About

Bruce A. Craig is a Professor of Statistics at Purdue University and serves as the Director of Statistical Consulting. His research interests encompass Bayesian methodology, bioinformatics and biologically related disciplines such as genomics, nutrition, proteomics, and statistical genetics. He also focuses on computational methods for statistical inference, experimental design, genomics, Markov Chain Monte Carlo, modeling and model selection, spatial analysis, and statistics education. Dr. Craig has supervised numerous Ph.D. students and has been recognized with several awards, including the College of Science Outstanding Service Award in 2022, the College of Science Leadership Award in 2017, and fellowships with the American Association for the Advancement of Science and the American Statistical Association. His educational background includes a B.A. in Mathematics and Economics from Washington University, and both M.S. and Ph.D. degrees in Statistics from the University of Wisconsin-Madison. Dr. Craig's contributions to the field are reflected in his leadership roles, awards, and extensive research in statistical methodology and applications.

Research topics

• Medicine
• Food science
• Physical medicine and rehabilitation
• Psychiatry
• Biology
• Gerontology
• Environmental health
• Surgery
• Physical therapy
• Geography
• Biomedical engineering
• Demography
• Pathology
• Medical emergency

Selected publications

• Toxicities of Brain and Spinal Cord Photon and Proton Radiation in Pediatric Brain Tumors Assessed by Real-World Data (RWD)

  International Journal of Radiation Oncology*Biology*Physics · 2025-09-01

  article1st authorCorresponding
  PublisherDOI

• Impact of a Novel Hybrid Computer-Based Simulation Adult Dysphagia Training on the Confidence and Clinical Knowledge of Speech-Language Pathology Graduate Students

  Teaching and Learning in Communication Sciences & Disorders · 2025-10-01

  articleOpen access

  The purpose of this project was to examine the impact of a novel hybrid computer-based dysphagia simulation training on student confidence and clinical decision-making in the area of adult dysphagia. We developed a pilot SBT database, including computer-based simulations utilizing real-life actors who simulated standardized adult patients with dysphagia. Students completed a comprehensive bedside swallowing evaluation and analysis of a videofluoroscopic swallow study (VFSS) as part of the training. We evaluated confidence and clinical decision-making through the following outcome measures: student self-rated confidence levels using a self-efficacy survey and agreement levels with our gold standard rater on clinical evaluation forms. Data for self-rated confidence were collected at three time points and organized into three composite areas (Treatment, Evaluation, and Personal & Interpersonal Skills). Clinical knowledge was measured at two time points using clinical evaluation forms, including a cranial nerve exam form, a VFSS analysis form, an Overall Impressions and Recommendations form, and the Functional Oral Intake Scale (FOIS). A repeated measures design was utilized to compare outcomes between time points. Results showed that student confidence in the evaluation, treatment, and personal and interpersonal composite areas of the self-efficacy survey increased significantly. While clinical performance for FOIS ratings and VFSS analysis did not significantly improve, ratings for the cranial nerve exam and choice of recommendations improved significantly between the first and second simulations. These results suggest that the students benefited from completing this novel hybrid computer-based SBT in dysphagia and support the use of our evidence-based methods for the development of future training approaches.

  PublisherOA PDFDOI

• A Research/Engagement Study Maximizing Nutrition Education to Meet Child and Parent Dietary Quality and Food Security Needs

  Journal of Nutrition Education and Behavior · 2025-08-01

  article
  PublisherDOI

• Effects of the Head Lift and Recline Exercise Regimens on the Neuromuscular Control of Functional Swallowing in Older Adults: An Electromyography Study Revealing Potential Differential Mechanisms

