Physical Sciences and Mathematics Commons^™

Wakening Past Concepts Without Past Data: Class-Incremental Learning From Online Placebos, Yaoyao Liu, Yingying Li, Bernt Schiele, Qianru Sun

Research Collection School Of Computing and Information Systems

Not forgetting old class knowledge is a key challenge for class-incremental learning (CIL) when the model continuously adapts to new classes. A common technique to address this is knowledge distillation (KD), which penalizes prediction inconsistencies between old and new models. Such prediction is made with almost new class data, as old class data is extremely scarce due to the strict memory limitation in CIL. In this paper, we take a deep dive into KD losses and find that "using new class data for KD"not only hinders the model adaption (for learning new classes) but also results in low efficiency for …

The Metaverse, Religious Practice And Wellbeing: A Narrative Review, Justin Thomas, Mohammad Amin Kuhail, Fahad Albeyahi

All Works

The metaverse is touted as the next phase in the evolution of the Internet. This emerging digital ecosystem is widely conceptualized as a persistent matrix of interconnected multiuser, massively scaled online environments optimally experienced through immersive digital technologies such as virtual reality (VR). Much of the prognostication about the social implications of the metaverse center on secular activities. For example, retail, entertainment (gaming/concerts), and social networking. Little attention has been given to how the metaverse might impact religion. This narrative review explores contemporary research into online religious practice and the use of immersive digital technologies for religious purposes. This focus …

Advancing The Understanding Of Clinical Sepsis Using Gene Expression–Driven Machine Learning To Improve Patient Outcomes, Asrar Rashid, Feras Al-Obeidat, Wael Hafez, Govind Benakatti, Rayaz A. Malik, Christos Koutentis, Javed Sharief, Joe Brierley, Nasir Quraishi, Zainab A. Malik, Arif Anwary, Hoda Alkhzaimi, Syed Ahmed Zaki, Praveen Khilnani, Raziya Kadwa, Rajesh Phatak, Maike Schumacher, M. Guftar Shaikh, Ahmed Al-Dubai, Amir Hussain

All Works

Sepsis remains a major challenge that necessitates improved approaches to enhance patient outcomes. This study explored the potential of machine learning (ML) techniques to bridge the gap between clinical data and gene expression information to better predict and understand sepsis. We discuss the application of ML algorithms, including neural networks, deep learning, and ensemble methods, to address key evidence gaps and overcome the challenges in sepsis research. The lack of a clear definition of sepsis is highlighted as a major hurdle, but ML models offer a workaround by focusing on endpoint prediction. We emphasize the significance of gene transcript information …

The Deep Bsde Method, Daniel Kovach

Masters Theses

"The curse of dimensionality is the non-linear growth in computing time as the dimension of a problem increases. Using the Deep Backwards Stochastic Differential Equation (Deep BSDE) method developed in [HJE18], I approximate the solution at an initial time to a one-dimensional diffusion equation. Although we only approximate a one-dimensional equation, this method extends well to higher dimensions because it overcomes the curse of dimensionality by evaluating the given partial differential equation along "random characteristics''. In addition to the implementation, I also present most of the mathematical theory needed to understand this method"-- Abstract, p. iii

Radiofrequency Interference Detection Using Lstmand Statistical Analysis Discriminator, Luke Smith

Masters Theses

"Wireless devices are becoming increasingly pervasive across all aspects of society. Examples of such devices include radios, routers, mobile phones, tablets, and more. As the number of radio frequency (RF) devices continues to rise, so does the amount of interference and noise increase. This is why an efficient approach to interference detection is explored. Most research within this area has been done strictly within the frequency domain as viewing a signal within this domain provides many insights into what makes the signal. This has, however, led to the time domain being underutilized for this area of research.

To explore the …

Branching Fractional Brownian Motion, Reece Beattie-Hauser

Masters Theses

"Fractional Brownian Motion (FBM) is a Gaussian process whose increments are correlated over long times. FBM is an example of anomalous diffusion, and recently it has been used to model the distribution of serotonergic fibers in the brain [1, 2]. To better represent these fibers, branching FBM (bFBM), where FBM trajectories may randomly split into two, is introduced. One-dimensional bFBM is studied in both sub diffusive and super diffusive regimes, examining three potential behaviors of the correlations (memory) in a branching event: both trajectories retain the memory of previous steps, only one keeps the memory, and neither keeps the memory. …

The Exponential Function In Discrete Fractional Calculus Under The Delta Operator, Brayton James Link

