Multi-Semantic-Stage Neural Networks For Robust And Interpretable Deep Learning, Christopher J. Menart

Browse all Theses and Dissertations

Deep neural networks have great representational power. However, most deep neural nets today optimize directly for performance on a single task defined only by labeled training data. This excludes potential sources of knowledge and ways of learning which could improve their performance, and address challenges, such as explainability, which are pressing to the field. We propose a framework for neural network architecture which generalizes it to a graph of many semantically-meaningful variables. We call it the Multi-Semantic-Stage Neural Network (MSSNN). An MSSNN models its domain as a web of conditional probabilities, i.e. a collection of inter-related tasks which can learn …

Robust Machine Learning Based Frameworks For Localization And Mmwave Beam Prediction In Massive Mimo Systems, Katarina Vuckovic

Graduate Thesis and Dissertation post-2024

The primary focus of this doctoral dissertation is the investigation of robust massive Multiple Input Multiple Output (MIMO) communication systems. The study centers on two closely related issues: i) single-site user localization, and ii) millimeter Wave (mmWave) beam prediction. The motivation is to develop a system that simultaneously offers users reliable communication and precise localization using a single infrastructure. This dissertation specifically addresses three key challenges in data-driven solution: i) the limited availability of position-labeled training datasets, ii) robustness in time-varying dynamic environments, and iii) the generalization of models to unseen scenarios. The study introduces three localization and two beam …

Effective Land Use Classification Through Hybrid Transformer Using Remote Sensing Imagery, Muhammad Zia Ur Rehman, Syed Mohammed Shamsul Islam, Anwaar Ulhaq, David Blake, Naeem Janjua

Research outputs 2022 to 2026

Recent advances in deep learning for hyperspectral image (HSI) classification have shown exceptional performance in resource management and environmental planning through land use classification. Despite these successes, challenges continue to persist in land use classification due to the complex topology of natural and man-made structures. The uneven distribution of land cover introduces spectral-spatial variability, causing inter- and intra-class similarity. To address this issue, this study adopts a hybrid approach that combines convolutional neural networks (CNNs) and a transformer model. The technique comprises three key components: a spectral-spatial convolutional module (SSCM), a spatial attention module (SAM), and a transformer module. Each …

Assigning Medical Professionals: Chatgpt's Contributions To Medical Education And Health Prediction, Maad M. Mijwil, Mostafa Abotaleb, Guma Ali, Klodian Dhoska

Mesopotamian Journal of Artificial Intelligence in Healthcare

Artificial intelligence is increasingly present in many applications that help humans accomplish many tasks. It can support improved results, increased productivity, and high efficiency in providing services. Systems developers seek to integrate improved artificial intelligence models into the development of healthcare services and make a significant qualitative shift in the medical field. One recently invented tool is ChatGPT, which deserves extensive advertising space. This tool provides sample answers that help both healthcare workers and patients answer all their questions. It is an important tool in medical education, increasing knowledge in making decisions that can improve the performance of medical professionals. …

Deep Learning Model For Hand Movement Rehabilitation, Reem D. Ismail, Qabas A. Hameed, Mustafa Abdulfattah Habeeb, Yahya Layth Khaleel, Fatimah N. Ameen

Mesopotamian Journal of Computer Science

Electroencephalography (EEG) can control machines for human purposes, especially for disabled people doing rehabilitation exercises or regular tasks. Brain-computer interface (BCI) for Robotic hand uses deep learning to convert (EEG) brain activity into orders for robotic hand allowing users to move their hands right or left by the movement imagining. It could enable paralyzed individuals to perform basic hand movements and help in rehabilitation robots that help stroke patients regain hand function by offering guided exercises based on machine learning interpretations of their movements and intents. Artificial intelligence algorithms, particularly deep learning, classify and recognize patterns and intents implicit brainwaves …

Credit Card Fraud Detection Based On Deep Learning Models, El-Sayed M. El-Kenawy, Ahmed Mohamed Zaki, Wei Hong Lim, Abdelhameed Ibrahim, Marwa M. Eid, Ahmed Osman Osman, Ahmed M. Elshewey

Mesopotamian Journal of Computer Science

Credit card fraud detection (FD) protects consumers and financial institutions by identifying suspicious or unauthorized transactions. To improve security and reduce false positives, fraud detection systems can analyze transaction data patterns in real time using advanced machine learning (ML) and deep learning (DL). This paper exploits DL models to detects transactional data which includes anomalies through Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) to verify data and mitigate fraud. The models used precision, recall, F1-score, and AUC on a balanced shared 559856-record Kaggle repository dataset. The RNN model detected anomalies with 99.39% accuracy, 0.9939 …

Potato Leaf Disease Detection Approach Based On Transfer Learning With Spatial Attention, Rima Grati, Emna Abdallah, Khouloud Boukadi, Ahmed Smaoui

All Works

No abstract provided.

