Scholars@Duke
Matthew M. Engelhard

Matthew M. Engelhard

Assistant Professor of Biostatistics & Bioinformatics
Biostatistics & Bioinformatics, Division of Translational Biomedical
2424 Erwin Rd, Durham, NC 27705

Scholarly Works - Conferences

Leveraging convenient wearable technology to assess adolescent sleep: physical and behavioral health correlates of single-channel sleep electroencephalogram in a community sample of youth.

Conference J Clin Sleep Med · July 13, 2026 STUDY OBJECTIVES: Sleep disturbances during adolescence heighten risk for physical and behavioral health problems, yet sleep physiological markers critical to health outcomes are rarely assessed in pediatric care. Wearable single-channel electroencephalogr ... Full text Link to item Cite

Abstract A137: Toward a Deeper Understanding of PREVENT for 10 Year Atherosclerotic Cardiovascular Risk: Subgroup Fairness and Predictive Value of Social Determinants of Health

Conference Stroke · February 2026 Background: The American Heart Association’s Predicting Risk of Cardiovascular Disease Events (PREVENT) model offers a modern, race-free approach to risk prediction from a large contem ... Full text Cite

Predicting Patient Preferences Using Large-Language-Models (LLM)

Conference PATIENT-PATIENT CENTERED OUTCOMES RESEARCH · September 2025 Link to item Cite

Predicting Patient Preferences Using Large-Language-Models (LLM)

Conference PATIENT-PATIENT CENTERED OUTCOMES RESEARCH · September 2025 Link to item Cite

Borrowing From the Future: Enhancing Early Risk Assessment through Contrastive Learning.

Conference Proc Mach Learn Res · August 2025 Risk assessments for a pediatric population are often conducted across multiple stages. For example, clinicians may evaluate risks prenatally, at birth, and during WellChild visits. While predictions at later stages typically achieve higher accuracy, it is ... Link to item Cite

Balancing Interpretability and Flexibility in Modeling Diagnostic Trajectories with an Embedded Neural Hawkes Process Model.

Conference Proc Mach Learn Res · August 2025 The Hawkes process (HP) is commonly used to model event sequences with self-reinforcing dynamics, including electronic health records (EHRs). Traditional HPs capture self-reinforcement via parametric impact functions that can be inspected to understand how ... Link to item Cite

Predicting Partially Observed Long-Term Outcomes with Adversarial Positive-Unlabeled Domain Adaptation.

Conference Proc Mach Learn Res · June 2025 Predicting long-term clinical outcomes often requires large-scale training data with sufficiently long follow-up. However, in electronic health records (EHR) data, long-term labels may not be available for contemporary patient cohorts. Given the dynamic na ... Link to item Cite

1017 Feasibility and Acceptability of SHEETS: A New Digital Intervention for Enhancing Sleep Regularity in Adolescents

Conference SLEEP · May 19, 2025 AbstractIntroductionAdolescence is associated with irregular sleep patterns, which in turn increases risk of onset and maintenance of ... Full text Cite

Synthetic Data Generation of Microbial Keratitis Slit Lamp Photos Using Limited Data

Conference INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE · June 2024 Link to item Cite

Common Event Tethering to Improve Prediction of Rare Clinical Events

Conference Proceedings of Machine Learning Research · January 1, 2024 Learning to predict rare medical events is difficult due to the inherent lack of signal in highly imbalanced datasets. Yet, oftentimes we also have access to surrogate or related outcomes that we believe share etiology or underlying risk factors with the e ... Cite

MALADE: Orchestration of LLM-powered Agents with Retrieval Augmented Generation for Pharmacovigilance

Conference Proceedings of Machine Learning Research · January 1, 2024 In the era of Large Language Models (LLMs), given their remarkable text understanding and generation abilities, there is an unprecedented opportunity to develop new, LLM-based methods for trustworthy medical knowledge synthesis, extraction, and summarizati ... Cite

On Neural Networks as Infinite Tree-Structured Probabilistic Graphical Models.

Conference Adv Neural Inf Process Syst · 2024 Deep neural networks (DNNs) lack the precise semantics and definitive probabilistic interpretation of probabilistic graphical models (PGMs). In this paper, we propose an innovative solution by constructing infinite tree-structured PGMs that correspond exac ... Link to item Cite

Sleep onset, duration, or regularity: which matters most for child adiposity outcomes?

