Computer Engineering Commons^™

Ensemble Lung Segmentation System Using Deep Neural Networks, Redha A. Ali, Russell C. Hardie, Hussin K. Ragb

Electrical and Computer Engineering Faculty Publications

Lung segmentation is a significant step in developing computer-aided diagnosis (CAD) using Chest Radiographs (CRs). CRs are used for diagnosis of the 2019 novel coronavirus disease (COVID-19), lung cancer, tuberculosis, and pneumonia. Hence, developing a Computer-Aided Detection (CAD) system would provide a second opinion to help radiologists in the reading process, increase objectivity, and reduce the workload. In this paper, we present the implementation of our ensemble deep learning model for lung segmentation. This model is based on the original DeepLabV3+, which is the extended model of DeepLabV3. Our model utilizes various architectures as a backbone of DeepLabV3+, such as …

Lm-Based Word Embeddings Improve Biomedical Named Entity Recognition: A Detailed Analysis, Liliya Akhtyamova, John Cardiff

Conference Papers

Recent studies have shown that contextualized word embeddings outperform other types of embeddings on a variety of tasks. However, there is little research done to evaluate their effectiveness in the biomedical domain under multi-task settings. We derive the contextualized word embeddings from the Flair framework and apply them to the task of biomedical NER on 5 benchmark datasets, yielding major improvements over the baseline and achieving competitive results over the current best systems. We analyze the sources of these improvements, reporting model performances over different combinations of word embeddings, and fine-tuning and casing modes.

Recent Developments In The General Atomic And Molecular Electronic Structure System, Guiseppe M.J. Barca, Colleen Bertoni, Laura Carrington, Dipayan Datta, Nuwan De Silva, J. Emillano Deustua, Dmitri G. Fedorov, Jeffrey R. Cour, Anastasia O. Gunina, Emilie Guidez, Taylor Harville, Stephan Irle, Joe Ivanic, Karol Kowalski, Sarom S. Leang, Wei Li, Jesse J. Lutz, Ilias Magoulas, Joani Mato, Vladimir Mironov, Hiroya Nakata, Buu Q. Pham, Piotr Piecuch, David Poole, Spencer R. Pruitt, Alistair P. Rendell, Luke B. Roskop, Klaus Ruedenberg, Tosaporn Sattasathuchana, Michael W. Schmidt, Jun Shen, Lyudmila Slipchenko, Masha Sosonkina, Vaibhav Sundriyal, Ananta Tiwari, Jorge L. Galvez Vallejo, Bryce Westheimer, Marta Włoch, Peng Xu, Federico Zahariev, Mark S. Gordon

Computational Modeling & Simulation Engineering Faculty Publications

A discussion of many of the recently implemented features of GAMESS (General Atomic and Molecular Electronic Structure System) and LibCChem (the C++ CPU/GPU library associated with GAMESS) is presented. These features include fragmentation methods such as the fragment molecular orbital, effective fragment potential and effective fragment molecular orbital methods, hybrid MPI/OpenMP approaches to Hartree-Fock, and resolution of the identity second order perturbation theory. Many new coupled cluster theory methods have been implemented in GAMESS, as have multiple levels of density functional/tight binding theory. The role of accelerators, especially graphical processing units, is discussed in the context of the new features …

Deformable Multisurface Segmentation Of The Spine For Orthopedic Surgery Planning And Simulation, Rabia Haq, Jérôme Schmid, Roderick Borgie, Joshua Cates, Michel Audette

Computational Modeling & Simulation Engineering Faculty Publications

Purpose: We describe a shape-aware multisurface simplex deformable model for the segmentation of healthy as well as pathological lumbar spine in medical image data.

Approach: This model provides an accurate and robust segmentation scheme for the identification of intervertebral disc pathologies to enable the minimally supervised planning and patient-specific simulation of spine surgery, in a manner that combines multisurface and shape statistics-based variants of the deformable simplex model. Statistical shape variation within the dataset has been captured by application of principal component analysis and incorporated during the segmentation process to refine results. In the case where shape statistics hinder detection …

Implementing Asynchronous Linear Solvers Using Non-Uniform Distributions, Erik Jensen, Evan C. Coleman, Masha Sosonkina

Computational Modeling & Simulation Engineering Faculty Publications

Asynchronous iterative methods present a mechanism to improve the performance of algorithms for highly parallel computational platforms by removing the overhead associated with synchronization among computing elements. This paper considers a class of asynchronous iterative linear system solvers that employ randomization to determine the component update orders, specifically focusing on the effects of drawing the order from non-uniform distributions. Results from shared-memory experiments with a two-dimensional finite-difference discrete Laplacian problem show that using distributions favoring the selection of components with a larger contribution to the residual may lead to faster convergence than selecting uniformly. Multiple implementations of the randomized asynchronous …

