Physics Commons^™

Ai-Driven Segmentation And Volumetric Response Modeling Of Liver Regions To Radiotherapy, Aashish Chandra Gupta

Dissertations and Theses (Open Access)

In liver-directed radiotherapy (RT), liver regions receiving higher doses typically undergo atrophy while contralateral/adjacent lower-dose regions may exhibit compensatory hypertrophy through regeneration of healthy tissue. Optimizing the RT plan to promote regional hypertrophy while minimizing the risk of developing atrophy has the potential to enhance post-RT liver function and long-term survivorship. However, current clinical practice largely relies on global liver dose-volume metrics during RT-planning, which may obscure favorable dose-response correlation and limit actionable guidance for clinicians. Therefore, we hypothesized that post-RT regional liver response is governed by a combination of region-specific dose-volume and patient clinical features, and that these responses …

Uranium Chemical Compound Classification Using Sub-Images And Statistical Machine Learning For Nuclear Forensics, Lee C. Lambert, Brett J. Borghetti, Abigail A. Bickley

Faculty Publications

Uranium particle analysis from Scanning Electron Microscope (SEM) imagery is a crucial tool in nuclear forensics. The particle morphology lexicon proposed by Tamasi et al. in J Radioanal Nucl Chem 307, 1611–1619 (2016) follows a standardized, manual identification process to identify particle morphology features. The present work seeks to mirror this methodology using computer feature selection from the scikit-image Python library rather than human classification. Using a random forest classifier, a 56% overall uranium true positive classification accuracy (a 39.6% balanced classification accuracy) was achieved on a test set outperforming a naïve (chance) model by 48%. The methodology introduced splits …

Applications Of Machine Learning In Gravitational-Wave Research With Current Interferometric Detectors, Elena Cuoco, Marco Cavaglià, Ik Siong Heng, David Keitel, Christopher Messenger

Physics Faculty Research & Creative Works

This article provides an overview of the current state of machine learning in gravitational-wave research with interferometric detectors. Such applications are often still in their early days but have reached sufficient popularity to warrant an assessment of their impact across various domains, including detector studies, noise and signal simulations, and the detection and interpretation of astrophysical signals. In detector studies, machine learning could be useful to optimize instruments like LIGO, Virgo, KAGRA, and future detectors. Algorithms could predict and help in mitigating environmental disturbances in real time, ensuring detectors operate at peak performance. Furthermore, machine-learning tools for characterizing and cleaning …

Fish-Spec: Fast Identification System For Handheld Spectroscopy And Species Classification, Mitchell Sueker, Nicholas Mackinnon, Gregory Bearman, Amanda Tabb, Diane Kim, Rosalee S. Hellberg, Alireza Akhbardeh, Hamid Reza Marateb, Jianwei Qin, Moon Kim, Fartash Vasefi, Hossein Kashani Zadeh

Food Science Faculty Articles and Research

Accurate fish species identification is critical to prevent mislabeling and fraud in the seafood industry. We present a handheld multi-mode point spectroscopy system that combines fluorescence (365 and 395 nm excitation) and reflectance measurements in the visible to near-infrared (∼350–900 nm) and short-wave infrared (∼900–1700 nm) regions for rapid, non-destructive classification of fish fillets. Tissue spectra were acquired at 25 positions on 68 fillets from 11 species, in both frozen and thawed states. Feature-level fusion across all four modes enabled higher classification accuracy than any single mode alone. A global machine-learning model classified all species with 85 ± 2.8 %, …

Investigating Characteristic Droplet Size Distributions In Large Eddy Simulations Of Stratocumulus Clouds, Nithin Allwayin, Daniel J. Miller, Kamal Kant Chandrakar, Michael L. Larsen, Raymond A. Shaw

