Dan Ricketts

Advances in artificial intelligence (AI) have enabled unprecedented capabilities, yet innovation teams struggle when envisioning AI concepts. Data science teams think of innovations users do not want, while domain experts think of innovations that cannot be built. A lack of effective ideation seems to be a breakdown point. How might multidisciplinary teams identify buildable and desirable use cases? This paper presents a first hand account of ideating AI concepts to improve critical care medicine. As a team of data scientists, clinicians, and HCI researchers, we conducted a series of design workshops to explore more effective approaches to AI concept ideation and problem formulation. We detail our process, the challenges we encountered, and practices and artifacts that proved effective. We discuss the research implications for improved collaboration and stakeholder engagement, and discuss the role HCI might play in reducing the high failure rate experienced in AI innovation.

Background: Acute kidney injury (AKI), the decline of kidney excretory function, occurs in up to 18% of hospitalized admissions. Progression of AKI may lead to irreversible kidney damage. Methods: This retrospective cohort study includes adult patients admitted to a non-intensive care unit at the University of Pittsburgh Medical Center (UPMC) (n = 46,815) and University of Florida Health (UFH) (n = 127,202). We developed and compared deep learning and conventional machine learning models to predict progression to Stage 2 or higher AKI within the next 48 hours. We trained local models for each site (UFH Model trained on UFH, UPMC Model trained on UPMC) and a separate model with a development cohort of patients from both sites (UFH-UPMC Model). We internally and externally validated the models on each site and performed subgroup analyses across sex and race. Results: Stage 2 or higher AKI occurred in 3% (n=3,257) and 8% (n=2,296) of UFH and UPMC patients, respectively. Area under the receiver operating curve values (AUROC) for the UFH test cohort ranged between 0.77 (UPMC Model) and 0.81 (UFH Model), while AUROC values ranged between 0.79 (UFH Model) and 0.83 (UPMC Model) for the UPMC test cohort. UFH-UPMC Model achieved an AUROC of 0.81 (95% confidence interval [CI] [0.80, 0.83]) for UFH and 0.82 (95% CI [0.81,0.84]) for UPMC test cohorts; an area under the precision recall curve values (AUPRC) of 0.6 (95% CI, [0.05, 0.06]) for UFH and 0.13 (95% CI, [0.11,0.15]) for UPMC test cohorts. Kinetic estimated glomerular filtration rate, nephrotoxic drug burden and blood urea nitrogen remained the top three features with the highest influence across the models and health centers. Conclusion: Locally developed models displayed marginally reduced discrimination when tested on another institution, while the top set of influencing features remained the same across the models and sites.

BACKGROUND AND OBJECTIVES: Tools to assist with drug management for both nephrotoxic medications and renally eliminated drugs are urgently needed. "Multi-hospital Electronic Decision Support for Drug-associated Acute Kidney Injury" (MEnD-AKI) aims to examine the effect of a pharmacist-led intervention augmented with predictive analytics in the form of electronic alerts delivered to pharmacists followed by drug management recommendations provided to physicians via telemedicine for the early management of patients at risk of developing AKI or progressing to higher AKI stages. DESIGN: Prospective, multi-site, cluster-randomized clinical trial. SETTING: Eight hospitals within the UPMC health system. PATIENTS: Attending physicians belonging to primary services other than intensive care or organ transplant will be eligible for participation in the study. The unit of randomization is physician hospital services (clusters), and outcomes will be assessed for patients cared for by these physicians. INTERVENTIONS: Researchers will randomize 38 hospital service clusters to receive: 1) electronic medical record (EMR)-based AKI passive alert, which is standard of care at UPMC; this alert provides the diagnosis and staging of AKI but without recommendations for management; or 2) protocolized, tiered pharmacist-led intervention augmented with near-realtime predictive analytics in the form of automated alerts incorporated into a web application delivered to pharmacists followed by drug management recommendations provided to physicians via telemedicine for consideration and approval. OUTCOMES: The primary outcome is major adverse kidney events (MAKE) measured within 30 days of admission. Secondary outcomes include progression of AKI, AKI duration, and nephrotoxic burden. CLINICAL TRIALS REGISTRATION: ClinicalTrials.govNCT06264752 (v2).