Masters Theses

"Previously the exponential problem in discrete fractional calculus under the nabla operator was solved with the discrete Mittag--Leffler function. We now show the solution to the exponential problem in discrete fractional calculus under the delta operator, providing multiple derivations of the solution with recursion and Laplace transforms. We also share some computational and numerical results of experiments with different orders of difference to display the nature of the solution" -- Abstract, p. iii

Comparative Study Of Crypto Volatility And Price Forecasting Using A Mixture Of Time Series And Machine Learning Models, Abhishek Kafle

Masters Theses

"Forecasting financial product volatility and price is crucial for informed decision-making in investment and risk management. The models considered include GARCH, LSTM, GRU, BiLSTM, and hybrid models that incorporate various combinations of these models. We present a comparative analysis of forecasting volatility and price using the aforementioned models.

We also introduce a user-friendly dashboard for model training and evaluation, enabling users to upload datasets and customize model parameters. The dashboard allows users to select the type of model, specify the dataset range for training, determine the number of epochs, adjust the number of layers for deep …

Ontogenetically Driven Changes In The Endocranial Anatomy Of Maiasaura Peeblesorum And Its Implications For Sensory Functions, Emma Puetz

Masters Theses

"Hadrosaurids exhibited extreme morphological diversity and behavioral characteristics that can be compared with other, closely related members of Dinosauria – particularly through the study of neuroanatomy. X-ray CT generated endocrania of the Late Cretaceous saurolophine hadrosaur Maiasaura peeblesorum, known as the “good mother” dinosaur, offers a unique opportunity to compare complex ecological behaviors between non-avian dinosaurs and their modern archosaur relatives. The 3D reconstruction software ORS Dragonfly provided linear and volumetric measurements taken from the digital cranial endocast and endosseous labyrinths of three ontogenetically varied M. peeblesorum specimens from the Two Medicine Formation in Montana that were used to …

A Gat-Bilstma Model For Weather-Aware Prediction Of Traffic Speed, Bikis Muhammed

Masters Theses

This thesis presented a method for incorporating the effect of weather conditions in the prediction of the average speed of vehicular traffic for each segment of a road network. The proposed approach utilized two different deep learning methods: graph attention networks and bidirectional long short-term memory with attention layers. The accuracy of predictions is increased by considering the real-world driving distance between road segments, in contrast to the Haversine distance used in several existing prediction methods. Categorization of input data as weekend or weekday further increased the prediction accuracy. The proposed approach was validated using two data sets published by …

A Reputation-Based Aodv Protocol For Blackhole And Malfunction Nodes Detection And Avoidance, Qussai M. Yaseen, Monther Aldwairi, Ahmad Manasrah

All Works

Enhancing the security of Wireless Sensor Networks (WSNs) improves the usability of their applications. Therefore, finding solutions to various attacks, such as the blackhole attack, is crucial for the success of WSN applications. This paper proposes an enhanced version of the AODV (Ad Hoc On-Demand Distance Vector) protocol capable of detecting blackholes and malfunctioning benign nodes in WSNs, thereby avoiding them when delivering packets. The proposed version employs a network-based reputation system to select the best and most secure path to a destination. To achieve this goal, the proposed version utilizes the Watchdogs/Pathrater mechanisms in AODV to gather and broadcast …

Digraph Enabled Digital Twin And Label-Encoding Machine Learning For Scada Network's Cyber Attack Analysis In Industry 5.0, Nabeel Al-Qirim, Anoud Bani-Hani, Munir Majdalawieh, Hussam Al Hamadi, Mohammad Kamrul Hasan

All Works

False-Data Injection Attack (FDIA), Remote-Tripping Command Injection (RTCI), and System Reconfiguration Attack (SRA) on SCADA (Supervisory Control and Data Acquisition) networks impact industry 5.0 enabled smart grid components such as intelligent-electronic-device (IED), circuit-breaker, network-switch, and power transmission lines. Since the SCADA-network-based cyber-attacking flow is not in digital-twin form, it is impossible to simulate the effects of the attack. Furthermore, the string nature of these affected components' data makes it challenging to incorporate into machine-learning-enabled intelligence (CTI) processes. To visualize the attacking flow of FDIA, RTCI, and SRA cyber-attacks on SCADA networks, this paper presents a novel "Digital Twin and Machine …

Fraud Detection In Medical Insurance Claims Using Majority Voting Of Multiple Unsupervised Algorithms, Mohamed Ahmed Abo El-Enen, Dina Tbaishat, Ahmed T. Sahlol, Amril Nazir, Khalid Almaymun, Mustafa Abdulrazek, Reem Muhammad, Fatima Adlan, Ravishankar Sharma

