Digital Commons Network^™

Low Doses Of Bisphenol S Affect Post-Translational Modifications Of Sperm Proteins In Male Mice, Hedvika Řimnáčová, Miriam Štiavnická, Jiří Moravec, Marouane Chemek, Yaroslav Kolinko, Olga García-Álvarez, Peter R. Mouton, Azalia Mariel Carranza Trejo, Tereza Fenclová, Nikola Eretová, Petr Hošek, Pavel Klein, Milena Králíčková, Jaroslav Petr, Jan Nevoral

Computer Science and Engineering Faculty Publications

Background: Bisphenol S (BPS) is increasingly used as a replacement for bisphenol A in the manufacture of products containing polycarbonates and epoxy resins. However, further studies of BPS exposure are needed for the assessment of health risks to humans. In this study we assessed the potential harmfulness of low-dose BPS on reproduction in male mice.

Methods: To simulate human exposure under experimental conditions, 8-week-old outbred ICR male mice received 8 weeks of drinking water containing a broad range of BPS doses [0.001, 1.0, or 100 µg/kg body weight (bw)/day, BPS1-3] or vehicle control. Mice were sacrificed and testicular tissue taken …

Deep Neural Ranking For Crowdsourced Geopolitical Event Forecasting, Giuseppe Nebbione, Derek Doran, Srikanth Nadella, Brandon Minnery

Computer Science and Engineering Faculty Publications

There are many examples of “wisdom of the crowd” effects in which the large number of participants imparts confidence in the collective judgment of the crowd. But how do we form an aggregated judgment when the size of the crowd is limited? Whose judgments do we include, and whose do we accord the most weight? This paper considers this problem in the context of geopolitical event forecasting, where volunteer analysts are queried to give their expertise, confidence, and predictions about the outcome of an event. We develop a forecast aggregation model that integrates topical information about a question, meta-data about …

Visual Entailment: A Novel Task For Fine-Grained Image Understanding, Ning Xie, Farley Lai, Derek Doran, Asim Kadav

Computer Science and Engineering Faculty Publications

Existing visual reasoning datasets such as Visual Question Answering (VQA), often suffer from biases conditioned on the question, image or answer distributions. The recently proposed CLEVR dataset addresses these limitations and requires fine-grained reasoning but the dataset is synthetic and consists of similar objects and sentence structures across the dataset. In this paper, we introduce a new inference task, Visual Entailment (VE) - consisting of image-sentence pairs whereby a premise is defined by an image, rather than a natural language sentence as in traditional Textual Entailment tasks. The goal of a trained VE model …

Augmenting Flight Imagery From Aerial Refueling, James D. Anderson, Scott Nykl, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

© 2019, This is a U.S. government work and not under copyright protection in the U.S.; foreign copyright protection may apply. When collecting real-world imagery, objects in the scene may be occluded by other objects from the perspective of the camera. However, in some circumstances an occluding object is absent from the scene either for practical reasons or the situation renders it infeasible. Utilizing augmented reality techniques, those images can be altered to examine the affect of the object’s occlusion. This project details a novel method for augmenting real images with virtual objects in a virtual environment. Specifically, images from …

An Interactive Game For Cultural Proficiencytraining Featuring Virtual Reality Immersion, Paul J. Hershberger, Blaine A. Klingler, Matt Davis, Sankalp Mishra, Miteshkumar Vasoya, Dixit Patel, Aishwarya Bositty, Tanuja Addanki, Frank A. Allen, Suneesh Menon, Sabrina Neeley, Angie Castle, Todd Pavlak, Yong Pei, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

No abstract provided.

Visual Entailment Task For Visually-Grounded Language Learning, Ning Xie, Farley Lai, Derek Doran, Asim Kadav

Computer Science and Engineering Faculty Publications

We introduce a new inference task - Visual Entailment (VE) - which differs from traditional Textual Entailment (TE) tasks whereby a premise is defined by an image, rather than a natural language sentence as in TE tasks. A novel dataset SNLI-VE (publicly available at https://github.com/necla-ml/SNLI-VE) is proposed for VE tasks based on the Stanford Natural Language Inference corpus and Flickr30k. We introduce a differentiable architecture called the Explainable Visual Entailment model (EVE) to tackle the VE problem. EVE and several other state-of-the-art visual question answering (VQA) based models are evaluated on the SNLI-VE dataset, facilitating grounded language understanding and providing …

Xr-Based Workforce Develop In The Southwestern Region Of Ohio, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

No abstract provided.

