BEGIN:VCALENDAR
VERSION:2.0
X-WR-CALNAME:event866a5e698eb5ce0
X-WR-CALDESC:Event Calendar
METHOD:PUBLISH
CALSCALE:GREGORIAN
PRODID:-//Sched.com Symposium//EN
X-WR-TIMEZONE:UTC
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T160000Z
DTEND:20261009T170000Z
SUMMARY:Breakfast
DESCRIPTION:\n
CATEGORIES:
LOCATION:Wieboldt 704\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:bb8f5c3528017b0331f7b1d79b9bab6f
URL:http://event866a5e698eb5ce0.sched.com/event/bb8f5c3528017b0331f7b1d79b9bab6f
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T160000Z
DTEND:20261009T200000Z
SUMMARY:Professional Headshots (Complimentary) - Dynamic Photo Systems
DESCRIPTION:\n
CATEGORIES:
LOCATION:Wieboldt 7th Floor Lounge\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:9fab8c6f556ba5880b094a565b0744ce
URL:http://event866a5e698eb5ce0.sched.com/event/9fab8c6f556ba5880b094a565b0744ce
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T170000Z
DTEND:20261009T174000Z
SUMMARY:Betting Big\, Starting Small: Quantum Computing's Capital and Its Classifiers
DESCRIPTION:Quantum computing sits at a crossroads: extraordinary capital investment on one side\, and modest near-term performance on the other. This session presents two complementary analyses that frame that tension.\n\n\nChoice 1 — The Cost of Belief: surveys the financial landscape of quantum computing — spending by Google\, IBM\, and Microsoft\, federal funding commitments\, and venture capital trends — concluding that transformative potential remains contingent on solving error correction at scale before commercial advantage arrives.\n\n\nChoice 2 — The Cost of Proof: grounds that promise in a concrete test\, benchmarking two quantum classifiers (VQC and a Qiskit-based QNN) against a classical logistic regression baseline on a real healthcare dataset — where neither quantum model yet beats the classical benchmark\, offering a grounded starting point for quantum curriculum development in Northwestern's MSDS program.\n\n
CATEGORIES:AI & DATA SCIENCE
LOCATION:Wieboldt 506\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:44d5b14da8bddd9f49f13fa03138e41a
URL:http://event866a5e698eb5ce0.sched.com/event/44d5b14da8bddd9f49f13fa03138e41a
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T170000Z
DTEND:20261009T174000Z
SUMMARY:Crisis\, Coordination\, & Adaptation: Organizational Theory in Action During Covid-19
DESCRIPTION:Public organizational theories often emphasize distinct models—hierarchy\, markets\, networks\, or human-centered management—as if any any given framework can adequately explain or guide effective government action. This session challenges that assumption through a practitioner-focused case study of Illinois’ COVID-19 response\, told from the presenter’s firsthand experience inside the governor’s office. It traces how a small\, siloed\, and decentralized organization rapidly reconfigured itself under crisis conditions into a highly coordinated and adaptive system. With limited initial information and no existing infrastructure for a statewide public health emergency\, leaders relied on improvisation\, cross-functional teamwork\, and ad hoc data systems to support decision-making. The case highlights how informal authority\, pre-existing relationships\, and practical problem-solving were often more important than formal structures\, enabling the rapid design and implementation of a large-scale testing strategy.\n\nAt its core\, the Illinois experience underscores that there is no “silver bullet” organizational model. In the public sector\, the actors and environment are often external to the leadership or needs\, creating what the author describes as a "patchwork quilt" of leadership\, management\, design\, and administration. Illinois serves as a powerful example of how organizations can become agile and effective under pressure\, not because they follow a single theory\, but because leaders draw from many tools and approaches to meet the demands of the moment.
