illustrate the changes in trends of the total cases in COVID-19 since they were first recorded in the state. Each of the 93 counties in the state has its COVID-19 dashboard to help inform decision makers and the public of spatial and temporal counts of total cases. Our analysis shows the differences in the relative spread between the counties and compares the evolution in time using Bayesian conditional autoregressive framework. The models are built under the Markov Chain Monte Carlo method and Moran spatial correlations. …

Fast Multiscale Functional Estimation In Optimal Emg Placement For Robotic Prosthesis Controllers, Jin Ren, Guohui Song, Lucia Tabacu, Yuesheng Xu

Mathematics & Statistics Faculty Publications

Electromyogram (EMG) signals play a significant role in decoding muscle contraction information for robotic hand prosthesis controllers. Widely applied decoders require a large amount of EMG signals sensors, resulting in complicated calculations and unsatisfactory predictions. By the biomechanical process of single degree-of-freedom human hand movements, only several EMG signals are essential for accurate predictions. Recently, a novel predictor of hand movements adopted a multistage sequential adaptive functional estimation (SAFE) method based on the historical functional linear model (FLM) to select important EMG signals and provide precise projections.

However, SAFE repeatedly performs matrix-vector multiplications with a dense representation matrix of the …

Robust Testing Of Paired Outcomes Incorporating Covariate Effects In Clustered Data With Informative Cluster Size, Sandipan Dutta

Mathematics & Statistics Faculty Publications

Paired outcomes are common in correlated clustered data where the main aim is to compare the distributions of the outcomes in a pair. In such clustered paired data, informative cluster sizes can occur when the number of pairs in a cluster (i.e., a cluster size) is correlated to the paired outcomes or the paired differences. There have been some attempts to develop robust rank-based tests for comparing paired outcomes in such complex clustered data. Most of these existing rank tests developed for paired outcomes in clustered data compare the marginal distributions in a pair and ignore any covariate effect on …

Multivariate Distributions Of Correlated Binary Variables Generated By Pair-Copulas, Huihui Lin, N. Rao Chaganty

Mathematics & Statistics Faculty Publications

Correlated binary data are prevalent in a wide range of scientific disciplines, including healthcare and medicine. The generalized estimating equations (GEEs) and the multivariate probit (MP) model are two of the popular methods for analyzing such data. However, both methods have some significant drawbacks. The GEEs may not have an underlying likelihood and the MP model may fail to generate a multivariate binary distribution with specified marginals and bivariate correlations. In this paper, we study multivariate binary distributions that are based on D-vine pair-copula models as a superior alternative to these methods. We elucidate the construction of these binary distributions …

Interaction Between Genetic Risk Scores For Reduced Pulmonary Function And Smoking, Asthma And Endotoxin, Sinjini Sikdar, Annah B. Wyss, Mi Kyeong Lee, Thanh T. Hoang, Marie Richards, Laura E. Beane Freeman, Christine Parks, Peter S. Thorne, John L. Hankinson, David M. Umbach, Alison Motsinger-Reif, Stephanie J. London

Mathematics & Statistics Faculty Publications

Rationale Genome-wide association studies (GWASs) have identified numerous loci associated with lower pulmonary function. Pulmonary function is strongly related to smoking and has also been associated with asthma and dust endotoxin. At the individual SNP level, genome-wide analyses of pulmonary function have not identified appreciable evidence for gene by environment interactions. Genetic Risk Scores (GRSs) may enhance power to identify gene–environment interactions, but studies are few.