Delta Radiomic Features Improve Prediction For Lung Cancer Incidence: A Nested Case–Control Analysis Of The National Lung Screening Trial, Dmitry Cherezov, Samuel H. Hawkins, Dmitry B. Goldgof, Lawrence O. Hall, Ying Liu, Qian Li, Yoganand Balagurunathan, Robert J. Gillies, Matthew B. Schabath

Computer Science and Engineering Faculty Publications

Background: Current guidelines for lung cancer screening increased a positive scan threshold to a 6 mm longest diameter. We extracted radiomic features from baseline and follow‐up screens and performed size‐specific analyses to predict lung cancer incidence using three nodule size classes (<6 mm [small], 6‐16 mm [intermediate], and ≥16 mm [large]).

Methods: We extracted 219 features from baseline (T0) nodules and 219 delta features which are the change from T0 to first follow‐up (T1). Nodules were identified for 160 incidence cases diagnosed with lung cancer at T1 or second follow‐up screen (T2) and for 307 nodule‐positive controls that had three consecutive positive screens not diagnosed as lung cancer. The …

Automatic Infants’ Pain Assessment By Dynamic Facial Representation: Effects Of Profile View, Gestational Age, Gender, And Race, Ruicong Zhi, Ghada Z. D. Zamzmi, Dmitry Goldgof, Terri Ashmeade, Yu Sun

Computer Science and Engineering Faculty Publications

Infants’ early exposure to painful procedures can have negative short and long-term effects on cognitive, neurological, and brain development. However, infants cannot express their subjective pain experience, as they do not communicate in any language. Facial expression is the most specific pain indicator, which has been effectively employed for automatic pain recognition. In this paper, dynamic pain facial expression representation and fusion scheme for automatic pain assessment in infants is proposed by combining temporal appearance facial features and temporal geometric facial features. We investigate the effects of various factors that influence pain reactivity in infants, such as individual variables of …

Modeling And Visualization Of Uncertainty-Aware Geometries Using Multi-Variate Normal Distributions, Christina Gillman, Thomas Wischgoll, Bernd Hamann, James Ahrens

Computer Science and Engineering Faculty Publications

Many applications are dealing with geometric data that are affected by uncertainty. It is important to analyze, visualize, and understand the properties of uncertain geometry. We present a methodology to model uncertain geometry based on multi-variate normal distributions. In addition, we propose a visualization technique to represent a hull for uncertain geometry capturing a user-defined percentage of the underlying uncertain geometry. To show the effectiveness of our approach, we have modeled and visualized uncertain datasets from different applications.

Usability Assessment For Caregiver Behavior Analysis Using Gaming Technology, Alexandrea C. Oliver, Tanvi Banerjee, Jennifer Hughes, Noah L. Schroeder

Computer Science and Engineering Faculty Publications

The proposed research focuses on developing a mobile application for Android systems that will detect changes in behavior and activity patterns of those who are primary caregivers for dementia patients. This application will be used to detect fluctuation in the behavior and the task performance of the caregivers as a measure of caregiver stress. By detecting these changes in behavior, the goal is to analyze the effects of caregiving to evaluate caregiver burnout. A usability study was conducted for this application to find the optimal design factors and features that benefit the target user: the caregiver.

The purpose of this …

An Uncertainty-Aware Workflow For Keyhole Surgery Planning Using Hierarchical Image Semantics, Christina Gillmann, Robin G.C. Maack, Tobias Post, Thomas Wischgoll, Hans Hagen

Computer Science and Engineering Faculty Publications

Keyhole surgeries become increasingly important in clinical daily routine as they help minimizing the damage of a patient's healthy tissue. The planning of keyhole surgeries is based on medical imaging and an important factor that influences the surgeries' success. Due to the image reconstruction process, medical image data contains uncertainty that exacerbates the planning of a keyhole surgery. In this paper we present a visual workfiow that helps clinicians to examine and compare different surgery paths as well as visualizing the patients' affected tissue. The analysis is based on the concept of hierarchical image semantics, that segment the underlying image …