CATEGORIES:CAREER & CLASSROOM
LOCATION:Wieboldt 517
SEQUENCE:0
UID:f160e0b049d1e4a90335227b91474a8f
URL:http://event866a5e698eb5ce0.sched.com/event/f160e0b049d1e4a90335227b91474a8f
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T170000Z
DTEND:20261009T174000Z
SUMMARY:Machine Learning for Early Prediction of Alzheimer's Disease Progression Using Multimodal ADNI Data
DESCRIPTION:Alzheimer’s Disease is projected to affect over 139 million people worldwide by 2050\, yet mostpatients are diagnosed after significant brain changes have already occurred. Early detectionexpands treatment options\, supports family planning\, and reduces caregiver burden\, butcurrent diagnostic pathways remain inconsistent\, expensive\, and frequently late.\nThis thesis evaluates whether machine learning applied to multimodal clinical data can improveearly prediction across the Alzheimer’s continuum. Using baseline data from approximately2\,100 participants in the Alzheimer’s Disease Neuroimaging Initiative (ADNI)\, we built andtested predictive models across three diagnostic stages: cognitively normal\, mild cognitiveimpairment (MCI)\, and dementia\, incorporating dozens of features spanning cognitiveassessments\, neuroimaging metrics\, and fluid biomarkers.\nA Random Forest classifier achieved high overall accuracy (approximately 89% in preliminaryanalyses)\, with about one third of MCI patients in the cohort progressing to dementia\,underscoring the clinical urgency of identifying this transitional group. The strongest predictorwas the Clinical Dementia Rating Sum of Boxes\, a measure of everyday functional decline\,followed by the Functional Activities Questionnaire and digital cognitive composites. Theseresults align with current clinical practice\, suggesting that machine learning can augment ratherthan replace clinician judgment.\nThe work also addresses real world data constraints. Initial class imbalance reduced sensitivityfor minority diagnostic groups (roughly 40–50% before rebalancing)\, and over threequarters ofcerebrospinal fluid biomarker entries were missing\, a common reality in clinical datasets thatany deployed model must accommodate.\nThis session will present our methodology\, key findings (with final metrics to be confirmed)\, andbroader implications for designing AI tools that perform reliably in clinical environments wheredata is incomplete\, imbalanced\, and variable.\n
CATEGORIES:HEALTHCARE AI AND DATA
LOCATION:Wieboldt 711\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:060ca716f57ccb40809d984b0b0032ac
URL:http://event866a5e698eb5ce0.sched.com/event/060ca716f57ccb40809d984b0b0032ac
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T170000Z
DTEND:20261009T174000Z
SUMMARY:Transfer Mobility: The Transfer Portal\, NIL\, & the Reshaping of College Football Careers
DESCRIPTION:College football has undergone a fundamental transformation in recent years due to two major policy changes: the introduction of the NCAA transfer portal and the emergence of Name\, Image\, and Likeness (NIL) compensation opportunities. Together\, these reforms have created a more open and dynamic marketplace in which student-athletes can transfer between institutions with greater freedom while also benefiting from new economic opportunities. This presentation examines how these changes have reshaped athlete mobility and career outcomes within NCAA Division I Football Bowl Subdivision (FBS) football.
CATEGORIES:SPORTS
LOCATION:Wieboldt 712
SEQUENCE:0
UID:dab0939311f6e0183a743fe27dc9acf1
URL:http://event866a5e698eb5ce0.sched.com/event/dab0939311f6e0183a743fe27dc9acf1
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T175000Z
DTEND:20261009T183000Z
SUMMARY:Community-Based Participatory Research: Tools & Techniques Using AI & Data
DESCRIPTION:This presentation examines the core principles of Community Based Participatory Research (CBPR) and demonstrates how artificial intelligence (AI) and data science methodologies can be integrated to support collaborative\, ethical\, and policy relevant research. CBPR provides a systematic framework for addressing complex social and policy challenges (e.g.\, education\, environmental conditions\, health\, and transportation) by actively involving community members and key stakeholders throughout the research process. More specifically\, centering the lived experiences and knowledge of these groups can strengthen the relevance and actionability of resulting policy recommendations. The addition of AI and data analytic tools further enhances the capacity to ensure that technological developments and data driven insights equitably serve diverse communities.\n\nUsing a case study format\, the presentation will guide attendees through strategies for engaging communities\, leveraging AI tools responsibly\, and analyzing data to generate policy insights that support local initiatives. The session adopts an interdisciplinary perspective\, drawing on concepts from economics\, ethics\, geography\, and public policy to illustrate how CBPR and AI can be combined to advance equitable and community driven research.