Media Haze Classification In Retinal Images Using Deep Learning, Jonathan O'Berry

UNF Graduate Theses and Dissertations

Media Haze (MH) is a condition that affects an individual’s quality of life by affecting their eyes. Current practice is to detect MH by manually examining retinal fundus (retinal) images. The analysis of images being used as the prevalent technique for identifying the MH condition strongly suggests that automation of this process may be possible. In recent years, machine learning, specifically computer vision, has allowed for the automation of tasks relating to image analysis. This ability to automate has also recently been shown in the medical field for some eye conditions and diseases. This thesis centers around the problem of …

A Transformer-Based Deep Learning Approach For Fairly Predicting Post-Liver Transplant Risk Factors, Can Li, Xiaoqian Jiang, Kai Zhang

Faculty, Staff and Student Publications

Liver transplantation is a life-saving procedure for patients with end-stage liver disease. There are two main challenges in liver transplant: finding the best matching patient for a donor and ensuring transplant equity among different subpopulations. The current MELD scoring system evaluates a patient's mortality risk if not receiving an organ within 90 days. However, the donor-patient matching should also consider post-transplant risk factors, such as cardiovascular disease, chronic rejection, etc., which are all common complications after transplant. Accurate prediction of these risk scores remains a significant challenge. In this study, we used predictive models to solve the above challenges. Specifically, …

Lesion Detection In Women Breast’S Dynamic Contrast-Enhanced Magnetic Resonance Imaging Using Deep Learning, Sudarshan Saikia, Tapas Si, Darpan Deb, Kangkana Bora, Saurav Mallik, Ujjwal Maulik, Zhongming Zhao

Faculty, Staff and Student Publications

Breast cancer is one of the most common cancers in women and the second foremost cause of cancer death in women after lung cancer. Recent technological advances in breast cancer treatment offer hope to millions of women in the world. Segmentation of the breast's Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI) is one of the necessary tasks in the diagnosis and detection of breast cancer. Currently, a popular deep learning model, U-Net is extensively used in biomedical image segmentation. This article aims to advance the state of the art and conduct a more in-depth analysis with a focus on the use …

Smart Applications And Resource Management In Internet Of Things, Zeinab Akhavan

Computer Science ETDs

Internet of Things (IoT) technologies are currently the principal solutions driving smart cities. These new technologies such as Cyber Physical Systems, 5G and data analytic have emerged to address various cities' infrastructure issues ranging from transportation and energy management to healthcare systems. An IoT setting primarily consists of a wide range of users and devices as a massive network interacting with different layers of the city infrastructure resulting in generating sheer volume of data to enable smart city services. The goal of smart city services is to create value for the entire ecosystem, whether this is health, education, transportation, energy, …

Utilizing Multitask Transfer Learning For Sonographic Rheumatoid Arthritis Synovitis Grading, Jordan Marie Claire Sanders

Doctoral Dissertations and Master's Theses

Classifying the four sonographic Rheumatoid Arthritis (RA) synovitis grades (Grade 0, Grade 1, Grade 2, and Grade 3) is a difficult problem due to the complexity of the relevant markers. Therefore, the current research proposes a Multitask Transfer Learning (MTL) framework for sonographic RA synovitis grading of Ultrasound (US) images in Brightness mode (B-Mode) and Power Doppler mode.

In the medical community, the lack of reliability of scoring these images has been an issue and reason for concern for doctors and other medical practitioners. The human/machine variability across the acquisition procedure of these US images creates an additional challenge that …

Neural Networks For Improved Signal Source Enumeration And Localization With Unsteered Antenna Arrays, John T. Rogers Ii

Theses and Dissertations

Direction of Arrival estimation using unsteered antenna arrays, unlike mechanically scanned or phased arrays, requires complex algorithms which perform poorly with small aperture arrays or without a large number of observations, or snapshots. In general, these algorithms compute a sample covriance matrix to obtain the direction of arrival and some require a prior estimate of the number of signal sources. Herein, artificial neural network architectures are proposed which demonstrate improved estimation of the number of signal sources, the true signal covariance matrix, and the direction of arrival. The proposed number of source estimation network demonstrates robust performance in the case …

Deep Learning Approaches For Chaotic Dynamics And High-Resolution Weather Simulations In The Us Midwest, Vlada Volyanskaya, Kabir Batra, Shubham Shrivastava

