About

Bruce Ankenman is a Professor of Industrial Engineering and Management Sciences at Northwestern University, affiliated with the Segal Design Institute. His research focuses on developing simple-to-use, yet statistically powerful tools for the design and analysis of physical and simulation-based experiments, driven by his experience as a design engineer in the automotive parts industry where efficient data collection and analysis are crucial. In 2016, he co-developed a course called Designing Your Life at Northwestern, inspired by a Stanford course of the same name, which has been highly praised and is offered multiple times per year to undergraduate students from across all schools at Northwestern. In 2022, he co-founded the Northwestern Personal Development StudioLab, which offers classes, resources, and events aimed at enhancing undergraduate students' personal development.

Research topics

• Computer Science
• Political Science
• Artificial Intelligence
• Medical emergency
• Mathematics
• Engineering management
• Programming language
• Engineering
• Emergency medicine
• Pedagogy
• Medicine
• Computational science
• Simulation
• Medical education
• Algorithm
• Internal medicine
• Physical therapy
• Psychology

Selected publications

• User-centred prototyping solutions to solve adult critical care issues: a scoping review

  BMJ Health & Care Informatics · 2026-01-01

  articleOpen access

  BACKGROUND: How user-centred prototyping is carried out to solve adult critical care issues depends on the unique characteristics of this context. This review aimed to characterise prototyping in the context of critical care in terms of the types of prototypes developed, activities used to generate prototypes and settings in which prototypes were generated. METHODS: Four databases (PubMed, CINAHL, SCOPUS and IEEExplore) were searched for articles published from inception to 25 September 2025, in English, that involved prototyping to address issues in adult critical care. Two reviewers independently screened the search results to identify eligible articles and reviewed retained articles. RESULTS: 22 of 860 articles met the eligibility criteria. Role, look and feel, implementation and integration prototype types which combined two or more of these prototypes were identified. Prototypes addressing both role and look and feel were most common. 10 prototyping activities were reported, namely sketching, storyboarding, interactivity simulation, digitalising and adapting paper-based forms, rank ordering, building a functional device model, survey for item selection, card sorting, adapting a predeveloped high-tech prototype to a low-tech version, and revising existing workflow. Six of 22 articles reported multiple activities. Sketching was the most often used activity, and the in-person hospital setting was the most reported. CONCLUSIONS: Overall, there was a lack of reporting on the details of the prototyping processes. Such details could help future researchers anticipate the unique challenges of prototyping to develop solutions to solve adult critical care issues, learn from prior successful experiences and better plan strategies to address these challenges.

  PublisherOA PDFDOI

• User-centred prototyping solutions to solve adult critical care issues: a scoping review

  UNC Libraries · 2026-01-16

  articleOpen access

  BACKGROUND: How user-centred prototyping is carried out to solve adult critical care issues depends on the unique characteristics of this context. This review aimed to characterise prototyping in the context of critical care in terms of the types of prototypes developed, activities used to generate prototypes and settings in which prototypes were generated. METHODS: Four databases (PubMed, CINAHL, SCOPUS and IEEExplore) were searched for articles published from inception to 25 September 2025, in English, that involved prototyping to address issues in adult critical care. Two reviewers independently screened the search results to identify eligible articles and reviewed retained articles. RESULTS: 22 of 860 articles met the eligibility criteria. Role, look and feel, implementation and integration prototype types which combined two or more of these prototypes were identified. Prototypes addressing both role and look and feel were most common. 10 prototyping activities were reported, namely sketching, storyboarding, interactivity simulation, digitalising and adapting paper-based forms, rank ordering, building a functional device model, survey for item selection, card sorting, adapting a predeveloped high-tech prototype to a low-tech version, and revising existing workflow. Six of 22 articles reported multiple activities. Sketching was the most often used activity, and the in-person hospital setting was the most reported. CONCLUSIONS: Overall, there was a lack of reporting on the details of the prototyping processes. Such details could help future researchers anticipate the unique challenges of prototyping to develop solutions to solve adult critical care issues, learn from prior successful experiences and better plan strategies to address these challenges.