Methods We analysed 2844 individuals of European ancestry with 1000 Genomes imputed GWAS data from a case–control study of adult asthma nested within a US agricultural cohort. Pulmonary function traits were FEV₁, …

Comparative Survival Of Asian And White Metastatic Castration-Resistant Prostate Cancer Men Treated With Docetaxel, Susan Halabi, Sandipan Dutta, Catherine M. Tangen, Mark Rosenthal, Daniel P. Petrylak, Ian M. Thompson Jr., Kim N. Chi, Johann S. De Bono, John C. Araujo, Christopher Logothetis, Mario A. Eisenberger, David I. Quinn, Karim Fizazi, Michael J. Morris, Celestia S. Higano, Ian F. Tannock, Eric J. Small, William Kevin Kelly

Mathematics & Statistics Faculty Publications

There are few data regarding disparities in overall survival (OS) between Asian and white men with metastatic castration-resistant prostate cancer (mCRPC). We compared OS of Asian and white mCRPC men treated in phase III clinical trials with docetaxel and prednisone (DP) or a DP-containing regimen. Individual participant data from 8820 men with mCRPC randomly assigned on nine phase III trials to receive DP or a DP-containing regimen were combined. Men enrolled in these trials had a diagnosis of prostate adenocarcinoma. The median overall survival was 18.8 months (95% confidence interval [CI] = 17.4 to 22.1 months) and 21.2 months (95% …

Classification Models Of Idiopathic Pulmonary Fibrosis Patients, Mohammed Alqawba, Luis R. Rodriguez, Norou Diawara, Rebecca T. Beuschel, Maryann Kaler, Amisha V. Barochia, Stewart J. Levine, Steven D. Nathan, Geraldine Grant

Mathematics & Statistics Faculty Publications

Idiopathic pulmonary fibrosis (IPF) is a chronic and fatal interstitial lung disease with no current cure. Progression of IPF is difficult to predict as the clinical course can be highly variable and range from a rapidly deteriorating state to a relatively stable state, or may be characterized by a slow progressive decline. Therefore, the need for an accurate diagnosis and improved tools for monitoring and managing IPF is of paramount importance, all for understanding the mitochondrial structure and the function played in the IPF. Mitochondrial DNA copy number (MtDCN) has been correlated with mortality in IPF patients and is a …

Quantifying The Varying Predictive Value Of Physical Activity Measures Obtained From Wearable Accelerometers On All-Cause Mortality Over Short To Medium Time Horizons In Nhanes 2003-2006, Lucia Tabacu, Mark Ledbetter, Andrew Leroux, Ciprian Crainiceanu, Ekaterina Smirnova

Mathematics & Statistics Faculty Publications

Physical activity measures derived from wearable accelerometers have been shown to be highly predictive of all-cause mortality. Prediction models based on traditional risk factors and accelerometry-derived physical activity measures are developed for five time horizons. The data set contains 2978 study participants between 50 and 85 years old with an average of 13.08 years of follow-up in the NHANES 2003–2004 and 2005–2006. Univariate and multivariate logistic regression models were fit separately for five datasets for one- to five-year all-cause mortality as outcome (number of events 46, 94, 155, 218, and 297, respectively). In univariate models the total activity count (TAC) …

Sparsity Promoting Regularization For Effective Noise Suppression In Spect Image Reconstruction, Wei Zheng, Si Li, Andrzej Krol, C. Ross Schmidtlein, Xueying Zeng, Yuesheng Xu

Mathematics & Statistics Faculty Publications

The purpose of this research is to develop an advanced reconstruction method for low-count, hence high-noise, Single-Photon Emission Computed Tomography (SPECT) image reconstruction. It consists of a novel reconstruction model to suppress noise while conducting reconstruction and an efficient algorithm to solve the model. A novel regularizer is introduced as the nonconvex denoising term based on the approximate sparsity of the image under a geometric tight frame transform domain. The deblurring term is based on the negative log-likelihood of the SPECT data model. To solve the resulting nonconvex optimization problem a Preconditioned Fixed-point Proximity Algorithm (PFPA) is introduced. We prove …

The Predictive Performance Of Objective Measures Of Physical Activity Derived From Accelerometry Data For 5-Year All-Cause Mortality In Older Adults: National Health And Nutritional Examination Survey 2003-2006, Ekaterina Smirnova, Andrew Leroux, Quy Cao, Lucia Tabacu, Vadim Zipunnikov, Ciprian Crainiceanu, Jacek Urbanek

Mathematics & Statistics Faculty Publications

Background: Declining physical activity (PA) is a hallmark of aging. Wearable technology provides reliable measures of the frequency, duration, intensity, and timing of PA. Accelerometry-derived measures of PA are compared to established predictors of 5-year all-cause mortality in older adults in terms of individual, relative, and combined predictive performance.