All Works

This paper addresses the critical challenge of fraud detection in medical insurance claims, a pervasive issue causing significant financial losses in healthcare. The primary goal is to develop an advanced fraud detection approach by integrating multiple unsupervised machines learning algorithms, leveraging their collective strengths through a majority voting mechanism, where labelling of data is unavailable. Central to this approach is the ensemble of 18 novel unsupervised algorithms, specifically, anomaly detection models. The novelty lies in the majority voting system employed to aggregate the decisions from these diverse algorithms, enhancing the reliability and accuracy of fraud detection. To validate the effectiveness …

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.

Moab Local Food Guide 2024, Roslynn Mccann, Shiree Duncan

All Current Publications

Thank you for picking up this Moab Utah Local Food Guide. Those growing and producing food within a 100 mile radius from Moab have been included in this guide to help YOU make locally sourced purchasing decisions!

Visualsphere: A Web-Based Interactive Visualization System For Clinical Research Data, Shiwei Lin, Shiqiang Tao, Wei-Chun Chou, Guo-Qiang Zhang, Xiaojin Li

Faculty, Staff and Student Publications

Clinical research data visualization is integral to making sense of biomedical research and healthcare data. The complexity and diversity of data, along with the need for solid programming skills, can hinder advances in clinical research data visualization. To overcome these challenges, we introduce VisualSphere, a web-based interactive visualization system that directly interfaces with clinical research data repositories, streamlining and simplifying the visualization workflow. VisualSphere is founded on three primary component modules: Connection, Configuration, and Visualization. An end-user can set up connections to the data repositories, create charts by selecting the desired tables and variables, and render visualization dashboards generated by …

Xuebijing Improves Intestinal Microcirculation Dysfunction In Septic Rats By Regulating The Vegf-A/Pi3k/Akt Signaling Pathway, A-Ling Tang, Yan Li, Li-Chao Sun, Xiao-Yu Liu, Nan Gao, Sheng-Tao Yan, Guo-Qiang Zhang

Faculty, Staff and Student Publications

BACKGROUND: This study aims to explore whether Xuebijing (XBJ) can improve intestinal microcirculation dysfunction in sepsis and its mechanism.

METHODS: A rat model of sepsis was established by cecal ligation and puncture (CLP). A total of 30 male SD rats were divided into four groups: sham group, CLP group, XBJ + axitinib group, and XBJ group. XBJ was intraperitoneally injected 2 h before CLP. Hemodynamic data (blood pressure and heart rate) were recorded. The intestinal microcirculation data of the rats were analyzed via microcirculation imaging. Enzyme-linked immunosorbent assay (ELISA) kits were used to detect the serum levels of interleukin-6 (IL-6), …

Statins Are Rarely Prescribed For Incidentally Discovered Covert Cerebrovascular Disease: A Retrospective Cohort In A Large Electronic Health Record (Ehr) Identified Using Natural Language Processing, Lester Y Leung, Eric Puttock, David F Kallmes, Patrick Luetmer, Sunyang Fu, Chengyi X Zheng, Hongfang Liu, Wansu Chen, David M Kent

Faculty, Staff and Student Publications

Introduction: While incidentally discovered covert cerebrovascular diseases (id-CCD) are associated with future stroke, it is not known if patients with id-CCD are prescribed statins.

Methods: Patients age ≥50 with id-CCD on neuroimaging from 2009 to 2019 with no prior ischaemic stroke, transient ischaemic attack or dementia were identified using natural language processing in a large real-world cohort. Robust Poisson multivariable regression was used to assess statin prescription among patients without prior statins.

Results: Among 2 41 050 patients, 74 975 patients (31.1%; 4.7% with covert brain infarcts (CBI); 29.0% with white matter disease (WMD)) had id-CCD. 53.5% (95% CI 53.2 …

The Enact Network Is Acting On Housing Instability And The Unhoused Using The Open Health Natural Language Processing Toolkit, Daniel R Harris, Sunyang Fu, Andrew Wen, Alexandria Corbeau, Darren Henderson, Jordan Hilsman, David Oniani, Yanshan Wang