CATEGORIES:AI & DATA SCIENCE
LOCATION:Wieboldt 506\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:89cbd39e32a16426844f99124ec8b51c
URL:http://event866a5e698eb5ce0.sched.com/event/89cbd39e32a16426844f99124ec8b51c
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T175000Z
DTEND:20261009T183000Z
SUMMARY:Built to Compete: What You're Pretending Not To Know About Your Own Next Move
DESCRIPTION:Most of us aren’t stuck because we don’t know what to do next. We’re stuck because the next move requires us to leave something comfortable behind. This session explores what keeps us waiting\, how expectations and the need for certainty get in our way\, and what it looks like to stop waiting for permission and start taking ownership of what comes next.cy to illustrate how CBPR and AI can be combined to advance equitable and community driven research.
CATEGORIES:CAREER & CLASSROOM
LOCATION:Wieboldt 517
SEQUENCE:0
UID:a26868882061d80630efd669089dfa31
URL:http://event866a5e698eb5ce0.sched.com/event/a26868882061d80630efd669089dfa31
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T175000Z
DTEND:20261009T183000Z
SUMMARY:Comparative Accuracy of Reasoning v. Non Reasoning Models for CPT Coding Assignments
DESCRIPTION:Does ‘thinking’ make a language model a better medical coder? We ran three frontier LLMs (Claude Opus 4.5\, GPT-5.2\, Gemini 3) across orthopedic operative notes\, with and without reasoning.&nbsp\;Come see&nbsp\;the accuracy\, consistency\, cost tradeoffs of LLM&nbsp\;assisted&nbsp\;medical coding&nbsp\;against surgeon-defined benchmarks.&nbsp\;
CATEGORIES:HEALTHCARE AI AND DATA
LOCATION:Wieboldt 711\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:255fbd7bc3620f4a417744cfe6f2a628
URL:http://event866a5e698eb5ce0.sched.com/event/255fbd7bc3620f4a417744cfe6f2a628
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T184000Z
DTEND:20261009T192000Z
SUMMARY:AI Panel
DESCRIPTION:Retrieval-Augmented Generation (RAG) systems are rapidly being integrated into healthcare workflows\, including clinical decision support\, claims processing\, and patient engagement. However\, real-world deployments reveal a critical gap: systems that perform well in controlled environments often fail in production due to subtle yet high-impact risks. These include exposure of Protected Health Information (PHI)\, hallucinated or clinically unsafe outputs\, and violations of regulatory requirements such as HIPAA—failures that can directly affect patient safety\, institutional trust\, and legal accountability.\nThis presentation introduces a governance-first\, security-centric architecture for production-grade healthcare RAG pipelines. The proposed framework embeds policy-driven access controls\, token-level redaction\, and context-aware retrieval filtering to prevent sensitive data leakage. It further integrates hybrid validation layers that combine deterministic rule-based checks with model-assisted reasoning to detect hallucinations and enforce output reliability. End-to-end auditability and traceability are incorporated across ingestion\, retrieval\, and generation stages to support compliance and operational transparency.\nTo enhance robustness\, the approach incorporates differential privacy techniques and encryption strategies for data in transit and at rest\, along with failure-injection testing to simulate real-world risks such as schema drift\, prompt injection\, and data leakage. Experimental and simulated evaluations demonstrate that embedding governance and security controls directly into the RAG pipeline significantly reduces PHI exposure\, improves output reliability\, and strengthens compliance readiness.\nThis work presents a practical blueprint for building trustworthy healthcare AI systems\, enabling organizations to mitigate risk and accelerate responsible GenAI adoption in regulated environments.\n
CATEGORIES:AI & DATA SCIENCE
LOCATION:Wieboldt 506\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:bd664beb49f6113a123c5516d95ee8a9
URL:http://event866a5e698eb5ce0.sched.com/event/bd664beb49f6113a123c5516d95ee8a9
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T184000Z
DTEND:20261009T192000Z
SUMMARY:Teaching Storytelling in the Age of AI: Reimagining Media Literacy Through Documentary Filmmaking
DESCRIPTION:As artificial intelligence and algorithm-driven platforms rapidly reshape communication\, education\, and culture\, media literacy education must evolve beyond traditional approaches focused solely on content consumption and analysis. This session explores documentary storytelling as an interdisciplinary information design framework for developing critical thinking\, communication\, and digital literacy skills in contemporary learning environments.\n\nDrawing from graduate research within Northwestern University’s Information Design & Strategy program\, this presentation examines how storytelling-centered learning experiences can help students become more intentional creators\, collaborators\, and interpreters of information. Through examples from the Young & Filming initiative\, participants will explore how documentary production practices—including interviewing\, narrative construction\, visual communication\, and ethical storytelling—can strengthen literacy comprehension\, emotional intelligence\, and audience awareness.\n\nThe session will also address the growing role of AI within creative and educational ecosystems\, examining both the opportunities and tensions surrounding generative technologies in storytelling workflows. Rather than positioning AI as a replacement for human creativity\, this presentation advocates for a human-centered design approach that emphasizes empathy\, critical inquiry\, and authentic communication.\n\nBy combining principles from learning design\, media studies\, documentary filmmaking\, and digital strategy\, this session offers an interdisciplinary model for preparing learners and professionals to navigate increasingly automated and information-saturated environments.\n\n