Methods: Participants between 50 and 85 years old from the 2003-2006 National Health and Nutritional Examination Survey (NHANES, n = 2978) wore a hip-worn accelerometer in the free-living environment for up to 7 days. A total of 33 predictors of 5-year all-cause mortality (number of events = 297), including 20 measures …

The Bivariate Erlang And Its Application In Modeling Recurrence Times Of Kidney Dialysis Data, Norou Diawara, S.H. Sathish Indika, Melva Grant, Edgard M. Maboudou-Tchao

Mathematics & Statistics Faculty Publications

Recent advances in computer modeling allows us to find closer fits to data. Our emphasis is on the interdependence between occurrence at kidney dialysis. The interdependence between kidney dialysis occurrences is modelled by a bivariate exponential that we propose in this article. The application is shown on the McGilchrist and Aisbett kidney data set with the use of the exponential distribution. The proposed bivariate exponential model has exponential marginal densities, correlated via a latent random variables and with finite probability of simultaneous occurrence. Extension of the model to a bivariate Erlang type distribution with same shape parameter is presented.

Symptoms Of Gastro-Oesophageal Reflux Disease And The Severity Of Obstructive Sleep Apnoea Syndrome Are Not Related In Sleep Disorders Center Patients, H.-N. Kim, R. D. Vorona, M. P. Winn, M. Doviak, D. A. Johnson, J. Catesby Ware

Mathematics & Statistics Faculty Publications

Background: Studies suggest obstructive sleep apnea syndrome (OSAS) frequently manifests in patients with gastroesophageal reflux disease (GERD) and that there may be a causal relationship.

Aim: To determine the relationship between OSAS and symptoms of GERD.

Methods: Consecutive patients referred to the Sleep Disorders Center (SDC) 18 years and older with polysomnographically defined OSAS were evaluated prospectively for GERD using a validated symptoms questionnaire. The GERD and OSAS relationship was assessed by 1) determining frequency of GERD in patients with and without OSAS; 2) ascertaining the relationship between OSAS severity categories and presence of GERD; 3) examining GERD score in …

A Comparison Of Esomeprazole And Lansoprazole For Control Of Intragastric Ph In Patients With Symptoms Of Gastro-Oesophageal Reflux Disease, D. A. Johnson, T. Stacy, M. Ryan, T. Wootton, J. Willia, K. Hornbuckle, W. Brooks, M. Doviak

Mathematics & Statistics Faculty Publications

Background: Intragastric acid suppression is the most direct measure of the pharmacodynamic efficacy of proton pump inhibitors, which are the most effective drugs for acid-related diseases.

Aim: To compare the effectiveness of once and twice daily dosing of lansoprazole and esomeprazole in controlling intragastric acidity ( target gastric pH > 4.0) over a 24-hour period.

Methods: In an open-label, two-way crossover study, 45 Helicobacter pylori-negative patients with gastro-oesophageal reflux disease were randomized to receive one of two regimens: 30 mg lansoprazole or esomeprazole 40 mg once daily. Intragastric pH was assessed by 24-hour pH monitoring on day 5 of each regimen. …

A Logistic Model Of Periodic Chemotherapy, J. C. Panetta

Mathematics & Statistics Faculty Publications

A logistic differential equation with a time-varying periodic parameter is used to model the growth of cells, in particular cancer cells, in the presences of chemotherapeutic drugs. The chemotherapeutic effects are modeled by a periodic parameter that modifies the growth rate of the cell tissue. A negative growth rate represents the detrimental effects of the drugs. A simple criterion is obtained for the behavior of the chemotherapy.