Observational studies link high net ultrafiltration (UFNET) rates during continuous kidney replacement therapy (CKRT) to increased mortality. The Restrictive versus Liberal Rate of Extracorporeal Volume Evaluation in Acute Kidney Injury trial evaluated the feasibility of a restrictive versus liberal UFNET rate strategy. This stepped-wedge cluster-randomized trial enrolled patients in ten ICUs across two healthcare systems from July 2022 to June 2024. Each ICU was a cluster, with 1 randomly transitioning from liberal (2.0–5.0 mL/kg/h) to restrictive (0.5–1.5 mL/kg/h) group every two months after the first six months. The coprimary outcomes included between-group separation in UFNET rates, protocol adherence, and recruitment rate. Of 97 patients (55 liberal, 42 restrictive) enrolled, the mean (SD) delivered UFNET rate did not differ between the groups (2.05 [0.83] vs. 1.81 [0.86] mL/kg/h; adjusted P = 0.4). In per-protocol analysis, there was a significant between-group separation in mean UFNET rates (2.24 [0.72] vs. 1.22 [0.32] mL/kg/h; P = 0.002). Protocol deviations were similar (9.1% vs.7.1%, P = 0.7), and the recruitment rate was 0.99 (0.27) patients per ICU per two months. The use of rescue UFNET was higher in the restrictive group (14.5% vs. 66.7%; P < 0.001). In conclusion, despite high protocol adherence, there was minimal separation in delivered UFNET rates. While both strategies were feasible in select patients, the high rates of hemodynamic instability, the need for rescue UFNET, and physician override orders suggest that UFNET is more often driven by dynamic patient physiology than fixed protocols. This makes it challenging to maintain distinct, alternative UFNET targets in clinical practice. Trial registration number: ClinicalTrials.gov Identifier: NCT05306964.

Key Points The developed dynamic model for predicting progression to stage 2 or higher AKI using multicenter data had robust performance. Clinical decision support implementation of the developed model could help prevent AKI progression. Kinetic eGFR, nephrotoxic drug burden, and BUN were the top features and remained the same across the models and sites. Background AKI is an abrupt decline in kidney function that occurs in about 20% of hospitalized admissions and may lead to irreversible kidney damage. Methods We developed and externally validated deep learning models to dynamically predict progression to stage 2 or higher AKI defined by Kidney Disease Improving Global Outcomes serum creatinine criteria within the next 48 hours. We used an extensive set of predictors including demographics, admission source, comorbidities, medications, laboratory, and vitals measurements. Results Our retrospective study includes adult noncritical care patients at the University of Pittsburgh Medical Center (UPMC; 2018–2022; n =39,755) and the University of Florida Health (UFH; 2012–2019; n =122,324). In the UFH and UPMC development cohort, the mean age was 55 and 71 and 55% ( n =47,350) and 54% ( n =15,128) were female, respectively. Stage 2 or higher AKI occurred in 3% ( n =3257) and 8% ( n =2296) of UFH and UPMC patients, respectively. Area under the receiver operating characteristic curve values with 95% confidence interval (CI) ranged between 0.77 (95% CI, 0.75 to 0.78; UPMC Model—model trained on UPMC patients) and 0.81 (95% CI, 0.79 to 0.82; UFH Model—model trained on University of Florida patients) for the UFH test cohort and between 0.79 (95% CI, 0.78 to 0.8; UFH Model) and 0.83 (95% CI, 0.82 to 0.84; UPMC Model) for the UPMC test cohort. UFH-UPMC Model achieved an area under the receiver operating characteristic curve of 0.81 (95% CI, 0.80 to 0.83) for UFH and 0.82 (95% CI, 0.81 to 0.84) for UPMC test cohorts. Kinetic eGFR, nephrotoxic drug burden, and BUN remained the features with the highest influence across the models and institutions. Conclusions The model developed using multicenter data had robust performance, suggesting that implementation could help prevent AKI progression.