Shelby Willis

Shelby Willis

Auto-generated profile, based on publicly available information

Shelby Willis

Oregon Health & Science University

Research Focus & Metrics

Academic Career Length

1Year (2026-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2026 in The Pacific Northwest Journal of Surgery
Verified
Journal Article
Open access
PDF file
PJPhillip D Jenkins
Phillip D Jenkins
SWShelby Willis
Shelby Willis
VAVictor Andujo
Victor Andujo
TLTiffany Lian
Tiffany Lian
RTRuchi Thanawala
Ruchi Thanawala
MKMichael R Kolesnikov
Michael R Kolesnikov
Introduction: Blunt thoracic aortic injury (BTAI) is a rare but lethal trauma mechanism. Mortality prediction has traditionally relied on logistic regression (LR), yet machine learning (ML) methods may better capture complex nonlinear relationships. Methods: Using the Aortic Trauma Foundation registry, we compared LR, extreme gradient boosting (XGB), and multilayer perceptron (MLP) for two tasks: (1) in-hospital mortality prediction and (2) classification of management strategy, open surgical repair (OSR), thoracic endovascular aortic repair (TEVAR), or medical management alone (MMA). Explanatory LR models identified independent predictors; predictive models targeted ~80% sensitivity in test sets. Performance metrics included area under the receiver operating characteristic curve (AUC), average precision, accuracy, F1 score, precision, recall, and Brier score. Results: Explanatory LR for mortality achieved excellent discrimination (AUC = 0.961), with higher Injury Severity Score, age, and ICU length of stay predicting increased mortality, and conversely, higher Glasgow Coma Scale, pH, and hospital stay protective. Predictive LR and XGB performed similarly (AUCs = 0.849 and 0.855; Brier = 0.083 and 0.101), both outperforming MLP (AUC = 0.757). For management strategy, explanatory LR identified limited predictors (e.g., ISS for OSR; age, heart rate, platelet count, and ISS for TEVAR). Predictive performance was modest: XGB (accuracy = 0.690; macro-AUC = 0.648) slightly outperformed MLP (accuracy = 0.667; macro-AUC = 0.565). Conclusions: Well-specified LR matched XGB for mortality prediction in BTAI, offering interpretability with high accuracy. Management strategy prediction remained limited across models, likely reflecting unmeasured institutional and procedural factors. Enriching datasets with imaging and practice-pattern variables may improve operative decision modeling.

9 September 2026 in Journal of Surgical Education
Verified
Journal Article
Closed access
SWShelby Willis
Shelby Willis
PJPhillip D. Jenkins
Phillip D. Jenkins
SBSteven Bedrick
Steven Bedrick
JJJonathan Jesneck
Jonathan Jesneck
JDJulie Doberne
Julie Doberne
RTRuchi M. Thanawala
Ruchi M. Thanawala

13 January 2026 in JMIR Perioperative Medicine
Verified
Journal Article
Open access
PJPhillip Jenkins
Phillip Jenkins
OXObert Xu
Obert Xu
SWShelby Willis
Shelby Willis
EKEthan Kimball
Ethan Kimball
RTRuchi Thanawala
Ruchi Thanawala
JDJulie Doberne
Julie Doberne

13 January 2026 in JMIR Preprints
Verified
Preprint
Open access
PJPhillip Jenkins
Phillip Jenkins
OXObert Xu
Obert Xu
SWShelby Willis
Shelby Willis
EKEthan Kimball
Ethan Kimball
RTRuchi Thanawala
Ruchi Thanawala
JDJulie Doberne
Julie Doberne
BACKGROUND Acute appendicitis is a common disease process typically requiring surgery, yet the workflow linking diagnostic imaging to surgical consultation varies substantially across emergency departments. Delays between imaging completion and consult acquisition may prolong care and contribute to avoidable clinical and operational inefficiencies. OBJECTIVE This study quantified real-world consultation delays for patients with radiology-confirmed appendicitis and evaluated the cumulative time impact on emergency department workflow. METHODS We performed a retrospective observational study of emergency department encounters from January 1, 2020, through December 31, 2025, in which abdominal imaging was obtained to evaluate possible appendicitis. All radiology impressions were manually adjudicated and classified as positive, indeterminate, or negative. The primary timing measure was the interval from imaging completion to surgical consultation order. Mann–Whitney U tests compared delays across imaging modalities and age groups. Logistic regression assessed predictors of prolonged delay (>30 minutes). RESULTS Among 1,422 encounters, 566 were classified as radiology-positive appendicitis. Surgical teams evaluated 565 of these patients (99.8 percent), demonstrating that positive radiology findings nearly always resulted in surgical involvement regardless of documentation of a formal consult order. Among 524 radiology-positive encounters with complete timestamps in a predefined plausible window (−60 to +360 minutes), the median time from imaging completion to consultation was 30.8 minutes (IQR 17.8–48.5). Delays were longer for CT than ultrasound (median 34.9 vs 21.2 minutes; p < 0.0001). CT was associated with prolonged delay (OR 2.29; 95% CI 1.08–4.86), while age group was not. Across a typical year, cumulative waiting time totaled approximately 58 patient-hours. CONCLUSIONS Radiology-confirmed appendicitis reliably triggered surgical evaluation, yet meaningful delays remained. Standardizing and automating consult activation for clear radiologic diagnoses may reduce avoidable workflow variation and improve the timeliness of surgical care.

30 September 2026 in Annals of Vascular Surgery
Verified
Journal Article
Closed access
SWShelby Willis
Shelby Willis
PJPhillip D. Jenkins
Phillip D. Jenkins
VAVictor Andujo
Victor Andujo
TLTiffany Lian
Tiffany Lian
SMSheila Markwardt
Sheila Markwardt
JDJulie Doberne
Julie Doberne