Discovery Undergraduate Interdisciplinary Research Internship

Weather prediction is indispensable across various sectors, from agriculture to disaster forecasting, deeply influencing daily life and work. Recent advancement of AI foundation models for weather and climate predictions makes it possible to perform a large number of predictions in reasonable time to support timesensitive policy- and decision-making. However, the uncertainty quantification, validation, and attribution of these models have not been well explored, and the lack of knowledge can eventually hinder the improvement of their prediction accuracy and precision. Our project is embarking on a two-fold approach leveraging deep learning techniques (LSTM and Transformer) architectures. Firstly, we model the Lorenz …

Context-Aware Temporal Embeddings For Text And Video Data, Ahnaf Farhan

Open Access Theses & Dissertations

Recent years have seen an exponential increase in unstructured data, primarily in the form of text, images, and videos. Extracting useful features and trends from large-scale unstructured datasets -- such as news outlets, scientific papers, and videos like security cameras or body cam recordings -- is faced with substantial challenges of volume, scalability, complexity, and semantic understanding. In analyzing trends, comprehending the temporal context is vital for uncovering patterns and narratives that are not apparent from a single video frame or text document. Despite its importance, many existing data mining and machine learning approaches overlook extracting evolutionary contextual features in …

Interpreting Codebert For Semantic Code Clone Detection, Shamsa Abid, Xuemeng Cai, Lingxiao Jiang

Research Collection School Of Computing and Information Systems

Accurate detection of semantic code clones has many applications in software engineering but is challenging because of lexical, syntactic, or structural dissimilarities in code. CodeBERT, a popular deep neural network based pre-trained code model, can detect code clones with a high accuracy. However, its performance on unseen data is reported to be lower. A challenge is to interpret CodeBERT's clone detection behavior and isolate the causes of mispredictions. In this paper, we evaluate CodeBERT and interpret its clone detection behavior on the SemanticCloneBench dataset focusing on Java and Python clone pairs. We introduce the use of a black-box model interpretation …

Correlating Deep Learning-Based Automated Reference Kidney Histomorphometry With Patient Demographics And Creatinine, Nicholas Lucarelli, Brandon Ginley, Jarcy Zee, Sayat Mimar, Anindya S. Paul, Sanjay Jain, Seung Seok Han, Luis Rodrigues, Tezcan Ozrazgat-Baslanti, Michelle L. Wong, Girish Nadkarni, William L. Clapp, Kuang-Yu Jen, Pinaki Sarder

2020-Current year OA Pubs

KEY POINTS: The authors leverage the unique benefits of panoptic segmentation to perform the largest ever quantitation of reference kidney morphometry. Kidney features vary with age and sex; and glomeruli size may intricately link to creatinine, defying prior notions.

BACKGROUND: Reference histomorphometric data of healthy human kidneys are largely lacking because of laborious quantitation requirements. Correlating histomorphometric features with clinical parameters through machine learning approaches can provide valuable information about natural population variance. To this end, we leveraged deep learning (DL), computational image analysis, and feature analysis to associate the relationship of histomorphometry with patient age, sex, serum creatinine (SCr), …

Assessing Chondrocyte Viability Of Articular Cartilage Using Label-Free Imaging Methods, Hongming Fan

All Dissertations

ABSTRACT

Articular cartilage covers the ends of bones, providing a protective layer that can endure forces during movement. Chondrocytes, the only cell type found in articular cartilage, produce collagen fibers and proteoglycans that maintain the extracellular matrix (ECM). Cartilage damage impairs chondrocytes and their ability to maintain tissue stability. There are various factors that can cause cell death, which can be broadly classified into two categories: programmed cell death (apoptosis) and non-programmed cell death (necrosis). The consequences of cell death can have significant impacts on the extracellular matrix (ECM). The ECM is a complex network of proteins and carbohydrates that …

Implementation Of Adas And Autonomy On Unlv Campus, Zillur Rahman

UNLV Theses, Dissertations, Professional Papers, and Capstones

The integration of Advanced Driving Assistance Systems (ADAS) and autonomous driving functionalities into contemporary vehicles has notably surged, driven by the remarkable progress in artificial intelligence (AI). These AI systems, capable of learning from real-world data, now exhibit the capability to perceive their surroundings via a suite of sensors, create optimal routes from source to destination, and execute vehicle control akin to a human driver.