Faculty, Staff and Student Publications

Housing is an environmental social determinant of health that is linked to mortality and clinical outcomes. We developed a lexicon of housing-related concepts and rule-based natural language processing methods for identifying these housing-related concepts within clinical text. We piloted our methods on several test cohorts: a synthetic cohort generated by ChatGPT for initial infrastructure testing, a cohort with substance use disorders (SUD), and a cohort diagnosed with problems related to housing and economic circumstances (HEC). Our methods successfully identified housing concepts in our ChatGPT notes (recall = 1.0, precision = 1.0), our SUD population (recall = 0.9798, precision = 0.9898), …

A Comparative Study Of Early Fusion And Multimodal Siamese Neural Network Using Image And Text Data In Food Classification, Kanokporn Sintarasirikulchai

Chulalongkorn University Theses and Dissertations (Chula ETD)

The economic development under capitalism has significantly transformed people's lifestyles, resulting in a fast-paced daily life. This shift has increased the consumption of convenient food options, leading to a preference for fast food, which is often high in carbohydrates and fats. Consequently, there has been a rise in obesity and related health issues, highlighting the importance of monitoring food intake. Automated systems utilizing artificial intelligence (AI) have emerged as potent tools for providing personalized dietary advice and monitoring. With the growing volume of food-related content on social media, including images and accompanying text, leveraging multimodal data has become essential for …

Leveraging Gpt-4 For Thai Land And Building Tax Calculation In The Bangkok Area, Sadanan Arsaibun

Chulalongkorn University Theses and Dissertations (Chula ETD)

This paper explores the application of large language models (LLMs), particularly GPT-4, in the context of Thai land and building tax computation. We address challenges arising from the complexity of Thai tax regulations, the scarcity of Thai-language training data, and the structured, multi-variable nature of tax computation tasks. To enhance model performance, we employ few-shot prompting, decomposition-based reasoning, and retrieval-augmented generation (RAG). We further introduce a JSON-style output format to support structured conversational interactions, allowing the model to identify missing user inputs and guide data collection. A secondary o3-mini pass is used as an answer verifier, helping to assess the …

Roles Of Reactants And Catalysts For Suzuki C-C Cross-Coupling Reaction Using Molecular Volcano Plots And Big-Data Analysis, Kwanchanok Kaewkwan

Chulalongkorn University Theses and Dissertations (Chula ETD)

The Suzuki carbon–carbon cross-coupling reaction is a well-known method for forming carbon–carbon bonds, catalyzed by transition metal complexes. Molecular volcano plots have been used to predict promising catalysts. While previous studies focused mainly on thermodynamic insight, with limited focus on kinetics, this work examines the influence of catalysts and reactants on catalytic activity by incorporating kinetic energy profiles. DFT calculations were performed to evaluate free energies of catalytic states, establish linear scaling relationships, and construct volcano plots. This study employed nickel and palladium complexes with five different ligands as catalysts, reacting with twenty-five substrates. Results show that catalysts have a …

Development Of Biosensors For Detection Of Human Papillomavirus Type 16 Dna, Panon Tungkunaruk

Chulalongkorn University Theses and Dissertations (Chula ETD)

This dissertation aims to develop two unconventional methods for the detection of a high-risk, cancer-causing human papillomavirus type 16 (HPV16) on miniaturized platforms for point-of-care testing (POCT). The first part focuses on a direct detection of HPV16 double-stranded DNA (dsDNA) using the non-self-pairing pyrrolidinyl peptide nucleic acid (acpcPNA) pair on a paper-based substrate coupled with fluorescence and electrochemical dual detections using a DNA-binding dye in a single device. The acpcPNA pair can perform double duplex invasion into dsDNA by hybridizing with each strand of DNA in the same region without the requirement of denaturing dsDNA to single-stranded DNA (ssDNA) before …

Flow Accelerated Corrosion In Economizer Tubes Of A Heat Recovery Steam Generator Using Computational Fluid Dynamics, Phuris Khunphakdee

Chulalongkorn University Theses and Dissertations (Chula ETD)

This study investigates the effects of operating and harp design parameters on flow accelerated corrosion (FAC) using CFD, based on a real case from a modern heat recovery steam generator (HRSG) where severe FAC occurred in an economizer harp after six years of service. Preliminary investigation found that contributions for this case include the construction material, which is carbon steel, the operating temperature range that close to FAC favourable temperature (150 °C), reduced water and disturbed flow geometry. Comparison of shear stress and wall thickness profiles confirmed that FAC is mass transfer dominated, driven by flow acceleration and recirculation near …

Development Of Electrochemical Sensor Based On Molecularly Imprinted Polymer For Determination Of Roxarsone, Pitranan Kunthanasap

Chulalongkorn University Theses and Dissertations (Chula ETD)

