Published Research

Featured Articles

More Articles

September 2026

• Innovation in Social Work Research: Integration of Wearable Technology and AI
  Discover how emerging technologies can help social work researchers engage participants more flexibly while reducing barriers to traditional research methods.
  

August 2026

• AI Edge: Q3
  Explore perspectives on institutional AI stewardship, emerging requirements for stablecoin issuers, and a new legal practice at the intersection of AI regulation and engineering.
• AI in Thyroid Nodules and Cancer: Clinical Validation and Real-World Performance
  How AI can improve thyroid nodule risk assessment, reduce unnecessary biopsies, and support more personalized clinical decision-making.
• Trusting to Continue or Resist: How AI Governance Shapes AI-Based Emotional Support Usages Across Countries
  Learn how AI governance and trust shape whether people embrace or resist AI-based emotional support across countries.
• Trust Calibration and Patient Decision in AI-Supported Infectious Disease Triage
  How trust in AI shapes whether patients accept or reject its recommendations in time-sensitive infectious disease triage.
• When Expectations Clash: Dynamic Trust Calibration and Expectation Violation in AI-Driven Sports Analytics DSS
  How unexpected AI predictions and the quality of their explanations shape whether users trust, or override, AI-driven sports analytics.
• Anticipatory Cognitive Offloading with GenAI
  How the expectation of AI assistance may change the way students study, think, and engage with learning before an assessment.
• Believing the Machine: Human Trust and the Impact of AI Language Confidence and Elaboration
  How an AI’s confidence and level of detail can shape user trust and attention, regardless of whether its recommendations are actually correct.
• Measuring and Detecting Harmful AI Sycophancy
  How researchers are measuring and detecting harmful AI sycophancy, when models change their responses simply to align with a user’s preferences.
• Now’s the Time: Computer Science Must Evolve to Emphasize Software and Systems Engineering With AI
  See how computer science education can evolve by treating programming, data structures, and algorithms as building blocks for broader systems and engineering skills.
• VisionDES: Robust and Explainable Dynamic Vision Ensemble
  How a new AI framework improves the accuracy, robustness, and explainability of image classification by dynamically selecting the most reliable models.
• Plain-language prompting improves readability of ChatGPT-generated patient education for meniscal surgery without loss of accuracy
  How plain-language prompting can make AI-generated patient education for meniscal surgery easier to understand without sacrificing accuracy.
• LLMs as Examinees and Graders in Simulated General Surgery Oral Board-Style Cases: A Psychometric Pilot Study
  How large language models perform as both examinees and graders in simulated surgical oral boards, and their potential role in surgical training and assessment.

July 2026

• Building a Custom Taxonomy of AI Skills and Tasks from the Ground Up with Job Postings
  Explore how LLMs can build more effective taxonomies of AI skills from job data, showing that carefully filtered inputs, not more data, lead to clearer and more meaningful results.
• LLM-Assisted Formalization for Deterministic Detection of Statutory Inconsistency in Tax Law
  Learn how combining LLMs with symbolic logic enables consistent, transparent detection of legal inconsistencies, overcoming the limits of probabilistic AI in complex tax law.
• Examining Students' Knowledge, Use, and Perceived Benefits of Artificial Intelligence Tools in Higher Education
  How students are using AI in higher education, the benefits they see, and the ethical concerns shaping their experiences.
• AI for radiation protection in medical imaging and radiotherapy: A perspective from the AI Working Party of ICRP Committee 3
  How AI can improve medical radiation protection while addressing critical challenges around safety, bias, accountability, and clinical oversight.

June 2026

• Can LLMs Write Correct TLA+ Specifications? Evaluating Natural-Language-to-TLA+ Generation
  This paper explores how emerging AI architectures are making systems more reliable, transparent, and ready for real-world use.
• Five Boundaries of Permissible Automation
  How to determine when outsourcing bureaucratic decisions to machine learning becomes ethically impermissible.
• AI Applications Versus Manual Methods For Literature Retrieval: A Comparative Analysis
  Learn how AI tools can support literature reviews while highlighting why human oversight remains essential for identifying reliable research.
• Levels of Understanding, World Models, and AI
  See what it means for humans and AI to truly “understand” the world, and how that understanding can deepen and expand.
• AI Applications Versus Manual Methods For Literature Retrieval: A Comparative Analysis
  How AI literature review tools compare with human researchers in identifying relevant sources, and where human oversight remains essential.

May 2026

• AI Edge: Q2
  AI is everywhere—but making it work is harder than it looks. These articles explore how AI can miss the human behind the data, why most agent initiatives stall before production, and how new approaches are making AI decisions more consistent and transparent.
• Explainable Multimodal Deep Learning Models for Variable-Length Sequences in Critically Ill Patients
  See how new biomedical informatics research on how data-driven methods can support health and clinical research.