Michigan Tech Publications

Cloud processes relevant to radiative and precipitation properties depend on the shape of the cloud droplet size distribution. Recent holographic observations revealed that cloud droplet populations do not have the same size distribution shapes throughout but form regions of characteristic distributions with similar microphysical properties. We investigate the existence and properties of these characteristic distributions within Large-Eddy Simulations of stratocumulus clouds using Lagrangian and bin microphysics schemes. Distribution types are identified, revealing localized characteristic distributions that vary on the scale of the largest convective cell for simulations with bin microphysics. The results from the Lagrangian microphysics scheme hint at similar …

Machine Learning Crime Prediction Models And The Gap Between Research And Implementation: A Systematic Review, Ricardo Huamantingo, Miguel Cano-Lengua, Ciro Rodriguez

Karbala International Journal of Modern Science

A crime is an illegal or violent act committed by one individual against another. The increasing crime rate has become a major concern as it negatively affects people's quality of life and generates significant social and economic costs. This study aims to identify the most widely used machine learning (ML) models for crime prediction, determine evaluation metrics for assessing model performance, and analyze key data characteristics to enhance real-world implementation. The study follows the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. A search string was formulated using the population, intervention, comparison, and outcomes (PICO) framework and applied …

Revised Unsteady Drag Force And Its Impact On Particle Dynamics At High Reynolds Numbers, Ahmad Talaei, Timothy J. Garrett

Physics Student Research

The accurate modeling of particle motion in viscous fluids at high Reynolds numbers remains a fundamental challenge, particularly under unsteady flow conditions where nonlinear effects dominate. This study introduces a revised unsteady drag formulation, extending its applicability beyond previous studies. By integrating classical solutions of the Navier–Stokes equations, the new formulation overcomes the limitations of the Maxey–Riley–Gatignol (MRG) equation, addressing non-physical memory effects associated with Basset drag and extending its applicability to higher Reynolds numbers.

Through experimental comparisons and direct numerical simulations (DNS), the revised equation of motion is compared to the MRG equation, both during steady-state conditions and non-equilibrium …

An Intercomparison Of Wall Fluxes In A Turbulent Thermal Convection Chamber: Direct Numerical Simulations And Wall-Modeled Large-Eddy Simulations Enhanced By Machine Learning, Aaron Wang, Silvio Schmalfuß, Kamal Kant Chandrakar, Hadi Zanganeh Kia, Fan Yang, Mikhail Ovchinnikov, Raymond A. Shaw, Yunsoo Choi

Michigan Tech Publications

Thermal convection in a closed chamber is driven by a warm bottom, a cold top, and side walls at various temperatures. Although wall fluxes are the source of convection energy, accurately modeling these fluxes (i.e., the wall model) is challenging. In large-eddy simulations (LESs), many wall models are traditionally derived from the canonical boundary layer, which may be unsuitable for thermal convection bounded by both horizontal and vertical walls. This study conducts a model intercomparison of dry convection in a cubic-meter chamber using three direct numerical simulations (DNSs) and four LESs with different wall models. The LESs employ traditional wall …

Parameter Estimation Of Binary Black Hole Coalescence Using Lstm Neural Networks, Lane Scheel

Undergraduate Honors Papers

The Advanced Laser Interferometer Gravitational Wave Observatory (aLIGO) made its first detection of gravitational waves in 2015. Since then, the rate of event detection has only increased, with a detection being made every 2-3 days during the current observing run, O4. This rapid influx of data has the potential to create bottle-necks in data analysis efforts, and can delay the scientific progress which require those efforts. Traditional gravitational wave data analysis techniques, such as matched filtering, require extremely large template banks of synthetic gravitational waveforms and can often fail to provide meaningful limits on system parameters. Not only is this …

Artificially Intelligent Maxwell's Demon For Optimal Control Of Open Quantum Systems, Paolo A. Erdman, Robert Czupryniak, Bibek Bhandari, Andrew N. Jordan, Frank Noé, Jens Eisert, Giacomo Guarnieri