CATEGORIES:CAREER & CLASSROOM
LOCATION:Wieboldt 517
SEQUENCE:0
UID:1fdc8ecdf6a995de5515474c818c4bb2
URL:http://event866a5e698eb5ce0.sched.com/event/1fdc8ecdf6a995de5515474c818c4bb2
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T184000Z
DTEND:20261009T192000Z
SUMMARY:When Healthcare AI Fails: Strengthening RAG Pipelines Against PHI Exposure\, Hallucinations\, and Compliance Risks
DESCRIPTION:Retrieval-Augmented Generation (RAG) systems are rapidly being integrated into healthcare workflows\, including clinical decision support\, claims processing\, and patient engagement. However\, real-world deployments reveal a critical gap: systems that perform well in controlled environments often fail in production due to subtle yet high-impact risks. These include exposure of Protected Health Information (PHI)\, hallucinated or clinically unsafe outputs\, and violations of regulatory requirements such as HIPAA—failures that can directly affect patient safety\, institutional trust\, and legal accountability.\nThis presentation introduces a governance-first\, security-centric architecture for production-grade healthcare RAG pipelines. The proposed framework embeds policy-driven access controls\, token-level redaction\, and context-aware retrieval filtering to prevent sensitive data leakage. It further integrates hybrid validation layers that combine deterministic rule-based checks with model-assisted reasoning to detect hallucinations and enforce output reliability. End-to-end auditability and traceability are incorporated across ingestion\, retrieval\, and generation stages to support compliance and operational transparency.\nTo enhance robustness\, the approach incorporates differential privacy techniques and encryption strategies for data in transit and at rest\, along with failure-injection testing to simulate real-world risks such as schema drift\, prompt injection\, and data leakage. Experimental and simulated evaluations demonstrate that embedding governance and security controls directly into the RAG pipeline significantly reduces PHI exposure\, improves output reliability\, and strengthens compliance readiness.\nThis work presents a practical blueprint for building trustworthy healthcare AI systems\, enabling organizations to mitigate risk and accelerate responsible GenAI adoption in regulated environments.\n
CATEGORIES:HEALTHCARE AI AND DATA
LOCATION:Wieboldt 711\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:02df7748c81114eaf97272c577c82931
URL:http://event866a5e698eb5ce0.sched.com/event/02df7748c81114eaf97272c577c82931
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T203000Z
DTEND:20261009T211000Z
SUMMARY:World Model Reinforcement Learning for Industrial Optimization
DESCRIPTION:This presentation explores how modern data science methods can be used to support industrial process optimization\, with a focus on applying world-model reinforcement learning to a fractionation-style process. Many industrial systems are nonlinear\, dynamic\, and highly interactive: a change to one controller or operating variable may influence temperatures\, pressures\, flows\, product quality\, and throughput over time. Because of this\, simple one-step prediction models may be useful for forecasting\, but they are often not enough to support practical operating guidance. The main concept presented here is the use of a learned world model (a data-driven simulator) that estimates how a process state may evolve after a proposed operating change.\nThe project demonstrates a general workflow for moving from prediction to decision support. First\, historical process data is used to train a target prediction model that estimates a key production or performance outcome. Next\, a separate dynamics model is trained to approximate short-term process response. This world model allows candidate operating moves to be tested through a short simulated rollout before the target outcome is evaluated. Reinforcement learning or policy optimization can then be used to search for bounded action patterns that may improve the predicted future state. This approach is valuable because it makes the recommendation process more realistic than static what-if analysis\, so the action is evaluated after the process has had time to respond.\nThe presentation also discusses broader architecture choices for industrial world models\, including feed-forward neural networks\, recurrent models\, temporal convolutional networks\, transformers\, physics-informed hybrids\, and probabilistic ensembles. Each architecture has tradeoffs in speed\, interpretability\, data requirements\, long-horizon accuracy\, and deployment complexity. Finally\, the discussion highlights opportunities across refining\, chemicals\, manufacturing\, utilities\, supply chain\, maintenance\, and other industrial domains. The central message is that world-model reinforcement learning should be viewed as a decision-support framework\, a way to screen scenarios\, improve understanding\, and guide expert review. This is not as an immediate replacement for engineering judgment or operational governance but is a tool to help enhance these skills.