April 2026

• Toward Experiential Training Program for AI Security and Privacy Practitioners
  From the 2025 CAE-CD Symposium, this highlight features Loyola research—including an extended abstract from Eric Chan-Tin—focused on experiential training to strengthen AI security, privacy, and real-world risk mitigation.
• LogosKG: Hardware-Optimized Scalable and Interpretable Knowledge Graph Retrieval
  Explore how LogosKG makes large-scale knowledge graph retrieval more scalable, efficient, and interpretable.
• Intro to Statistics in Criminal Justice and Criminology: A Practical Approach to Calculating, Using, and Interpreting Data
  A practical, plain-English intro to stats for criminal justice and criminology. Learn to calculate, interpret, and use data with confidence.
• Molecular Contrastive Learning with Graph Attention Network (MoCL-GAT) for Enhanced Molecular Representation
  A new graph-based approach to stronger molecular representations—see what improves and why it matters.
• DTA-GNN: A Toolkit for Constructing Target-Specific Drug–Target Affinity Datasets and Training Graph Neural Networks
  A hands-on toolkit for building drug–target affinity datasets and training GNN models—ready to plug into your workflow.
• Now's the Time: Computer Science Must Evolve to Emphasize Software and Systems Engineering with AI
  Why computer science education needs to shift from “learning to code” to engineering AI-enabled systems.
• Transforming FHIR into an OWL Knowledge Graph for Schema-Grounded Natural-Language Querying and Exploratory Data Analysis
  Turning complex clinical records into a knowledge graph so data is easier to search, explore, and analyze.

March 2026

• AI Edge: Q1
  Explore perspectives on institutional AI stewardship, emerging requirements for stablecoin issuers, and a new legal practice at the intersection of AI regulation and engineering.
• The Digital Scholar: Artificial Intelligence in Social Work Research
  Read how AI can support social work research while raising important questions about ethics and responsible use.
• An Empirical Investigation of Pre-Trained Deep Learning Model Reuse in the Scientific Process
  See how scientists are reusing pre-trained deep learning models across the research process.
• A Novel Patient-Specific Solution for Merkel Cell Carcinoma in High-Dose-Rate Brachytherapy
  A 3D-printed, patient-specific hand applicator made complex HDR brachytherapy practical—see the workflow and 1-year outcome.

February 2026

• Machine Learning Models Powered by Emergency Medical Services Data Enhance Stroke Triage in Prehospital Settings
  See how machine learning models using real-time EMS data can improve early stroke detection.
• From Moderation to Mediation: Can LLMs Serve as Mediators in Online Flame Wars?
  Can LLMs cool down online fights? See what happens when models try to mediate, not just moderate.
• Machine Learning Models Powered by Emergency Medical Services Data Enhance Stroke Triage in Prehospital Settings
  Faster stroke triage starts in the ambulance. See how EMS data + ML boosts early detection.
• Uncovering eHealth Engagement Patterns Through Latent Class Analysis and SHAP: A Data Mining Perspective on Telehealth Access
  Telehealth isn’t one-size-fits-all. Explore four digital engagement profiles—and what predicts access.
• Predicting Employees' Organizational Citizenship Behavior: A Comparative Study Using AI Techniques
  Can AI predict who goes the extra mile at work? This study shows how HR can use data-driven models to forecast and foster organizational citizenship behavior.

January 2026

• Toward AI Agent Behavior Research: A Behavioral Science Approach to Machine Decision-making in AI Interaction Platform
  See what happens to attention when only AI agents run the platform—and all optimize the same way.
• Scaling Biomedical Knowledge Graph Retrieval for Interpretable Reasoning: Applications to Clinical Diagnosis Prediction
  Turn billion-edge biomedical graphs into fast, interpretable diagnostic clues with LogosKG.
• AI Nudging and Decision Quality: Evidence from Randomized Experiments in Online Recommendation Setting
  How do AI “nudges” like Amazon’s Choice shape what we buy, how confident we feel, and how much we trust recommendations?