Mathematics, Physics, and Computer Science Faculty Articles and Research

Feedback control of open quantum systems is of fundamental importance for practical applications in various contexts, ranging from quantum computation to quantum error correction and quantum metrology. Its use in the context of thermodynamics further enables the study of the interplay between information and energy. However, deriving optimal feedback control strategies is highly challenging, as it involves the optimal control of open quantum systems, the stochastic nature of quantum measurement, and the inclusion of policies that maximize a long-term time- and trajectory-averaged goal. In this work, we employ a reinforcement learning approach to automate and capture the role of a …

Isotopic Analysis Of Lithium Hydroxide Monohydrate Using Laser-Induced Breakdown Self-Reversal Isotopic Spectrometry (Libris) And Machine Learning, Madison R. Moran

Theses and Dissertations

High-resolution Laser-Induced Breakdown Self-Reversal Isotopic Spectrometry (LIBRIS) is implemented to record the 15.8 pm Li I 670.8 nm isotopic shift in LiOH · H₂O samples of varying 6, 7Li abundance. A simple univariate linear regression demonstrates an acquired isotopic shift of 13.813 ± 1.21 pm in samples varying from 3 to 95 ⁶Li at%. Supervised machine learning regressions are trained on self-reversal wavelength locations in order to quantify ⁶Li abundance. A stacked ensemble using two base learners yields the superlative characterization of 6Li abundance with RMSE of 5.66 at% and detection limit of 18.8 at%. Using …

Deep Learning-Based Superconductivity Prediction And Experimental Tests, Daniel Kaplan, Adam Zheng, Joanna Blawat, Rongying Jin, Robert J. Cava, Viktor Oudovenko, Gabriel Kotliar, Anirvan M. Sengupta, Weiwei Xie

Faculty Publications

The discovery of novel superconducting materials is a long-standing challenge in materials science, with a wealth of potential for applications in energy, transportation and computing. Recent advances in artificial intelligence (AI) have enabled expediting the search for new materials by efficiently utilizing vast materials databases. In this study, we developed an approach based on deep learning (DL) to predict new superconducting materials. We have synthesized a compound derived from our DL network and confirmed its superconducting properties in agreement with our prediction. Our approach is also compared to previous work based on random forests (RFs). In particular, RFs require knowledge …

Ai For Nuclear Physics: The Exclaim Project, S. Liuti, D. Adams, M. Boër, G. W. Chern, M. Cuic, M. Engelhardt, G. R. Goldstein, B. Kriesten, Y. Li, H. W. Lin, M. Sievert, D. Sivers

Computer Science Faculty Publications

An overview of the recent activity of the newly funded EXCLusives with AI and Machine learning (EXCLAIM) collaboration is presented. The main goal of the collaboration is to develop a framework to implement AI and machine learning techniques in problems emerging from the phenomenology of high energy exclusive scattering processes from nucleons and nuclei, maximizing the information that can be extracted from various sets of experimental data, while implementing theoretical constraints from lattice QCD. A specific perspective embraced by EXCLAIM is to use the methods of theoretical physics to understand the working of ML, beyond its standardized applications to physics …

Optimal Manipulation Motion Action Planner Enabled By Physics Informed Neural Networks, Jonathon E. Slightam, Logan E. Beaver

Mechanical & Aerospace Engineering Faculty Publications

Autonomous robotic manipulation in unstructured environments faces many challenges and is hindered by capabilities that bridge the gap between perception and acting on the world. Action plans that are centric to object motion rather than end-of-arm tooling behavior may aid this. This paper presents an autonomous action planner for a feedback linearizeable system comprised of three base motions that can be leveraged on their own or in combination to give custom motion plans. The optimization routine for the three different types of motion are presented, which are integrated into physics informed neural networks. A component of this is the autonomy …

Harnessing The Power Of Gradient-Based Simulations For Multi-Objective Optimization In Particle Accelerators, Kishansingh Rajput, Malachi Schram, Auralee Edelen, Jonathan Colen, Armen Kasparian, Ryan Roussel, Adam Carpenter, He Zhang, Jay Benesch