Within the context of this thesis, we undertake a comprehensive exploration of three distinct yet interrelated ADAS and Autonomy projects. Our central objective is the implementation of autonomous driving(AD) technology at UNLV campus, culminating in …

Identifying The Optimal Deep Learning Architecture And Parameters For Automatic Beam Aperture Definition In 3d Radiotherapy, Skylar S Gay, Kelly D Kisling, Brian M Anderson, Lifei Zhang, Dong Joo Rhee, Callistus Nguyen, Tucker Netherton, Jinzhong Yang, Kristy Brock, Anuja Jhingran, Hannah Simonds, Ann Klopp, Beth M Beadle, Laurence E Court, Carlos E Cardenas

Faculty, Staff and Student Publications

PURPOSE: Two-dimensional radiotherapy is often used to treat cervical cancer in low- and middle-income countries, but treatment planning can be challenging and time-consuming. Neural networks offer the potential to greatly decrease planning time through automation, but the impact of the wide range of hyperparameters to be set during training on model accuracy has not been exhaustively investigated. In the current study, we evaluated the effect of several convolutional neural network architectures and hyperparameters on 2D radiotherapy treatment field delineation.

METHODS: Six commonly used deep learning architectures were trained to delineate four-field box apertures on digitally reconstructed radiographs for cervical cancer …

Deep Learning Approaches With Optimum Alpha For Energy Usage Forecasting, Aji Prasetya Wibawa, Agung Bella Putra Utama, Ade Kurnia Ganesh Akbari, Akhmad Fanny Fadhilla, Alfiansyah Putra Pertama Triono, Andien Khansa’A Iffat Paramarta, Faradini Usha Setyaputri, Leonel Hernandez

Knowledge Engineering and Data Science

Energy use is an essential aspect of many human activities, from individual to industrial scale. However, increasing global energy demand and the challenges posed by environmental change make understanding energy use patterns crucial. Accurate predictions of future energy consumption can greatly influence decision-making, supply-demand stability and energy efficiency. Energy use data often exhibits time-series patterns, which creates complexity in forecasting. To address this complexity, this research utilizes Deep Learning (DL), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-term Memory (LSTM), Bidirectional LSTM (Bi-LSTM), and Gated Recurrent Unit (GRU) models. The main objective is to improve the accuracy of …

Underwater Image Enhancement: A Pipeline For Underwater Computer Visions, Humberto Lebron Rivera

Theses and Dissertations

Ocean exploration has surged in popularity and significance in recent years, including diverse areas like maritime archeology, underwater resources, and submerged structure inspection. The activities mentioned above heavily depend on vision and imagery, a challenge in the unpredictable marine world. This thesis presents a conditional generative adversarial network model for image-to-image translation problems. We designed and trained the model with the end goal of enhancing underwater images. Five metrics were employed for validation to quantify our model’s resulting enhanced images. By doing so, we aim to establish a pipeline that can leverage aerial computer vision algorithms for marine applications.

Our …

Deep Learning Frameworks For Accelerated Magnetic Resonance Image Reconstruction Without Ground Truths, Ibsa Kumara Jalata

Graduate Theses and Dissertations

Magnetic Resonance Imaging (MRI) is typically a slow process because of its sequential data acquisition. To speed up this process, MR acquisition is often accelerated by undersampling k-space signals and solving an ill-posed problem through a constrained optimization process. Image reconstruction from under-sampled data is posed as an inverse problem in traditional model-based learning paradigms. While traditional methods use image priors as constraints, modern deep learning methods use supervised learning with ground truth images to learn image features and priors. However, in some cases, ground truth images are not available, making supervised learning impractical. Recent data-centric learning frameworks such as …

Deep Learning For Photovoltaic Characterization, Adrian Manuel De Luis Garcia

Graduate Theses and Dissertations

This thesis introduces a novel approach to Photovoltaic (PV) installation segmentation by proposing a new architecture to understand and identify PV modules from overhead imagery. Pivotal to this concept is the creation of a new Transformer-based network, S3Former, which focuses on small object characterization and modelling intra- and inter- object differentiation inside an image. Accurate mapping of PV installations is pivotal for understanding their adoption and guiding energy policy decisions. Drawing insights from current Deep Learning methodologies for image segmentation and building upon State-of-the-Art (SOTA) techniques in solar cell mapping, this work puts forth S3Former with the following enhancements: 1. …

Single-Cell Analysis Of Chromatin Accessibility In The Adult Mouse Brain, Songpeng Zu, Yang Eric Li, Kangli Wang, Ethan J Armand, Sainath Mamde, Maria Luisa Amaral, Yuelai Wang, Andre Chu, Yang Xie, Michael Miller, Jie Xu, Zhaoning Wang, Kai Zhang, Bojing Jia, Xiaomeng Hou, Lin Lin, Qian Yang, Seoyeon Lee, Bin Li, Samantha Kuan, Hanqing Liu, Jingtian Zhou, Antonio Pinto-Duarte, Jacinta Lucero, Julia Osteen, Michael Nunn, Kimberly A Smith, Bosiljka Tasic, Zizhen Yao, Hongkui Zeng, Zihan Wang, Jingbo Shang, M Margarita Behrens, Joseph R Ecker, Allen Wang, Sebastian Preissl, Bing Ren