Conference Int J Obes (Lond) · August 2022 BACKGROUND/OBJECTIVES: Sleep measures, such as duration and onset timing, are associated with adiposity outcomes among children. Recent research among adults has considered variability in sleep and wake onset times, with the Sleep Regularity Index (SRI) as ... Full text Link to item Cite

5.22 Improving the Predictive Value of Self-Report in Adult ADHD Diagnosis

Conference Journal of the American Academy of Child & Adolescent Psychiatry · October 2021 Full text Cite

SpanPredict: Extraction of Predictive Document Spans with Neural Attention

Conference Naacl Hlt 2021 2021 Conference of the North American Chapter of the Association for Computational Linguistics Human Language Technologies Proceedings of the Conference · January 1, 2021 In many natural language processing applications, identifying predictive text can be as important as the predictions themselves. When predicting medical diagnoses, for example, identifying predictive content in clinical notes not only enhances interpretabi ... Full text Cite

Exposure to Smoking Context Potentiates Habitual Motor Response

Conference NEUROPSYCHOPHARMACOLOGY · December 2020 Link to item Cite

Neural Conditional Event Time Models

Conference Proceedings of Machine Learning Research · January 1, 2020 Event time models predict occurrence times of an event of interest based on known features. Recent work has demonstrated that neural networks achieve state-of-the-art event time predictions in biomedical applications, where event time models are frequently ... Cite

A generic algorithm for sleep-wake cycle detection using unlabeled actigraphy data

Conference 2019 IEEE EMBS International Conference on Biomedical and Health Informatics Bhi 2019 Proceedings · May 1, 2019 One key component when analyzing actigraphy data for sleep studies is sleep-wake cycle detection. Most detection algorithms rely on accurate sleep diary labels to generate supervised classifiers, with parameters optimized for a particular dataset. However, ... Full text Cite

Predicting Smoking Events with a Time-Varying Semi-Parametric Hawkes Process Model

Conference Proceedings of Machine Learning Research · January 1, 2018 Health risks from cigarette smoking - the leading cause of preventable death in the United States - can be substantially reduced by quitting. Although most smokers are motivated to quit, the majority of quit attempts fail. A number of studies have explored ... Cite

Demonstrating the real-world significance of the mid-swing to heel strike part of the gait cycle using spectral features

Conference 2017 IEEE 14th International Conference on Wearable and Implantable Body Sensor Networks Bsn 2017 · May 30, 2017 Multiple sclerosis (MS) interrupts communication between the brain and other parts of the body causing functional deterioration. Gait impairment is a common finding in MS, one caused by several neurological symptoms. We perform an event-specific analysis t ... Full text Cite

Relationship between kernel density function estimates of gait time series and clinical data

Conference 2017 IEEE EMBS International Conference on Biomedical and Health Informatics Bhi 2017 · April 11, 2017 Multiple sclerosis (MS) is a neurological disorder which interrupts the communication between the brain and other parts of the body resulting in neurologic and physical and functional limitations. Gait deterioration is one of the most common problems and h ... Full text Cite

Adaptive symptom reporting for mobile patient-reported disability assessment

Conference 2016 IEEE Wireless Health Wh 2016 · December 1, 2016 Featured Publication Mobile symptom reporting apps can conveniently gather health-related information at low cost from day to day, fundamentally altering the relationship between patients, health data, and care providers. However, current mobile systems face a difficult trade- ... Full text Cite

Determining physiological significance of inertial gait features in multiple sclerosis

Conference Bsn 2016 13th Annual Body Sensor Networks Conference · July 18, 2016 Gait impairment in Multiple Sclerosis (MS) can result from imbalance, physical fatigue, weakness, and other symptoms. Walking speed is the primary measure of gait impairment used by clinical researchers, but inertial gait features from body-worn sensors ha ... Full text Cite

Correlations between inertial body sensor measures and clinical measures in multiple sclerosis

Conference Bodynets International Conference on Body Area Networks · January 1, 2015 Gait assessment using inertial body sensors is becoming popular as an outcome measure in multiple sclerosis (MS) research, supplementing clinical observations and patient-reported outcomes with precise, objective measures. Although numerous research report ... Full text Cite

Toward detection and monitoring of gait pathology using inertial sensors under rotation, scale, and offset invariant dynamic time warping

Conference Bodynets International Conference on Body Area Networks · January 1, 2015 Walking ability can be degraded by a number of pathologies, including movement disorders, stroke, and injury. Personal activity tracking devices gather inertial data needed to measure walking quality, but the required algorithmic methods are an active area ... Full text Cite