  PublisherDOI

• Inter-clinician diagnostic agreement of shock etiology: a multicenter observational study

  Health Information Science and Systems · 2026-01-12

  articleOpen access

  Abstract Purpose We sought to (1) quantify lack of inter-clinician diagnostic agreement of shock etiology and (2) predict patients without complete inter-clinician diagnostic agreement of shock etiology. Methods This multicenter retrospective, cohort study identified patients evaluated by two or more clinicians who documented a shock diagnosis from 2018 to 2023 across intensive care units (ICU) at 9 acute care hospitals. Shock etiology was abstracted using regular expression from clinician notes in the electronic health record then was made into a 9-dimensional vector representing 9 different shock etiologies. Inter-clinician diagnostic agreement of these vectors was calculated for each patient using Cosine Similarity Scores. Measure of agreement was based on cosine similarity of etiology vectors, not clinical adjudication. Patients without complete inter-clinician diagnostic agreement (Cosine Similarity Score < 1) were compared to patients with diagnostic agreement. Machine learning models were tested to predict patients without complete inter-clinician diagnostic agreement. Results Of 7302 patients, 1327 (18.2%) never had complete inter-clinician diagnostic agreement. Patients without diagnostic agreement were more often Black (20.5 vs 16.2%, p = 0.011), with more comorbidities (Elixhauser Comorbidity Index > 10; 39.1 vs 31.6%, p < 0.001), and Sequential Organ Failure Assessment (SOFA) score > 15 (12.1 vs 7.6%, p < 0.001). Patients without diagnostic agreement less frequently had improvements in SOFA scores between ICU days 0 and 4 (34.7 vs 41.9%, p < 0.001), and more often died in-hospital (41.5 vs. 27.6%, p < 0.001). Machine learning models that most accurately predicted patients without diagnostic agreement were logistic regression (Accuracy: 0.8597, F1-Score: 0.9117, AUC-ROC: 0.9257), random forest (Accuracy: 0.8658, F1-Score: 0.9201, AUC-ROC: 0.9255), and gradient boosting (Accuracy: 0.8515, F1-Score: 0.9138, AUC-ROC: 0.9227). Conclusion Patients without complete inter-clinician diagnostic agreement of shock etiology can be successfully predicted.

  PublisherOA PDFDOI

• 1012-P: Codesigning Solutions for CGM Use in Community Health Centers

  Diabetes · 2025-06-13

  article

  Introduction and Objective: Co-design is a research method to help develop solutions for complex healthcare problems. Continuous glucose monitoring (CGM) helps optimize diabetes mellitus (DM) care, yet uptake in community health centers (CHCs) is low. This study aimed to identify barriers to CGM use and propose user-centered solutions to increase CGM use in CHCs. Methods: DM team members involved in implementation or delivery of DM care in CHCs in the US were recruited to participate in 1-2 sessions via videoconferencing, which were audio recorded and transcribed. Session 1 elicited barriers and facilitators, guided by the Consolidated Framework for Implementation Research (CFIR) and System Engineering Initiative for Patient Safety (SEIPS) model. Session 2 (design) focused on ideating, categorizing, and prioritizing solutions. Transcripts were coded deductively using these constructs, and inductive codes were generated from data. A code matrix was used to organize coded data and develop themes. Solutions were categorized at the individual, clinic, or external level. Results: Twenty-two participants spanning roles (quality, clinical), across 5 states, participated. Analyses identified barriers, facilitators, and solutions to CGM initiation and data monitoring. CGM initiation was hindered by insurance requirements, inadequate staffing, and low confidence in managing patients with CGM. Remote monitoring of CGM data was limited by a perceived lack of patient access to necessary technologies (cell phones, internet) and limited integration of data into the EHR and workflows. Facilitators (dedicated expertise in CGM, utilization of CGM samples) and solutions were largely focused on initiation. Solutions were at the individual (trainings), clinic (partnerships with CGM companies, utilization of diabetes prevention program infrastructure), and external levels (broader coverage, receiver redesign). Conclusion: Multilevel solutions are needed to ensure that healthcare teams and patients benefit from DM technologies. Disclosure E.L. Lam: None. M.E. Wolter: None. A. Pack: Consultant; Gilead Sciences, Inc. J. Gacki-Smith: None. S.E. DeLacey: None. S. Agarwal: Research Support; Dexcom, Inc. B. Ankenman: None. D.W. Gatchell: None. A. Berry: None. C.T. Schaefer: None. M.J. O'Brien: None. A. Wallia: Research Support; UnitedHealth Group. Funding NIH NIDDK (P30DK020541-4751); Chicago Center for Diabetes Translation Research (P30 DK092949)

  PublisherDOI

• Leveraging User-Centered Design and Usability and Skills Testing for a Novel Diabetes Survival Skills Toolkit