Faculty, Staff and Student Publications

Recent advances in single-cell technologies have led to the discovery of thousands of brain cell types; however, our understanding of the gene regulatory programs in these cell types is far from complete1-4. Here we report a comprehensive atlas of candidate cis-regulatory DNA elements (cCREs) in the adult mouse brain, generated by analysing chromatin accessibility in 2.3 million individual brain cells from 117 anatomical dissections. The atlas includes approximately 1 million cCREs and their chromatin accessibility across 1,482 distinct brain cell populations, adding over 446,000 cCREs to the most recent such annotation in the mouse genome. The mouse brain cCREs are …

Domain Specific Feature Representation Learning For Diverse Temporal Data, Farhan Asif Chowdhury

Computer Science ETDs

Humans can leverage domain context to recognize novel patterns and categories based on limited known examples. In contrast, computational learning methods are not adept at exploiting context and require sufficient labeled examples to achieve similar accuracy. Many temporal data domain, for example, seismic signals and oil mining sensor data, requires domain expert annotation, which is both costly and time-consuming. The dependency on training data limits the applicability of machine learning algorithms for domains with limited labeled data. This dissertation aims to address this gap by developing temporal mining algorithms that exploit domain context to learn discriminative feature representation from limited …

Automation Of Crack Detection And Quantification In Civil Infrastructure Facilities Using Deep Learning Techniques, Luqman Ali

Thesis/ Dissertation Defenses

Cracks are the earliest signs of structural deterioration that reduce the lifespan and reliability of structures and can lead to severe damage. Assessment and monitoring of the facilities are required for lifetime maintenance and failure prediction. Structure condition information can be obtained manually, i.e., through subjective visual inspection and evaluation by human experts. Manual inspection techniques are labor-intensive, time-consuming, and inspector-dependent, i.e., vulnerable to the inspector’s perceptiveness. Automatic crack detection is crucial at the earliest stage to avoid further structure degradation and allow fast intervention. Deep Learning algorithms have become more popular in crack detection systems in recent years. However, …

Deciphering Trends And Tactics: Data-Driven Techniques For Forecasting Information Spread And Detecting Coordinated Campaigns In Social Media, Kin Wai Ng Lugo

USF Tampa Graduate Theses and Dissertations

The main objective of this dissertation is to develop models that predict and investigate the spread of information in social media over time. In this context, we consider topics of discussions as the information that spreads. Thus, we are interested in forecasting the number of messages per day in a future interval of time. We take a data-driven approach, in which we compare our results with real datasets from a multitude of socio-political contexts and from multiple social media platforms, specifically, Twitter and YouTube.

We identified a number of challenges related to forecasting social media time series per topic. First, …

Metaformer Baselines For Vision, Weihao Yu, Chenyang Si, Pan Zhou, Mi Luo, Yichen Zhou, Jiashi Feng, Shuicheng Yan, Xinchao Wang

Research Collection School Of Computing and Information Systems

Abstract—MetaFormer, the abstracted architecture of Transformer, has been found to play a significant role in achieving competitive performance. In this paper, we further explore the capacity of MetaFormer, again, by migrating our focus away from the token mixer design: we introduce several baseline models under MetaFormer using the most basic or common mixers, and demonstrate their gratifying performance. We summarize our observations as follows: (1) MetaFormer ensures solid lower bound of performance. By merely adopting identity mapping as the token mixer, the MetaFormer model, termed IdentityFormer, achieves >80% accuracy on ImageNet-1K. (2) MetaFormer works well with arbitrary token mixers. When …

Real-Time Object Tracking With Yolov5 And Recurrent Network On A Hardware Platform, Mohammed Abdulhakeem Alameri

Theses

The advancement of computer vision, particularly in the domain of object tracking, involves the integration of conventional feature-based approaches with contemporary deep learning methodologies. The initial phase of object detection plays a fundamental role in generating prospective objects for further tracking, and its level of success has a direct influence on the overall effectiveness of the tracking process. Many of the methods used to generate candidate objects start with object detection, and then object tracking algorithms are developed to link the object instances together to create the trajectories.
The primary difficulties in object tracking pertain to the establishment of reliable …