Data Science Faculty Publications

Particle accelerator operation requires simultaneous optimization of multiple objectives. Multi-objective optimization (MOO) is particularly challenging due to trade-offs between the objectives. Evolutionary algorithms, such as genetic algorithms (GAs), have been leveraged for many optimization problems, however, they do not apply to complex control problems by design. This paper demonstrates the power of differentiability for solving MOO problems in particle accelerators using a deep differentiable reinforcement learning (DDRL) algorithm. We compare the DDRL algorithm with model-free reinforcement learning (MFRL), GA, and Bayesian optimization (BO) for simultaneous optimization of heat load and trip rates in the continuous electron beam accelerator facility. The …

Data-Driven Gradient Optimization For Field Emission Management In A Superconducting Radio-Frequency Linac, S. Goldenberg, K. Ahammed, A. Carpenter, J. Li, R. Suleiman, C. Tennant

Electrical & Computer Engineering Faculty Publications

Field emission can cause significant problems in superconducting radio-frequency linear accelerators (linacs). When cavity gradients are pushed higher, radiation levels within the linacs may rise exponentially, causing degradation of many nearby systems. This research aims to utilize machine learning with uncertainty quantification to predict radiation levels at multiple locations throughout the linacs and ultimately optimize cavity gradients to reduce field emission-induced radiation while maintaining the total linac energy gain necessary for the experimental physics program. The optimized solutions show over 40% reductions for both neutron and gamma radiation from the standard operational settings.

Point Cloud-Based Diffusion Models For The Electron-Ion Collider, Jack Y. Araz, Vinicius Mikuni, Felix Ringer, Nobuo Sato, Fernando Torales Acosta, Richard Whitehill

Physics Faculty Publications

At high-energy collider experiments, generative models can be used for a wide range of tasks, including fast detector simulations, unfolding, searches of physics beyond the Standard Model, and inference tasks. In particular, it has been demonstrated that score-based diffusion models can generate high-fidelity and accurate samples of jets or collider events. This work expands on previous generative models in three distinct ways. First, our model is trained to generate entire collider events, including all particle species with complete kinematic information. We quantify how well the model learns event-wide constraints such as the conservation of momentum and discrete quantum numbers. We …

Detecting Anomalous Srf Cavity Behavior With Unsupervised Learning, Hal Ferguson, Jiang Li, Adam Carpenter, Chris Tennant, Dillon Thomas, Dennis Turner

Electrical & Computer Engineering Faculty Publications

We present an unsupervised learning framework for detecting anomalous superconducting radio-frequency (SRF) cavity behavior at the Continuous Electron Beam Accelerator Facility (CEBAF), emphasizing its initial performance and effectiveness. Key to the system’s success was the development of data acquisition systems (DAQs) that capture fast-sampled, information-rich signals, essential for detecting transient effects. The approach involves creating daily cavity-specific models using principal component analysis to handle variations in rf signal behavior and mitigate performance degradation from data drift. This unsupervised method eliminates the need for expensive labeling by continuously updating models with recent data. Deployed and operational for 3 months before a …

Applying Machine‐Learning Methods To Laser Acceleration Of Protons: Lessons Learned From Synthetic Data, Ronak Desai, Thomas Zhang, J. J. Felice, Ricky Oropeza, Joseph R. Smith, Alona Kryshchenko, Chris Orban, Michael L. Dexter, Anil K. Patnaik

Faculty Publications

In this study, we consider three different machine-learning methods—a three-hidden-layer neural network, support vector regression, and Gaussian process regression—and compare how well they can learn from a synthetic data set for proton acceleration in the Target Normal Sheath Acceleration regime. The synthetic data set was generated from a previously published theoretical model by Fuchs et al. 2005 that we modified. Once trained, these machine-learning methods can assist with efforts to maximize the peak proton energy, or with the more general problem of configuring the laser system to produce a proton energy spectrum with desired characteristics. In our study, we focus …