Trust In Visualization (And What It Has To Do With Theory), Thomas Wischgoll

Computer Science and Engineering Faculty Publications

There are different issues with trust involved when working with domain experts to visualize their data. There may be limitations with the data that require special precautions, such as sensitivity or security limitations. It may have taken a lot of effort to collect or create the data so that a certain level of trust is required for the domain expert to share the data. At the same time, the domain expert needs to be able to trust in the final visualization results. This presentation discusses these issues with trust and what requirements for a theoretical foundation this results in. Furthermore, …

Visual Analytics Of Cascaded Bottlenecksin Planar Flow Networks, Tobias Post, Thomas Wischgoll, Bernd Hamann, Hans Hagen

Computer Science and Engineering Faculty Publications

Finding bottlenecks and eliminating them to in-crease the overall flow of a network often appears in real world applications, such as production planning, factory layout, flowrelated physical approaches, and even cyber security. In many cases, several edges can form a bottleneck (cascaded bottlenecks). This work presents a visual analytics methodology to analyze these cascaded bottlenecks. The methodology consists of multiple steps: identification of bottlenecks, identification of potential improvements, communication of bottlenecks, interactive adaption of bottlenecks, and a feedback loop that allows users to adapt flow networks and their resulting bottlenecks until they are satisfied with the flow network configuration. To …

Comparing And Enhancing The Analytical Model For Exposure Of A Retail Facility Layout With Human Performance, Bradley R. Guthrie, Pratik Parikh, Tyler Whitlock, Madison Glines, Thomas Wischgoll, John Flach, Scott Watamaniuk

Computer Science and Engineering Faculty Publications

Recent research in retail facility layout has focused on developing analytical models to estimate visibility measures of novel rack layouts based on assumptions about a shopper’s field of view. However, because of the human element involved in the shopping experience, it is vital to compare these models relative to actual human performance. In this study, we evaluate the predictions of our previously developed analytical model (that estimates exposure of every location on a given rack layout assuming expected head movement) in a 3D Virtual Environment (VE). We conducted trials with 18 participants who were asked to find targets strategically placed …

Towards An Image-Based Indicator For Pad Classification And Localization, Christina Gillmann, Johh H. Matsuura, Hans Hagen, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

Peripheral Artery Disease (PAD) is an often occurring problem caused by narrowed veins. With this type of disease, mostly the legs receive an insufficient supply of blood to sustain their functions. This can result in an amputation of extremities or strokes. In order to quantify the risks, doctors onsult a classification table which is based on the pain response of a patient. This classification is subjective and does not indicate the exact origin of the PAD symptoms. Resulting from this, complications can occur unprompted. We present the first results for an image-based indicator assisting medical doctors in estimating the stage …

An Uncertainty-Aware Visual System For Image Pre-Processing, Christina Gillmann, Pablo Arbelaez, Jose Tiberio Hernandez, Hans Hagen, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

Due to image reconstruction process of all image capturing methods, image data is inherently affected by uncertainty. This is caused by the underlying image reconstruction model, that is not capable to map all physical properties in its entirety. In order to be aware of these effects, image uncertainty needs to be quantified and propagated along the entire image processing pipeline. In classical image processing methodologies, pre-processing algorithms do not consider this information. Therefore, this paper presents an uncertainty-aware image pre-processing paradigm, that is aware of the input image’s uncertainty and propagates it trough the entire pipeline. To accomplish this, we …

Caregiver Assessment Using Smart Gaming Technology: A Preliminary Approach, Garrett Goodman, Tanvi Banerjee, William Romine, Cogan Shimizu, Jennifer Hughes

Computer Science and Engineering Faculty Publications

As pre-diagnostic technologies are becoming increasingly accessible, using them to improve the quality of care available to dementia patients and their caregivers is of increasing interest. Specifically, we aim to develop a tool for non- invasively assessing task performance in a simple gaming application. To address this, we have developed Caregiver Assessment using Smart Technology (CAST), a mobile application that personalizes a traditional word scramble game. Its core functionality uses a Fuzzy Inference System (FIS) optimized via a Genetic Algorithm (GA) to provide customized performance measures for each user of the system. With CAST, we match the relative level of …

Understanding Sleep In Pediatric Patients With Sickle Cell Disease Admitted For Vaso-Occlusive Pain Crisis Through Objective Data, Kalindi Narine, Fan Yang, Tanvi Banerjee, Jude Jonassaint, Nirmish Shah