CATEGORIES:AI & DATA SCIENCE
LOCATION:Wieboldt 506\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:173230c64531869e0f463ff344c32f80
URL:http://event866a5e698eb5ce0.sched.com/event/173230c64531869e0f463ff344c32f80
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T203000Z
DTEND:20261009T211000Z
SUMMARY:GenAI in Group Projects: Bridging Faculty Expectations & Student Realities
DESCRIPTION:Group projects remain one of the most effective ways for students to apply disciplinary knowledge to authentic\, collaborative challenges. The rapid adoption of large language models (LLMs) and other generative AI tools has introduced new opportunities and new questions about how students work together\, contribute expertise\, and demonstrate learning. \n\nJoin SPS instructors and graduate students for a conversation about navigating the use of genAI tools in group projects. Hear perspectives from both sides of the classroom about deciding where and when to use AI and how to collaborate with others during that process. \n\nIn addition to sharing real-life examples of what’s working for both instructors and students\, we’ll introduce a framework for assessing the collaboration of genAI and humans in your group project through the lenses of accuracy\, authenticity\, expertise\, and impact. Attendees will leave with ideas for setting expectations\, facilitating productive teamwork\, and assessing the contributions of both students and AI tools in group-based learning environments.
CATEGORIES:CAREER & CLASSROOM
LOCATION:Wieboldt 517
SEQUENCE:0
UID:3cb261a536ad70e22209a0dc77e3d738
URL:http://event866a5e698eb5ce0.sched.com/event/3cb261a536ad70e22209a0dc77e3d738
END:VEVENT
BEGIN:VEVENT
DTSTAMP:20260820T091941Z
DTSTART:20261009T203000Z
DTEND:20261009T211000Z
SUMMARY:Evaluating Sequencing and Tokenization Strategies for EHR Foundation Models
DESCRIPTION:Electronic Health Records (EHRs) provide a rich\, longitudinal view of patient health\, but their high dimensionality\, irregular temporal structure\, and heterogeneous data types present significant challenges for machine learning. Transformer-based foundation models offer a promising approach by representing patient histories as sequences of clinical events. However\, key design decisions\, such as how to represent clinical concepts\, encode continuous values\, and structure temporal information\, remain largely underexplored.\nThis presentation systematically evaluates how different ways of organizing and representing EHR data affect model performance on downstream clinical tasks. Patient histories are converted into sequences of events (diagnoses\, medications\, lab results\, etc.) that a model can process similarly to language. This study compares several key design choices\, including how detailed diagnosis codes should be\, how to represent numerical values such as lab results\, and how to best capture the timing and structure of healthcare encounters.\nBy quantifying the impact of each design decision\, this presentation provides practical guidance for constructing effective EHR foundation models and highlights representation design as a key driver of performance in healthcare AI.\n
CATEGORIES:HEALTHCARE AI AND DATA
LOCATION:Wieboldt 711\, Kellogg Wieboldt Hall\, 340 E Superior St\, Chicago\, IL 60611
SEQUENCE:0
UID:e982ffa1785e0efd6abc1ece642bad0e
URL:http://event866a5e698eb5ce0.sched.com/event/e982ffa1785e0efd6abc1ece642bad0e
END:VEVENT
END:VCALENDAR