  Journal of Speech Language and Hearing Research · 2025-07-11

  articleOpen access

  Purpose: The head lift exercise (HLE) and recline exercise (RE) are rehabilitative exercises that have been shown to elicit similar biomechanical swallowing changes in older adults. However, the neuromuscular mechanisms underlying these changes are unknown and could elucidate the physiological mechanisms these exercises target. Method: A randomized clinical trial with two arms—a 6-week HLE or RE regimen—was conducted. Data were collected on older adults pre- and posttreatment and at 6-week follow-up using videofluoroscopic swallow studies (VFSSs) and surface electromyography (sEMG). Results of the VFSS showing similar hyolaryngeal excursion gains post both regimens have been published. This follow-up study presents the sEMG data and includes data from 18 older adults (age range: 60–82 years; HLE = 9, RE = 9) from the pre- and posttreatment time points. sEMG activity measures (normalized mean amplitude, burst duration, and time to peak sEMG amplitude) were collected from submental muscles during standardized swallow tasks. Results: Normalized mean amplitude, burst duration, and time to peak amplitude during swallowing did not significantly change posttreatment for either group. Post hoc correlational analysis of percent change between outcome measures from pre- to posttreatment revealed a strong negative relationship between normalized mean amplitude and time to peak amplitude for liquids ( r = −.926, p = .0001) and pudding ( r = −.901, p = .0001), indicating that participants who required greater levels of muscular contraction to functionally swallow posttreatment may need less time to reach that contraction level and vice versa. Conclusions: In combination with the Fujiki, Oliver, Malandraki, et al. (2019) results, these findings support that older adults show improvements in anterior and superior hyoid excursion post both HLE and RE without the need for greater submental musculature contraction. The inverse relationship identified post hoc between the percent change of amplitude and time to peak amplitude may indicate different neuromuscular mechanisms for biomechanical gains and needs further exploration for future personalized treatments.

  PublisherOA PDFDOI

• <scp>4D</scp> flow <scp>MRI</scp> velocity uncertainty quantification

  Magnetic Resonance in Medicine · 2024-09-13 · 4 citations

  articleOpen access

  PURPOSE: An automatic method is presented for estimating 4D flow MRI velocity measurement uncertainty in each voxel. The velocity distance (VD) metric, a statistical distance between the measured velocity and local error distribution, is introduced as a novel measure of 4D flow MRI velocity measurement quality. METHODS: The method uses mass conservation to assess the local velocity error variance and the standardized difference of means (SDM) velocity to estimate the velocity error correlations. VD is evaluated as the Mahalanobis distance between the local velocity measurement and the local error distribution. The uncertainty model is validated synthetically and tested in vitro under different flow resolutions and noise levels. The VD's application is demonstrated on two in vivo thoracic vasculature 4D flow datasets. RESULTS: Synthetic results show the proposed uncertainty quantification method is sensitive to aliased regions across various velocity-to-noise ratios and assesses velocity error correlations in four- and six-point acquisitions with correlation errors at or under 3.2%. In vitro results demonstrate the method's sensitivity to spatial resolution, venc settings, partial volume effects, and phase wrapping error sources. Applying VD to assess in vivo 4D flow MRI in the aorta demonstrates the expected increase in measured velocity quality with contrast administration and systolic flow. CONCLUSION: The proposed 4D flow MRI uncertainty quantification method assesses velocity measurement error owing to sources including noise, intravoxel phase dispersion, and velocity aliasing. This method enables rigorous comparison of 4D flow MRI datasets obtained in longitudinal studies, across patient populations, and with different MRI systems.

  PublisherOA PDFDOI

• Is Parent Diet Quality Related to Child Diet Quality? A Cross-Sectional Pilot Study Among Low-Income Families From Indiana

  Current Developments in Nutrition · 2024-06-29

  articleOpen access

  Objectives: Evidence has shown inconclusive relationships between parent and child diets, including diet quality, despite an expected link. The objective of this pilot study was to determine whether parent diet quality predicts child diet quality among low-income families in Indiana, United States. Methods: Baseline data collected from April 2023 to January 2024 from 49 paired parents 26-75 years and their children 5-17 years (one parent and one child) from low-income households throughout Indiana was utilized for this secondary analysis. The parent completed the U.S. Household Food Security Survey Module and parent and child each completed one to four Automated Self-Administered 24-hour dietary recalls (86% of the parent-child pairs reported intakes on the same day). The Healthy Eating Index (HEI-2015) derived using the simple method for individuals indicated diet quality for each parent and child score in a Pearson correlation coefficient test and parent HEI was the independent and child HEI the dependent variables in multiple linear regression models adjusted for food security status; child’s sex, ethnicity, and age; and parent’s age and smoking status. Results: The HEI-2015 total scores indicated poor dietary quality for both parents and children (50.9±11.6 and 51.5±12.1, respectively). The parent-child correlations for HEI-2015 that were statistically significant were weak or moderate (p< 0.05: r=0.4 for total HEI, total fruits, greens and beans, and saturated fats; r=0.3 for whole fruits, total vegetables, and fatty acids; and r=0.5 for whole grains). Parent diet quality was not significantly related to child diet quality for the total HEI-2015 score but was significantly related for components such that an increase in one HEI point in the parent score, on average, was related to a small increase of 0.4 points for greens and beans (p=0.01), whole grains (p=0.004), and saturated fats (p=0.01); and 0.3 points for fatty acids (p=0.02) in the child scores. Conclusions: A weak or moderate relationship existed between parent-child diets for several aspects of dietary quality. Factors other than parental dietary quality may potentially play a role in child dietary quality among low-income households. Funding Sources: United States Department of Agriculture, National Institute of Food and Agriculture USDA NIFA.