  Journal of the Endocrine Society · 2025-06-24 · 2 citations

  articleOpen access

  Purpose: With the expansion of telemedicine, patient-centered approaches for delivering diabetes mellitus (DM) self-care education in both in-person and remote settings are needed. A novel Diabetes Survival Skills Toolkit (Kit) (physical toolkit, website, paper guide) was developed, using a user-centered design approach. The aim of this study was to develop a hybrid protocol to assess the perceived usability of the Kit and the skills attainment of its users. Methods: Adults without prior exposure to DM self-care were recruited. User tests were conducted between January 2021 and July 2022. Initially, the usability of the website alone was tested. Then, usability and skills attainment tests were conducted with all 3 components delivered together. Usability was measured by the System Usability Scale (SUS) and skills attainment was measured thorough simulated insulin injection and lancing device use. Results: User tests (N = 43) were conducted remotely (27/43; 63%) and in-person (16/43; 37%). SUS scores were largely excellent (35%) or acceptable (47%). Users who completed skills attainment testing (N = 32) all successfully injected insulin with simulation supplies. However, SUS scores and skills attainment were poorly correlated: users with unacceptable SUS scores (4/32, 13%) successfully attained the tested skills, while 2 of the 3 users who did not demonstrate successful lancing device use had excellent SUS scores. Conclusion: Hybrid user testing of a multi-component Kit to teach DM survival skills showed high skills attainment among adult users new to DM self-care. Pairing usability and skills attainment testing can help optimize the design of DM education interventions.

  PublisherOA PDFDOI

• Employing user-centered design and education sciences to inform training of diabetes survival skills

  Journal of Clinical & Translational Endocrinology · 2024-08-07 · 2 citations

  articleOpen access

  Background: Patients newly diagnosed with diabetes mellitus (diabetes), who require insulin must acquire diabetes "survival" skills prior to discharge home. COVID-19 revealed considerable limitations of traditional in-person, time-intensive delivery of diabetes education and survival skills training (diabetes survival skills training). Furthermore, diabetes survival skills training has not been designed to meet the specific learning needs of patients with diabetes and their caregivers, particularly if delivered by telehealth. The objective of the study was to identify and understand the needs of users (patients newly prescribed insulin and their caregivers) to inform the design of a diabetes survival skills training, specifically for telehealth delivery, through the application of user-centered design and adult learning and education principles. Methods: Users included patients newly prescribed insulin, their caregivers, and laypersons without diabetes. In semi-structured interviews, users were asked about experienced or perceived challenges in learning diabetes survival skills. Interviews were audio-recorded and transcribed. Investigators performed iterative rounds of coding of interview transcripts utilizing a constant comparative method to identify themes describing the dominant challenges users experienced. Themes were then mapped to adult learning and education principles to identify novel educational design solutions that can be applied to telehealth-based learning. Results: We interviewed 18 users: patients (N = 6, 33 %), caregivers (N = 4, 22 %), and laypersons (N = 8, 44 %). Users consistently described challenges in understanding diabetes survival skills while hospitalized; in preparing needed supplies to execute diabetes survival skills; and in executing diabetes survival skills at home. The challenges mapped to three educational strategies: (1) spiral learning; (2) repetitive goal directed practice and feedback, which have the potential to translate into design solutions supporting remote/virtual learning; and (3) form fits function organizer, which supports safe organization and use of supplies to execute diabetes survival skills independently. Conclusion: Learning complex tasks, such as diabetes survival skills, requires time, repetition, and continued support. The combination of a user-centered design approach to uncover learning needs as well as identification of relevant adult learning and education principles could inform the design of more user-centered, feasible, effective, and sustainable diabetes survival skills training for telehealth delivery.

  PublisherDOI

• SAT116 Utilizing User-centered Design And Usability And Skills Testing For Remote Diabetes Survival Skills Training