Influential Factors And Predicting Dose Delivery Accuracy For Imaging And Radiation Oncology Core’S Phantom Program Using Machine Learning, Hunter Mehrens

Dissertations and Theses (Open Access)

IROC’s mission is to help ensure consistent and comparable, high-quality radiotherapy across clinics that participate in national clinical trials. To obtain this mission, IROC’s phantom program provides a third-party end-to-end check of the clinical workflow of a patient receiving radiotherapy. The goal of the phantom audit is to compare the dose delivered to the dose planned by the treatment system ensuring dose delivery accuracy. While IROC’s phantoms are better equipped to catch dose delivery errors compared to a clinic’s QA process, the end-to-end process and reporting of results is time-consuming creating a bottleneck for clinical trial participation. Furthermore, IROC’s passing …

Mesostructure Reconstruction Of Prepreg Platelet Molded Composite With Artificial Intelligence, Richard Larson

Mechanical & Aerospace Engineering Theses & Dissertations

Prepreg platelet molded composites (PPMC) are long, discontinuous fiber reinforced polymer materials. PPMC are an important subcategory of composite materials as they are processible into geometrically complex structures and can be produced via high-throughput manufacturing processes, however they have higher stiffness and strength as compared to traditional discontinuous fiber reinforced polymers. However, there is inherent randomness in the structure of PPMCs and as such, PPMC parts frequently require per part testing that is cost prohibitive.

Herein, a method using artificial intelligence is proposed as a more cost-effective method of inspecting PPMC parts. Different artificial intelligence (AI) architectures are explored to …

On The Ubiquity, Properties And Evolution Of Small-Scale Magnetic Flux Ropes In The Heliosphere, Hameedullah Farooki

Dissertations

The solar wind is a plasma constantly blowing out from the Sun with a large-scale magnetic field having significant local complexity at small scales. Small-scale magnetic flux ropes (SMFRs), plasma structures with twisted field lines, are an important element of this complexity. This dissertation contributes several studies that further our understanding of SMFRs. The first study applies machine learning to measurements from Wind labeled by the presence of SMFRs and magnetic clouds (MCs). MCs were distinguished from non-MFRs with an AUC of 94% and SMFRs with an AUC of 89% and had distinctive plasma properties, whereas SMFRs appeared to be …

Computational Microscopy For Biomedical Imaging With Deep Learning Assisted Image Analysis, Yuwei Liu

Dissertations

Microscopy plays a crucial role across various scientific fields by enabling structural and functional imaging with microscopic resolution. In biomedicine, microscopy contributes to basic research and clinical diagnosis. Conventionally, optical microscopy derives its contrast from the amplitude of the optical wave and provides visualization of the physical structure of the sample qualitatively. To understand the function at the cellular or tissue level, there is a need to characterize the sample quantitatively and explore contrast mechanisms other than light intensity. Image enhancement or reconstruction from microscopic imaging systems is known as computational microscopy, and it involves the application of computational techniques …

A Multivariate Analysis Of The Gravitational Wave Signal Landscape From Core Collapse Supernovae, Raul Alberto Espinosa Perez

Theses and Dissertations

Core collapse supernovae (CCSN) are highly anticipated sources of gravitational waves (GW) during the on-going fourth observation run (O4) of GW detectors like LIGO and the future observation runs. The GW signal from the CCSN cannot be modeled mathematically. Several groups around the world have engaged in simulation of the predicted GW signals from CCSN sources. These simulations are carried out in supercomputers, and they incorporate general relativity, hydrodynamics, neutrino physics, mass and angular momentum of the stellar progenitor and nuclear equations of state (EoS). The output consists of simulated signals with varying duration, peak frequency, GW energy and time-frequency …

Predicting Biomolecular Properties And Interactions Using Numerical, Statistical And Machine Learning Methods, Elyssa Sliheet

Mathematics Theses and Dissertations

We investigate machine learning and electrostatic methods to predict biophysical properties of proteins, such as solvation energy and protein ligand binding affinity, for the purpose of drug discovery/development. We focus on the Poisson-Boltzmann model and various high performance computing considerations such as parallelization schemes.