Computer Science and Engineering Faculty Publications

Sickle cell disease (SCD) is an inherited red cell disorder that leads to sickling of red blood cells, anemia and vaso-occlusion. The most common reason for hospitalization and morbidity in children is pain due to vaso-occlusive crisis (VOC). Importantly, poor sleep quality can lead to increased pain the subsequent day and nocturnal pain leads to reduced deep sleep, both which can then modify pain sensitivity. Studies using sleep diaries have shown this cyclical relationship between sleep and pain. Frequent occurrences of restless sleep are therefore believed to contribute to an increased severity and intensity of pain episodes. There is very …

Teaching Image Processing And Visualization Principles To Medicine Students, Christina Gillmann, Thomas Wischgoll, Jose T. Hernandez, Hans Hagen

Computer Science and Engineering Faculty Publications

Although image processing becomes increasingly important in most applications such as medicine, image processing and visualization is usually not a part of the medical education and therefore not widely spread in clinical daily routine. Contrary to students from computer science, medical students are usually not familiar to computational models or algorithms and require a different view of the algorithms instead of knowing each computational detail. To solve this problem this paper presents the concept of a lecture that aims to impart image processing and visualization principals for students in medicine in order to pioneer a higher acceptance and propagation of …

What Does Explainable Ai Really Mean? A New Conceptualization Of Perspectives, Derek Doran, Sarah Schulz, Tarek R. Besold

Computer Science and Engineering Faculty Publications

We characterize three notions of explainable AI that cut across research fields: opaque systems that offer no insight into its algo- rithmic mechanisms; interpretable systems where users can mathemat- ically analyze its algorithmic mechanisms; and comprehensible systems that emit symbols enabling user-driven explanations of how a conclusion is reached. The paper is motivated by a corpus analysis of NIPS, ACL, COGSCI, and ICCV/ECCV paper titles showing differences in how work on explainable AI is positioned in various fields. We close by introducing a fourth notion: truly explainable systems, where automated reasoning is central to output crafted explanations without requiring human …

Explaining Trained Neural Networks With Semantic Web Technologies: First Steps, Md Kamruzzaman Sarker, Ning Xie, Derek Doran, Michael L. Raymer, Pascal Hitzler

Computer Science and Engineering Faculty Publications

The ever increasing prevalence of publicly available struc-tured data on the World Wide Web enables new applications in a varietyof domains. In this paper, we provide a conceptual approach that lever-ages such data in order to explain the input-output behavior of trainedartificial neural networks. We apply existing Semantic Web technologiesin order to provide an experimental proof of concept.

Intuitive Error Space Exploration Of Medical Image Data In Clinical Daily Routine, Christina Gillmann, Pablo Arbeláez, José Tiberio Hernández Peñaloza, Hans Hagen, Thomas Wischgoll

Computer Science and Engineering Faculty Publications

Medical image data can be affected by several image errors. These errors can lead to uncertain or wrong diagnosis in clinical daily routine. A large variety of image error metrics are available that target different aspects of image quality forming a highdimensional error space, which cannot be reviewed trivially. To solve this problem, this paper presents a novel error space exploration technique that is suitable for clinical daily routine. Therefore, the clinical workflow for reviewing medical data is extended by error space cluster information, that can be explored by user-defined selections. The presented tool was applied to two real-world datasets …

Panel: Teaching To Increase Diversity And Equity In Stem, Helen H. Hu, Douglas Blank, Albert Chan, Travis E. Doom

Computer Science and Engineering Faculty Publications

TIDES (Teaching to Increase Diversity and Equity in STEM) is a three-year initiative to transform colleges and universities by changing what STEM faculty, especially CS instructors, are doing in the classroom to encourage the success of their students, particularly those that have been traditionally underrepresented in computer science.Each of the twenty projects selected proposed new inter-disciplinary curricula and adopted culturally sensitive pedagogies, with an eye towards departmental and institutional change. The four panelists will each speak about their TIDES projects, which all involved educating faculty about cultural competency. Three of the panelists infused introductory CS courses with applications from other …

Semi-Supervised Approach To Monitoring Clinical Depressive Symptoms In Social Media, Amir Hossein Yazdavar, Hussein S. Al-Olimat, Monireh Ebrahimi, Goonmeet Bajaj, Tanvi Banerjee, Krishnaprasad Thirunarayan, Jyotishman Pathak, Amit Sheth