  Journal of the Endocrine Society · 2023-10-01

  articleOpen access

  Abstract Disclosure: S.J. Freeman: None. B. Radonski: None. L. Lecka: Employee; Self; Doximity. Stock Owner; Self; Doximity. K. Davis: None. G. Prince: None. K. Carthy: None. J.J. Seley: Speaker; Self; Lifescan Diabetes Institute. J. Song: None. J. Lee: None. S.C. Bailey: Consulting Fee; Self; Merck, Lundbeck, Sanofi-Aventis, Pfizer, Inc., Luto, University of Westminster, Gilead. Grant Recipient; Self; Merck, Eli Lilly & Company, Pfizer, Inc., Lundbeck, Gordon and Betty Moore Foundation, National Institutes of Health, Gilead. R. Khorzad: None. D. Gatchell: None. B. Ankenman: None. D.R. Lewis: Grant Recipient; Self; Pfizer, Inc., Spencer Foundation, National Institutes of Health. J. Holl: None. A. Wallia: Consulting Fee; Self; Eli Lilly & Company. Grant Recipient; Self; Novo Nordisk. Research Investigator; Self; UnitedHealth Group, Eli Lilly & Company. Patient-centered approaches for teaching diabetes mellitus (DM) survival skills are essential. Furthermore, in the peri-COVID era, interventions also need to be amenable to remote care delivery. User-Centered design (UCD) including usability testing is a key strategy to optimize adoption and engagement of interventions. We developed a Diabetes Survival Skills Toolkit (website, paper guide, and a physical Kit with simulation supplies) using UCD (> 50 sessions), followed by administration of system usability surveys (SUS) (scored as unacceptable, acceptable, or excellent) and, in a subset, additional skills testing. Skills testing included simulated blood glucose checks and insulin administration, conducted by 2 trained observers. Forty-three participants with no prior history of DM were recruited between 01/2021-07/2022 to independently learn survival skills using different Toolkit components [website only (N=11), Kit + paper guide (N=28), and Kit + website (N=4)]. Purposive sampling for age and highest education level resulted in 33% being ≥ 65 years and 35% having < 4-year degree. Overall, SUS scores were deemed excellent (N=15/43 [35%]) or acceptable (N=20/43 [47%]). Unacceptable scores were noted in 8/43 (19%) [4 website only (all > 4-year degree) and 4 Kit + paper guide (3 of 4 > 65 years, all < 4-year degree)]. Use of the website alone resulted in a higher rate of unacceptable SUS scores (37%) compared to use of the Kit with either the paper guide or website (13%). SUS-score category was not associated with age (82% acceptable/excellent among <45 years, 86% among 45-64 years, and 79% among >=65 years; Fishers’ p=1.00) nor highest education level (80% acceptable/excellent among <4-year degree and 82% among >=4-year degree; Fisher’s p=0.69). Participants who completed skills testing (N= 28 Kit + paper guide, 4 Kit + website), regardless of their SUS score, all correctly demonstrated the ability to inject insulin with simulation supplies. However, 4/32 (13%) (all SUS scores acceptable/excellent) were unable to navigate all steps independently and 9/32 (28%) (2 SUS unacceptable) did not use the recommended instructional pathway. All 4 participants (3 with > age 65 and < 4-year degree) who completed skills testing but had unacceptable SUS scores still correctly demonstrated the ability to measure blood glucose and inject insulin. In conclusion, a Survival Skills Toolkit, resulted in excellent rates of successful survival skills performance when tested with laypersons of diverse ages and education levels. Subjective usability (SUS scores) did differ among users of different Toolkit components; however, they did not align with actual skill performance. Design preferences and usability tests as well as subsequent skills testing are critical to optimally design tools for diabetes survival skills training. Presentation: Saturday, June 17, 2023

  PublisherDOI

• Improving Prehospital Stroke Diagnosis Using Natural Language Processing of Paramedic Reports

  Stroke · 2021 · 31 citations

  • Medicine
  • Emergency medicine
  • Physical therapy

  Background and Purpose: Accurate prehospital diagnosis of stroke by emergency medical services (EMS) can increase treatments rates, mitigate disability, and reduce stroke deaths. We aimed to develop a model that utilizes natural language processing of EMS reports and machine learning to improve prehospital stroke identification. Methods: We conducted a retrospective study of patients transported by the Chicago EMS to 17 regional primary and comprehensive stroke centers. Patients who were suspected of stroke by the EMS or had hospital-diagnosed stroke were included in our cohort. Text within EMS reports were converted to unigram features, which were given as input to a support-vector machine classifier that was trained on 70% of the cohort and tested on the remaining 30%. Outcomes included final diagnosis of stroke versus nonstroke, large vessel occlusion, severe stroke (National Institutes of Health Stroke Scale score >5), and comprehensive stroke center-eligible stroke (large vessel occlusion or hemorrhagic stroke). Results: Of 965 patients, 580 (60%) had confirmed acute stroke. In a test set of 289 patients, the text-based model predicted stroke nominally better than models based on the Cincinnati Prehospital Stroke Scale (c-statistic: 0.73 versus 0.67, P=0.165) and was superior to the 3-Item Stroke Scale (c-statistic: 0.73 versus 0.53, P<0.001) scores. Improvements in discrimination were also observed for the other outcomes. Conclusions: We derived a model that utilizes clinical text from paramedic reports to identify stroke. Our results require validation but have the potential of improving prehospital routing protocols.