December 2025

• AI Edge: Q4
  A quick, practical scan of what’s working in AI right now—beyond the hype: smarter strategy, more reliable results, and audit-ready compliance.
• Predicting Failures of LLMs to Link Biomedical Ontology Terms to Identifiers: Evidence Across Models and Ontologies
  Why do LLMs miss ontology IDs? This study pinpoints the features that predict failure across models and ontologies.
• Healthcare AI's Unlearned Intermediaries
  Explore how deference to FDA processes can quietly block injured patients from recovering in AI device cases.
• AI in Medication Management and Drug Dosing in Critical Care
  See how AI is being used to optimize medication dosing in intensive care.
• Evaluating the Use of LLMs to Answer Patient-Facing Clinical Trial Questions
  Not all LLMs are equal: learn why rigorous testing is critical before letting AI talk to patients.
• 3D Shape Cues Affect Human and Artificial Recognition Systems Differently
  Do humans and neural networks really “see” the same object? Shape, texture, and 3D cues tell a different story.
• AI Foundation Model Automates Cryo-EM Structure Determination
  Meet Cryo-IEF: a foundation model that automates cryo-EM structure determination.
• Human at the Center: A Framework for Human-Driven AI Development
  When humans and AI work together, performance often drops. See how rethinking design can turn that around.

November 2025

• An Interactive Agentic HPC Tutor for Lesson Planning, Teaching, and Assessment
  From prompts to parallel programming. Meet an interactive, multi-agent HPC tutor built for classrooms and labs.
• Bridging the Gap: AI Technology Competence in Contemporary Legal Education
  Are tomorrow’s lawyers ready for AI—or already behind?

October 2025

• SysLLMatic: Large Language Models are Software System Optimizers
  From microbenchmarks to full apps, discover how SysLLMatic scales LLM code optimization to production-size software.
• Extracting O*NET Features from the NLx Corpus to Build Public Use Aggregate Labor Market Data
  A new natural language processing toolkit, funded by the Washington Center for Equitable Growth, extracts over 10 billion data points from 155 million job ads, transforming the chaos of online postings into a powerful, transparent, public-use dataset for analyzing the modern labor market.

September 2025

• The Association of Area Deprivation Index and Blood Pressure Control and Therapeutic Inertia Among Older Adults with Hypertension
  Discover how neighborhood deprivation shapes hypertension care in older adults—finding modest links to treatment inertia, even when blood pressure control itself shows no clear association.
• Prior Knowledge Shapes Success When Large Language Models Are Fine-Tuned for Biomedical Term Normalization
  Research uncovers a "reactive middle zone" where an LLM's existing knowledge determines if fine-tuning will succeed, revealing the secret to a model's ability to learn new biomedical associations.
• Artificial Intelligence in the Management of Leukemia
  AI can already help detect leukemia—can it also help prevent relapse?
• AI-Driven Systems in MTSS: Potential Opportunities and Challenges
  As schools face limited resources, AI offers both promise and pitfalls for implementing MTSS. Learn how educators and school psychologists can harness AI responsibly to support every learner.

August 2025

• AI Edge: Q3
  AI is transforming work through new tools, cloud infrastructure, and automation while highlighting that human relationships, collaboration, and creativity remain essential in the workplace.

July 2025

• Strengthening Scientific Integrity: Digital Forensics for Biomedical Research Imaging
  See how digital forensics is exposing manipulation in biomedical research images.
• Use of a Deep Learning Neural Network to Generate Bone Suppressed Images for Markerless Lung Tumor Tracking
  A deep learning model generates synthetic bone-suppressed images, enabling accurate markerless lung tumor tracking with a single X-ray exposure.
• Towards Human-Centered AI: The Role of Social Sciences
  The social sciences are essential, not only for analyzing the transformations brought by Artificial Intelligence but primarily for guiding its future development to be human-centered, focusing on its impact on interpersonal relations, social equality, and political regimes/public policy.

June 2025

• Climate Media Amidst Technopolitical Change
  Climate media research must move beyond outdated frameworks to address the complex, shifting realities of modern media and politics.

May 2025

• AI Edge: Q2
  From AI inspired by human movement to data-driven retail and the generational shift shaping corporate innovation, these three articles reveal how businesses are adapting—and thriving—in an intelligent, fast-changing world.
• Improving the Reproducibility of Deep Learning Software
  Reproducibility remains a critical challenge in deep learning. This study offers practical guidelines and a case study showing how researchers can close the gap between innovation and reliable replication.
• Preprocessing of Physician Notes by LLMs Improves Clinical Concept Extraction Without Information Loss
  LLMs can clean up messy physician notes—boosting concept extraction by 50% without losing key clinical info.

April 2025

• Artificial Intelligence and Academic Integrity: Legislate or Educate?
  Are AI rules enough—or do we need to teach better?
• Is AI Revolutionizing Climate Change Education?
  Learn how ChatGPT’s climate responses often promote passive learning, vague concern, and limited transparency—raising questions for educators and students alike.

February 2025

• AI Edge: Q1
  From workforce disruption to ethical dilemmas, AI is reshaping industries at an unprecedented pace. This issue explores the evolving job landscape, the urgent need to address AI bias, and the latest advancements in automation. Discover what these shifts mean for businesses, workers, and the future of technology-driven decision-making.