Computer Science and Engineering Faculty Publications

With the rise of social media, millions of people are routinely expressing their moods, feelings, and daily struggles with mental health issues on social media platforms like Twitter. Unlike traditional observational cohort studies conducted through questionnaires and self-reported surveys, we explore the reliable detection of clinical depression from tweets obtained unobtrusively. Based on the analysis of tweets crawled from users with self-reported depressive symptoms in their Twitter profiles, we demonstrate the potential for detecting clinical depression symptoms which emulate the PHQ-9 questionnaire clinicians use today. Our study uses a semi-supervised statistical model to evaluate how the duration of these symptoms …

An Industrial Vision System To Analyze The Wear Of Cutting Tools, Christina Gillmann, Tobias Post, Benjamin Kirsch, Thomas Wischgoll, Jörg Hartig, Bernd Hamann, Hans Hagen, Jan C. Aurich

Computer Science and Engineering Faculty Publications

The wear behavior of cutting tools directly affects the quality of the machined part. The measurement and evaluation of wear is a time consuming process and is subjective. Therefore, an image-based wear measurement that can be computed automatically based on given image series of cutting tools and an objective way to review the resulting wear is presented in this paper. The presented method follows the industrial vision system pipeline where images of cutting tools are used as input which are then transformed through suitable image processing methods to prepare them for the computation of a novel image based wear measurement. …

A High-Dimensional Data Quality Metric Using Pareto Optimality, Tobias Post, Thomas Wischgoll, Bernd Hamann, Hans Hagen

Computer Science and Engineering Faculty Publications

The representation of data quality within established high-dimensional data visualization techniques such as scatterplots and parallel coordinates is still an open problem. This work offers a scale-invariant measure based on Pareto optimality that is able to indicate the quality of data points with respect to the Pareto front. In cases where datasets contain noise or parameters that cannot easily be expressed or evaluated mathematically, the presented measure provides a visual encoding of the environment of a Pareto front to enable an enhanced visual inspection.

Energy-Efficient Multicast Transmission For Underlay Device-To-Device Communications: A Social-Aware Perspective, Fan Jiang, Yao Liu, Chenbi Li, Changyin Sun

Computer Science and Engineering Faculty Publications

In this paper, by utilizing the social relationships among mobile users, we present a framework of energy-efficient cluster formation and resource allocation for multicast D2D transmission. In particular, we first deal with D2D multicast cluster/group formation strategy from both physical distance and social trust level. Then we aim to maximize the overall energy-efficiency of D2D multicast groups through resource allocation and power control scheme, which considers the quality-of-service (QoS) requirements of both cellular user equipment and D2D groups. A heuristic algorithm is proposed to solve above energy-efficiency problem with less complexity. After that, considering the limited battery capacity of mobile …

Predicting Malignant Nodules From Screening Ct Scans, Samuel Hawkins, Hua Wang, Ying Liu, Alberto Garcia, Olya Stringfield, Henry Krewer, Qiang Li, Dmitry Cherezov, Matthew Schabath, Lawrence O. Hall, Robert J. Gillies

Computer Science and Engineering Faculty Publications

Objectives

The aim of this study was to determine whether quantitative analyses (“radiomics”) of low-dose computed tomography lung cancer screening images at baseline can predict subsequent emergence of cancer.

Methods

Public data from the National Lung Screening Trial (ACRIN 6684) were assembled into two cohorts of 104 and 92 patients with screen-detected lung cancer and then matched with cohorts of 208 and 196 screening subjects with benign pulmonary nodules. Image features were extracted from each nodule and used to predict the subsequent emergence of cancer.

Results

The best models used 23 stable features in a random forests classifier and could …

Deep Feature Transfer Learning In Combination With Traditional Features Predicts Survival Among Patients With Lung Adenocarcinoma, Rahul Paul, Samuel H. Hawkings, Matthew B. Schabath, Robert J. Gilies, Lawrence O. Hall, Dmitry Goldgof

Computer Science and Engineering Faculty Publications

Lung cancer is the most common cause of cancer-related deaths in the USA. It can be detected and diagnosed using computed tomography images. For an automated classifier, identifying predictive features from medical images is a key concern. Deep feature extraction using pretrained convolutional neural networks (CNNs) has recently been successfully applied in some image domains. Here, we applied a pretrained CNN to extract deep features from 40 computed tomography images, with contrast, of non-small cell adenocarcinoma lung cancer, and combined deep features with traditional image features and trained classifiers to predict short- and long-term survivors. We experimented with several pretrained …