An Evaluation Of Text Representation Techniques For Fake News Detection Using: Tf-Idf, Word Embeddings, Sentence Embeddings With Linear Support Vector Machine., Sangita Sriram

Dissertations

In a world where anybody can share their views, opinions and make it sound like these are facts about the current situation of the world, Fake News poses a huge threat especially to the reputation of people with high stature and to organizations. In the political world, this could lead to opposition parties making use of this opportunity to gain popularity in their elections. In the medical world, a fake scandalous message about a medicine giving side effects, hospital treatment gone wrong or even a false message against a practicing doctor could become a big menace to everyone involved in …

Drug Reviews: Cross-Condition And Cross-Source Analysis By Review Quantification Using Regional Cnn-Lstm Models, Ajith Mathew Thoomkuzhy

Dissertations

Pharmaceutical drugs are usually rated by customers or patients (i.e. in a scale from 1 to 10). Often, they also give reviews or comments on the drug and its side effects. It is desirable to quantify the reviews to help analyze drug favorability in the market, in the absence of ratings. Since these reviews are in the form of text, we should use lexical methods for the analysis. The intent of this study was two-fold: First, to understand how better the efficiency will be if CNN-LSTM models are used to predict ratings or sentiment from reviews. These models are known …

Adapting Microservices In The Cloud With Faas, Mateusz Pietraszewski

Dissertations

This project involves benchmarking, microservices and Function-as-a-service (FaaS) across the dimensions of performance and cost. In order to do a comparison this paper proposes a benchmark framework.

Lightgwas: A Novel Genome-Wide Association Study Procedure, Bruno Ambrozio

Dissertations

This dissertation proposes LightGWAS, a novel machine learning procedure for genome-wide association study (GWAS) based on LightGBM and k-fold cross-validation. The conducted literature review identified that the currently available GWAS implementations rely on massive manual quality control steps to address statistical issues, such as controlling for false-positive inflation and power reduction. It also showed they demand a specific GWAS method for each type of genomic dataset morphology, which consequently increases the human dependency and open margins for misleadings. LightGWAS is a potential single, resilient, autonomous and scalable solution to address such concerns. Through this research, LightGWAS was contrasted against the …

Physical Randomness Can Help In Computations, Olga Kosheleva, Vladik Kreinovich

Departmental Technical Reports (CS)

Can we use some so-far-unused physical phenomena to compute something that usual computers cannot? Researchers have been proposing many schemes that may lead to such computations. These schemes use different physical phenomena ranging from quantum-related to gravity-related to using hypothetical time machines. In this paper, we show that, in principle, there is no need to look into state-of-the-art physics to develop such a scheme: computability beyond the usual computations naturally appears if we consider such a basic notion as randomness.

Nyku: A Social Robot For Children With Autism Spectrum Disorders, Dan Stephan Stoianovici

Electronic Theses and Dissertations

The continued growth of Autism Spectrum Disorders (ASD) around the world has spurred a growth in new therapeutic methods to increase the positive outcomes of an ASD diagnosis. It has been agreed that the early detection and intervention of ASD disorders leads to greatly increased positive outcomes for individuals living with the disorders. Among these new therapeutic methods, Robot-Assisted Therapy (RAT) has become a hot area of study. Recent works have shown that high functioning ASD children have an affinity for interacting with robots versus humans. It is proposed that this is due to a less complex set of communication …

Microgrid-Enabled Reactive Power Support To Enhance Grid Economics, Sarhan Hasan

Electronic Theses and Dissertations

Reactive power plays an essential role in voltage control and stability in electric power systems. Various Volt/VAR techniques are utilized in electric power systems to maintain the voltage profile within defined acceptable limits and accordingly provide reliability and stability. Reactive power has been commonly generated through large-scale synchronous generators or distributed capacitor banks to provide proper transmission and distribution level system management, however, reactive power can be further used as an effective means to reduce total system operation cost. In this dissertation, an optimal reactive power model is proposed to determine the optimal nodal reactive powers that result in the …

Deep Siamese Neural Networks For Facial Expression Recognition In The Wild, Wassan Hayale

Electronic Theses and Dissertations

The variation of facial images in the wild conditions due to head pose, face illumination, and occlusion can significantly affect the Facial Expression Recognition (FER) performance. Moreover, between subject variation introduced by age, gender, ethnic backgrounds, and identity can also influence the FER performance. This Ph.D. dissertation presents a novel algorithm for end-to-end facial expression recognition, valence and arousal estimation, and visual object matching based on deep Siamese Neural Networks to handle the extreme variation that exists in a facial dataset. In our main Siamese Neural Networks for facial expression recognition, the first network represents the classification framework, where we …