  PublisherOA PDFDOI

• Abstract P225: Improving Prehospital Stroke Identification Using Natural Language Processing of Paramedic Reports

  Stroke · 2021-03-01

  article

  Introduction: Early identification of stroke by emergency medical services (EMS) providers in the prehospital setting is associated with increased treatment rates, improved functional outcomes, and reduced mortality. We hypothesize that a predictive model utilizing machine learning and natural language processing (NLP) techniques can be developed to analyze EMS run reports to identify stroke patients accurately. Methods: We analyzed EMS data from the Chicago Fire Department matched with inpatient data on confirmed and suspected strokes from 17 Chicago hospitals in the Get With The Guidelines-Stroke (GWTG-Stroke) registry from 11/28/2018 to 5/31/2019. Using features derived from paramedic notes, we developed a support vector machine classifier to predict the following categories: any stroke, AIS-LVO, severe stroke (NIHSS&gt;5), and CSC-eligible stroke (AIS-LVO or ICH/SAH). Individuals were randomly assigned into model derivation (70%) and validation cohorts (30%). C-statistics were used to evaluate discrimination of the classifier for stroke categories. Results: A total of 965 patients were included for analysis. In a validation cohort of 289 patients, the text-based model predicted stroke better than models trained using the Cincinnati Prehospital Stroke Scale (CPSS, c-statistic: 0.73 vs. 0.67, P=0.165) and the 3-Item Stroke Scale (3I-SS, c-statistic: 0.73 vs. 0.53, P &lt;0.001) scores. The text-based model also demonstrated improved performance over the CPSS and 3I-SS models in discriminating patients with other stroke categories (Table 1). Conclusion: We derived a predictive model using clinical text from paramedic reports that has superior performance to existing prehospital clinical screening tools to identify stroke in the prehospital setting. Future studies can evaluate the implementation of an NLP-based decision tool to assist in prehospital stroke evaluation and destination decision-making.

  PublisherDOI

• Employing User-Centered Design and Learning Science Theory to Enhance Remote Delivery of Diabetes Education and Survival Skills at Hospital Discharge

  Journal of the Endocrine Society · 2021-05-01 · 2 citations

  articleOpen access

  Abstract Learning diabetes mellitus (DM) survival skills is critically important, especially for those newly diagnosed upon discharge. COVID-19 has created new educational challenges, as DM self-management education and support is difficult to deliver remotely and can be time intensive. Content and format have not been re-designed for remote delivery; however, learning sciences research can help us create effective remote education strategies. We conducted interviews with users to identify critical needs in assuming immediate DM self-care at discharge from the hospital. We then mapped these user needs to relevant learning science theories to inform potential re-designs for remote delivery of DM education and survival skills at discharge. We conducted 12 semi-structured interviews with “users,” which included 18 participants (8 minority; 6&amp;gt;65 years): patients newly diagnosed with DM at discharge (N=6 [33%]), their caregivers (N=4 [22%]), and laypersons new to DM (N=8 [45%]). Users were asked about their discharge needs, laypersons about perceived needs. Three investigators performed iterative rounds of inductive coding of the transcripts (using MAXQDA software), utilizing a constant comparative method to identify codes describing dominant user needs. Learning science theory was applied to identify potential re-designs for remote delivery. Dominant user needs during hospitalization included being overwhelmed with DM self-care information (6/12 sessions) and difficulty organizing self-care equipment (5/12 sessions). Dominant user needs at home included remembering DM self-care steps (6/12 sessions), understanding correct insulin dosing (9/12 sessions), feeling fearful injecting insulin (9/12 sessions), with some noting difficulty in tracking glucose (4/12 sessions) and confusing insulin types (4/12 sessions). When learning science theory was applied, analysis mapped to three discrete educational strategies, most dominant of which is the spiral design approach—cycles of teaching the same topic but with increasing complexity. This design follows the pre-teaching principle—curriculum-based conceptual overview of self-care. Self-care at home mapped to the need for segmented learning and goal directed practice and feedback, with the potential need for behavioral therapies to reduce fear. Learning sciences has demonstrated that learning complex procedures and concepts, such as DM self-care, requires time, repetition, and continued support. With short hospital stays and the complexity of learning DM self-care, patients cannot gain needed knowledge structures to organize the information received during hospitalization. This study suggests specific learning science strategies for the design of an effective remote delivery of DM education and skills.

  PublisherOA PDFDOI

Frequent coauthors

• Roth Elliot

  University of Wisconsin–Madison

  36 shared

• Phillip Jacob

  Shirley Ryan AbilityLab

  36 shared

• Barry L. Nelson

  Northwestern University

  29 shared

• Susan M. Sanchez

  Naval Postgraduate School

  13 shared

• Collin Erickson

  Northwestern University

  12 shared

• Jane L. Holl

  University of Chicago

  9 shared

• Mustafa H. Tongarlak

  Western University

  8 shared

• Stacy Benjamin

  Northwestern University

  8 shared