Roxarsone (ROX), a growth enhancer in poultry and swine, has the potential to improve weight gain and prevent coccidiosis. Excessive use of ROX may lead to its accumulation in animal tissues, potentially contaminating food products and posing health risks to humans through consumption. This research aims to develop an electrochemical sensor based on a magnetic screen-printed graphene electrode (mSPGE) modified with a highly selective molecularly imprinted polymer (MIP) for the determination of ROX with enhanced sensitivity and selectivity. Typically, an applied magnetic field at the mSPGE could accelerate mass transport at the electrode surface through the magnetohydrodynamic (MHD) effect, leading …

Aza-Piancatelli And Friedel-Crafts Reactions Under Microdroplet Conditions, Phanomsak Yukhet

Chulalongkorn University Theses and Dissertations (Chula ETD)

The development of sustainable, catalyst-free synthetic methodologies represents a significant advance in modern organic chemistry, which offers environmentally benign alternatives to conventional protocols that often rely on metal catalysts or harsh reagents. This dissertation presents two complementary studies to explore and establish novel catalyst-free synthetic strategies in both microdroplet and bulk solution phases and highlight their potential applications, mechanistic features, and practical limitations. The first part of this work investigates the rearrangement of furfural with secondary amines to yield 4,5-diaminosubstituted cyclopentenones (4,5-DCPs) within the unique reaction environment of electrospray-generated charged microdroplets. The reaction is proposed to proceed via a Stenhouse …

Geology Of The Bishop 7.5-Minute Quadrangle And Geophysical Modeling To Constrain Syndepositional Deformation Of The Black Warrior Foreland Basin, Joseph Edward Lane

Graduate Theses/Dissertations

The Black Warrior basin is a foreland basin formed by the Ouachita–Appalachian orogeny during the late Paleozoic Era. Mississippian strata record the transition from the carbonate platform within the Tuscumbia Limestone to the clastic-rich units within the overlying Pride Mountain Formation and Hartselle Sandstone. The western and southern-most fringes of the Black Warrior basin are covered by thin, post orogenic Cretaceous sediments of the East Gulf Coastal Plain. Geological mapping of the Bishop 7.5-Minute quadrangle revealed subtle deformation in Mississippian strata in the form of gentle folds (

End-To-End Development Of Mandible Reconstruction Using Machine Learning, Nattapon Kamboonsri

Chulalongkorn University Theses and Dissertations (Chula ETD)

Virtual surgical planning (VSP) is a critical step in mandible reconstruction surgery, which involves preoperative planning and implant design to restore mandibular defects. This study focuses on two essential components of VSP: (1) mandible segmentation, specifically the separation of healthy and defective regions, which currently relies on manual annotation, and (2) generation of the complete mandible, where traditional approaches such as mirroring the contralateral side often fail in scenarios involving midline-crossing defects. While recent automated methods have addressed mandible segmentation from CT scans, they typically focus only on binary segmentation and often utilize conventional UNet-based architectures that suffer from limited …

Automated Cecum Identification In Colonoscopy Using Deep Learning Approach, Kittipoom Sutthinuntakorn

Chulalongkorn University Theses and Dissertations (Chula ETD)

Colonoscopy is essential for the early detection and prevention of colorectal cancer. Identifying the cecum is a vital element of this process. However, most existing automated methods rely on still images or temporal cues alone, without incorporating camera motion awareness, and are rarely suitable for real-time use. In this paper, we present a real-time cecum detection method that integrates spatial features, temporal modeling, and camera motion cues. We deploy ConvNeXtV2 for spatial feature extraction, LTContext for temporal modeling, and Depth Anything in Robotic Endoscopic Surgery (DARES) to enhance understanding of endoscope positioning and camera motion. The dataset used in this …

Craftjai: Llm-Enhanced Generative Ai For Gamified Mood Journal, Kingrak Phairoh

Chulalongkorn University Theses and Dissertations (Chula ETD)

Mood journaling helps promote emotional self-awareness, but maintaining user motivation remains difficult. This study presents CRAFTJAI, a generative AI system that enhances journaling through emotionally aligned visual feedback. Instead of replacing diary entries, the system adds character-based imagery to support intrinsic motivation and symbolic emotional reflection. It consists of emotional feature extraction using a large language model, emotion-labeled dataset construction based on eight core moods, and LoRA fine-tuning of a diffusion-based text-to-image model to improve emotional fidelity and stylistic consistency. A web-based prototype was developed with Gradio for real-time evaluation. Drawing on gamification and self-determination theory, the system integrates emotional …