  PublisherOA PDFDOI

• The current landscape of academic statistical and data science collaboration units with examples

  Stat · 2024-07-26 · 2 citations

  articleOpen access

  The delivery of academic statistical collaboration resources can vary among types of institutions and across time. In particular, this variation might occur in the management of infrastructure and the business model, the staffing model and opportunities for staff development. In this manuscript, we present examples of these three themes in modern academic statistical collaboration units and describe key advantages and challenges.

  PublisherOA PDFDOI

• A Research/Engagement Study Maximizing Nutrition Education to Meet Child and Parent Dietary Quality and Food Security Needs

  Journal of Nutrition Education and Behavior · 2024-07-26

  article
  PublisherDOI

• Is Parent Diet Quality Related to Child Diet Quality? A Cross-Sectional Pilot Study Among Low-Income Families from Indiana

  Journal of Nutrition Education and Behavior · 2024-07-26

  article
  PublisherDOI

• Automatic 4D Flow MRI Segmentation Using the Standardized Difference of Means Velocity

  IEEE Transactions on Medical Imaging · 2023-03-02 · 9 citations

  articleOpen access

  We present a method to automatically segment 4D flow magnetic resonance imaging (MRI) by identifying net flow effects using the standardized difference of means (SDM) velocity. The SDM velocity quantifies the ratio between the net flow and observed flow pulsatility in each voxel. Vessel segmentation is performed using an F-test, identifying voxels with significantly higher SDM velocity values than background voxels. We compare the SDM segmentation algorithm against pseudo-complex difference (PCD) intensity segmentation of 4D flow measurements in in vitro cerebral aneurysm models and 10 in vitro Circle of Willis (CoW) datasets. We also compared the SDM algorithm to convolutional neural network (CNN) segmentation in 5 thoracic vasculature datasets. The in vitro flow phantom geometry is known, while the ground truth geometries for the CoW and thoracic aortas are derived from high-resolution time-of-flight (TOF) magnetic resonance angiography and manual segmentation, respectively. The SDM algorithm demonstrates greater robustness than PCD and CNN approaches and can be applied to 4D flow data from other vascular territories. The SDM to PCD comparison demonstrated an approximate 48% increase in sensitivity in vitro and 70% increase in the CoW, respectively; the SDM and CNN sensitivities were similar. The vessel surface derived from the SDM method was 46% closer to the in vitro surfaces and 72% closer to the in vitro TOF surfaces than the PCD approach. The SDM and CNN approaches both accurately identify vessel surfaces. The SDM algorithm is a repeatable segmentation method, enabling reliable computation of hemodynamic metrics associated with cardiovascular disease.

  PublisherOA PDFDOI

Recent grants

• NIH Grant R03EY012254

  NIH · $112k · 2000

• NIH Grant R01AI048067

  NIH · $144k · 2004

Frequent coauthors

• Michael Weiner

  University of Indianapolis

  58 shared

• Laura P. Sands

  Virginia Tech

  51 shared

• Marc B. Rosenman

  Indiana University School of Medicine

  46 shared

• Huiping Xu

  Indiana University Health

  43 shared

• Joseph Thomas

  42 shared

• W. W. Campbell

  Purdue University West Lafayette

  29 shared

• Caroline Carney Doebbeling

  Healthwise

  23 shared

• Deborah W. Knapp

  Purdue University West Lafayette

  22 shared

Education

• B.A., Mathematics and Economics

  Washington University

  1989

• M.S., Statistics

  University of Wisconsin-Madison

  1991

• Ph.D., Statistics

  University of Wisconsin-Madison

  1996

Awards & honors

• College of Science, Outstanding Service Award, 2022
• College of Science Leadership Award, 2017
• College of Science Team Award, 2015
• Fellow of the American Association for the Advancement of Sc…
• College of Science Engagement Award, 2014