Probing The Ising Model’S Thermodynamics Through Restricted Boltzmann Machines, Xiaobei (Emma) Zhang

HMC Senior Theses

This thesis explores the connection between physics and machine learning by using Restricted Boltzmann Machines (RBMs) to study the thermodynamic properties of the Ising model. The Ising model is a simple but realistic model that captures the magnetic behavior of a system, where spins occupy a lattice of sites and different spin configurations correspond to different energies. The model exhibits phase transitions between ferromagnetic and paramagnetic phases as a function of temperature. RBMs are two-layered neural networks that can learn probability distributions over binary spins. The study generates 2D Ising model data at different temperatures using Monte Carlo simulations, including …

A Data-Driven Machine Learning Approach For Electron-Molecule Ionization Cross Sections, Allison Harris, Josh Nepomuceno

Faculty publications – Physics

Despite their importance in a wide variety of applications, the estimation of ionization cross sections for large molecules continues to present challenges for both experiment and theory. Machine learning (ML) algorithms have been shown to be an effective mechanism for estimating cross section data for atomic targets and a select number of molecular targets. We present an efficient ML model for predicting ionization cross sections for a broad array of molecular targets. Our model is a 3-layer neural network that is trained using published experimental datasets. There is minimal input to the network, making it widely applicable. We show that …

The Application Of Novel Machine Learning Algorithms To Study Multi-Dimensional Fragmentation Functions Of Hadrons In Jets At Star, Hannah A. Harrison-Smith

Theses and Dissertations--Physics and Astronomy

Hadronization, the process by which colored quarks and gluons shower from high energy collisions and recombine to form stable, experimentally-observable particles, is a fundamental aspect of Quantum Chromodynamics (QCD) that is not yet fully understood. Fragmentation functions, typically measured in electron-positron collisions, encapsulate this hadronization process well for quarks. Studying proton-proton collisions offers direct access to gluon fragmentation that other channels like electron-positron do not. Recent theoretical developments have proposed the study of hadronic showers in groupings called jets, introducing the concept of multi-dimensional jet fragmentation functions. This thesis presents the extraction of collinear and transverse momentum-dependent fragmentation functions for …

Dynamic Exchange-Correlation Functional For Bandgap Optimization: Reparametrization And Machine Learning, Viviana Faride Dovale Farelo

Graduate Theses, Dissertations, and Problem Reports (ETD)

This dissertation explores predicting the physical properties of solids using first-principles methods, with a focus on Density Functional Theory (DFT). DFT uses the electronic density within a material to predict its properties, simplifying the treatment of electron-electron interactions and allowing the study of realistic systems with a balanced treatment between accuracy and computational efficiency. Additionally, machine learning (ML) is employed to create correlations between some physical properties of solids and other properties or parameters that are more difficult to calculate.

The main problem addressed in this study is adjusting the parameters in the Strongly Constrained and Appropriately Normed (SCAN) semilocal …

Accelerating Markov Chain Monte Carlo Sampling With Diffusion Models, N. T. Hunt-Smith, W. Melnitchouk, F. Ringer, N. Sato, A. W. Thomas, M. J. White

Physics Faculty Publications

Global fits of physics models require efficient methods for exploring high-dimensional and/or multimodal posterior functions. We introduce a novel method for accelerating Markov Chain Monte Carlo (MCMC) sampling by pairing a Metropolis-Hastings algorithm with a diffusion model that can draw global samples with the aim of approximating the posterior. We briefly review diffusion models in the context of image synthesis before providing a streamlined diffusion model tailored towards low-dimensional data arrays. We then present our adapted Metropolis-Hastings algorithm which combines local proposals with global proposals taken from a diffusion model that is regularly trained on the samples produced during the …