Facial Action Unit Detection With Deep Convolutional Neural Networks, Siddhesh Padwal

Electronic Theses and Dissertations

The facial features are the most important tool to understand an individual's state of mind. Automated recognition of facial expressions and particularly Facial Action Units defined by Facial Action Coding System (FACS) is challenging research problem in the field of computer vision and machine learning. Researchers are working on deep learning algorithms to improve state of the art in the area. Automated recognition of facial action units has man applications ranging from developmental psychology to human robot interface design where companies are using this technology to improve their consumer devices (like unlocking phone) and for entertainment like FaceApp. Recent studies …

Acue Certificate In Effective College Instruction Curated Reflections, Fadi Muheidat

Q2S Enhancing Pedagogy

The ACUE course is full of useful, best practice, experts’ opinions, and elaborated research on effective teaching. I feel I need another round of reading and watching the videos and classroom scenarios and go through the state-of-the-art techniques and work samples. My peers who participated in this course enriched the discussion and their own experiences best practices. Samples of reflections and implementation strategies listed below.

Developing An Inclusive K-12 Outreach Model, Karen Nolan, Roisin Faherty, Keith Quille, Brett A. Becker, Susan Bergin

Conference Papers

This paper outlines the longitudinal development of a K-12 outreachmodel, to promote Computer Science in Ireland. Over a three-yearperiod, it has been piloted to just under 9700 K-12 students fromalmost every county in Ireland. The model consists of a two-hourcamp that introduces students to a range of Computer Sciencetopics: addressing computing perceptions, introduction to codingand exploration of computational thinking. The model incorporateson-site school delivery and is available at no cost to any interestedschool across Ireland. The pilot study so far collected over 3400surveys (pre- and post-outreach delivery).Schools from all over Ireland self-selected to participate, includ-ing male only, female only and …

Named Entity Recognition In Spanish Biomedical Literature: Short Review And Bert Model, Liliya Akhtyamova

Conference Papers

Named Entity Recognition (NER) is the rst step for knowledge acquisition when we deal with an unknown corpus of texts. Having received these entities, we have an opportunity to form parameters space and to solve problems of text mining as concept normalization, speech recognition, etc. The recent advances in NER are related to the technology of word embeddings, which transforms text to the form being effective for Deep Learning. In the paper, we show how NER detects pharmacological substances, compounds, and proteins in the dataset obtained from the Spanish Clinical Case Corpus (SPACCC). To achieve this goal, we use contextualized …

Annual Report 2019-2020, Depaul University College Of Computing And Digital Media

CDM Annual Reports

LETTER FROM THE DEAN

As I write this letter wrapping up the 2019-20 academic year, we remain in a global pandemic that has profoundly altered our lives. While many things have changed, some stayed the same: our CDM community worked hard, showed up for one another, and continued to advance their respective fields. A year that began like many others changed swiftly on March 11th when the University announced that spring classes would run remotely. By March 28th, the first day of spring quarter, we had moved 500 CDM courses online thanks to the diligent work of our faculty, staff, …

Timcc: On Data Freshness In Privacy-Preserving Incentive Mechanism Design For Continuous Crowdsensing Using Reverse Auction, Xiaoqiang Ma, Weiwei Deng, Feng Wang, Menglan Hu, Fei Chen, Mohammad Mehedi Hassan

Faculty and Student Publications

© 2013 IEEE. As an emerging paradigm that leverages the wisdom and efforts of the crowd, mobile crowdsensing has shown its great potential to collect distributed data. The crowd may incur such costs and risks as energy consumption, memory consumption, and privacy leakage when performing various tasks, so they may not be willing to participate in crowdsensing tasks unless they are well-paid. Hence, a proper privacy-preserving incentive mechanism is of great significance to motivate users to join, which has attracted a lot of research efforts. Most of the existing works regard tasks as one-shot tasks, which may not work very …

Promoting A Growth Mindset In Cs1: Does One Size Fit All? A Pilot Study, Keith Quille, Susan Bergin

Conference Papers

This paper describes a pilot intervention conducted in CS1, in the
academic year of 2016-2017. The intervention was based on the
work of Dweck, promoting a growth Mindset in an effort to in-
crease performance in introductory programming. The study also
examined data from a previous year (as a control group) to compare
and contrast the results. Multiple factors related to programming
performance were recorded with the control and treatment group,
which were measured at multiple intervals throughout the course,
to monitor changes as the pilot intervention was implemented.
This study found a significant increase in programming perfor-
mance when …