January 2025

• Ethical Concerns in Contemporary VR and Frameworks for Pursuing Responsible Use
  Discover the ethical gaps in Virtual Reality and how existing frameworks—from IRB to care ethics—can be synthesized into a new model for responsible VR use.
• The TRIPOD-LLM Reporting Guideline for Studies Using Large Language Models
  Bringing order to the chaos of LLM-in-healthcare studies—meet TRIPOD-LLM.
• In Moderation: Automation in the Digital Public Sphere
  Explore the Principle of Epistemic Resilience, a new approach to guiding digital platforms' content policies, with insights on how moderation and algorithm changes can foster healthy debate.

December 2024

• AI Edge: Q4
  From overcoming adoption challenges to reshaping the workforce, this issue delves into the transformative power of AI. Explore practical strategies for scaling solutions, ethical considerations for the future of work, and hiring trends shaping AI roles. Whether you're navigating implementation or keeping pace with industry demands, these insights offer a roadmap to thrive in the AI-driven era.
• Humans Outside the Loop
  AI may seem autonomous, but human choices shape it at every step—yet when things go wrong, current laws offer little recourse. This piece reveals why today’s liability systems fall short and what could change.
• Noise2Inverse for 3D Low-Dose Cone-Beam Computed Tomography
  Extending Noise2Inverse to 3D CBCT volumes, this deep learning approach reduces noise in low-dose scans, improving image quality without extra radiation exposure.
• AI and the Future of Digital Public Squares
  See how AI could reshape online public discourse by fostering more participatory dialogue while addressing risks to healthy digital communities.

November 2024

• The Routledge Handbook of Artificial Intelligence and Philanthropy
  A comprehensive guide to AI and philanthropy, this handbook explores ethical AI integration, philanthropy's role in shaping AI, and practical case studies, fostering collaboration and advancing research.

October 2024

• Revealing the Limits of Cybersecurity Law for Healthcare AI
  Healthcare tech holds sensitive data, but weak cybersecurity laws leave it exposed.
• Calculation of Alpha Particle Single-Event Spectra Using a Neural Network
  A neural network accurately predicts alpha-particle single-event spectra, streamlining microdosimetry calculations across diverse cell and tissue configurations.

September 2024

• AI Employment Interviews: Examining Limitations, Biases, and Perceptions
  What happens when an algorithm decides if you get the job?
• The New EU–US Data Protection Framework’s Implications for Healthcare
  The new U.S.-EU Data Privacy Framework aims to ease health data transfers—but gaps and stricter EU rules may still hinder cross-border healthcare and research.

August 2024

• AI Edge: Q3
  Explore how the latest advancements in AI and machine learning are transforming industries across the board. From aligning strategic investments to drive long-term profitability, to the cutting-edge innovations in financial services, and the revolutionary impact of Gen AI-powered Digital Knowledge Workers™ in manufacturing—discover how these technologies are enhancing decision-making, boosting productivity, and driving the next wave of business success.

May 2024

• AI Edge: Q2
  Topics include generative AI's impact on manufacturing design, maintenance, and quality control; the role of objective functions in defining AI's capabilities; and how large language models are transforming AI with autonomous agents for improved decision-making and customer experiences.
• Synthetic Realities and AI-Generated Contents
  When AI can fake anything, how do we trust what we see?
• Prescribing Exploitation
  AI-powered medical devices can quietly exploit patients, especially when reliance is unavoidable and data collection is constant. This piece argues for legal safeguards to protect users from excessive harm.

April 2024

• What’s Democratic About ‘Democratizing AI’?
  By invoking the language of democracy without considering how best to ensure collective governance, tech companies fail at offering a truly democratic AI.
• A Computer Vision Solution to Cross-cultural Food Image Classification and Nutrition Logging
  What if snapping a photo could give anyone, anywhere, an accurate nutrition log—no matter the cuisine?
• Advancing the Frontiers of Deep Learning for Low-Dose 3D Cone-Beam Computed Tomography Reconstruction
  A large-scale benchmark study reveals how deep learning can enhance low-dose 3D CT reconstructions, cutting radiation without compromising image quality.
• Low Dose CBCT Denoising Using a 3D U-Net
  A 3D deep learning approach effectively denoises low-dose CBCT scans, reducing radiation exposure while preserving image quality.

March 2024

• AI Edge: Q1
  Topics include: Retrieval-Augmented Generation (RAG) models and their potential for various industries; the transformative role of AI in supply chain management, and machine learning insights for non-machine learning engineers.
• Improving Clinical NLP Performance through Language Model-Generated Synthetic Clinical Data
  See how synthetic clinical data generated by language models could strengthen clinical NLP performance in healthcare research.