Kılıçkaya, Oğuz

Abstract Rationale Clear and accurate communication is essential in critical care, yet patients with non-English language preferences (NELP) often face limited access to professional interpreters. Although machine translation has been explored in healthcare, most studies have focused on written translation. Their use in real-time, two-way interpretation specifically for high-stakes ICU conversations remains largely unexamined. To bridge this gap, we developed AI-TransLATE (AI-enhanced Transition to Language-Agnostic Transcultural Engagement), a voice-to-voice multilingual interpretation tool designed to support real-time interactions in critical care. Following successful initial validation, we evaluated clinicians’ perceptions of its usability in simulated ICU encounters and compared its performance with commercially available AI tools, including Gemini and ChatGPT. Methods We developed three standardized ICU role-play scenarios focused on shared decision-making regarding ECMO, tracheostomy, and comfort care, designed to encourage natural, unscripted dialogue. Clinicians were asked to use the AI translation tools during these encounters in Spanish, Chinese, or Arabic, while a bilingual research team member acted as the patient’s family member. After each simulation, participants completed a post-session survey evaluating communication quality, satisfaction (using a 5-point Likert scale), and usability through the System Usability Scale (SUS). Results Eighteen clinicians participated, including residents, fellows, and attending physicians with an average of 3.6 ± 1.2 years of postgraduate experience. Fifty-four post-simulation surveys were completed across Arabic, Chinese, and Spanish languages comparing AI-TransLATE, ChatGPT, and Google Gemini. Participants rated the simulation as realistic across all cases (ECMO = 3.94 ± 0.66; Tracheostomy = 4.22 ± 0.43; Comfort Care = 4.06 ± 0.42; p = 0.32). Satisfaction was highest for AI-TransLATE (4.11 ± 0.68) and ChatGPT (4.00 ± 0.69) but lowest for Gemini (2.06 ± 1.00; p < 0.001). Willingness to use in real life followed the same trend. (AI-TransLATE = 4.00 ± 0.69, ChatGPT = 3.89 ± 0.96, Gemini = 2.00 ± 1.03 (p < 0.001)). SUS scores supported these findings (AI-TransLATE = 71.25 ± 13.35; ChatGPT = 72.22 ± 15.57; Gemini = 40.28 ± 21.30; p < 0.001), corresponding to good usability for AI-TransLATE and ChatGPT and poor usability for Gemini. Clinicians mentioned interruptions, uneven translation flow, and difficulty judging accuracy, but appreciated it when the AI-TransLATE conveyed tone and intonation, enabling smoother, more natural communication. Conclusion Overall, AI-TransLATE and ChatGPT demonstrated strong usability and satisfaction, showing promise as clinical communication tools; however, further validation in real-world ICU settings is warranted, and data privacy and confidentiality remain important considerations. This abstract is funded by: Critical care research committee mayo clinic

Abstract Introduction Large Language Models (LLMs) have proven effective at analyzing unstructured and structured data from electronic health records (EHR). Their reliance on static EHR data means they miss the dynamic, high-resolution physiological signals that are the first indicators of patient decline, creating a fundamental data lag as clinical documentation of an event may occur hours after its onset. This represents a critical gap in the ability to develop truly effective triage tools. Furthermore, while LLMs can synthesize multi-source data for acuity assessment, they exhibit known limitations, including hallucinations that constrain autonomous use in safety-critical settings. In contrast, a Vision-Language Model (VLM) with strong image-grounded clinical reasoning can interpret vital sign trend images, surpassing the limitations of current EHR-based models. This approach provides immediate triage evaluations that lead to earlier detection of patient deterioration. Methods This mixed-methods study deployed a VLM (Gemini 2.5) integrated within the eICU dashboard. We used near-real-time two-hour trends in heart rate, blood pressure, respiratory rate, and oxygen saturation from consecutive patient cases during alerts in hospital floor units monitored by eICU that admit acute medical and surgical patients. To increase the diversity of the sample, we sought routine cases from patients without alerts and emergent cases from patients admitted to the medical, surgical, trauma, and mixed ICUs. A structured context prompt was used to generate a standardized summary including clinical interpretation, data quality and artifact assessment, triage recommendation, and the model’s chain-of-thought reasoning. Results We included patients from Mayo Clinic Rochester, MN, using an 800-patient cohort for prompt development and a 100-patient cohort for validation against clinically adjudicated results. In the validation cohort the VLM achieved 80% accuracy (80/100). An analysis of the 20 incorrect predictions revealed a clear pattern of conservative over-triage: 80% (16/20) of errors were clinically acceptable adjacent-class discrepancies (routine↔urgent or urgent↔emergent). Only 4/20 discrepancies represented a substantive misclassification. Most errors therefore reflected conservative prioritization rather than incorrect assessment. The model demonstrated a perfect safety profile, underestimating patient acuity in 0% of cases, and thus avoiding the most clinically dangerous error. These results confirm the feasibility of VLM-based approach and provide a strong foundation for the proposed work. Conclusion This VLM performs cross-modal reasoning by fusing deep visual perception with broad clinical context. The result is a fully contextualized and explainable triage prioritization, which provides direct decision support and integrates into the clinical workflow, allowing for faster and more confident acute care assessments. This abstract is funded by: none

Abstract Effective communication in the intensive care unit (ICU) is essential, particularly for patients with non-English language preference, yet timely access to professional interpreters remains limited. While artificial intelligence (AI)-based translation tools have been explored in outpatient and nonacute care settings, studies evaluating their use in acute care, environments such as the ICU remain limited. To address this gap, we developed AI-TransLATE (AI-enhanced Transition to Language-Agnostic Transcultural Engagement), a speech-based translation tool designed for multilingual communication in critical care settings. This study aimed to assess the interpretation quality of AI-TransLATE across four languages—Spanish, Chinese, Arabic, and Turkish—using scripted ICU scenarios. We created ICU communication scripts and recorded bilingual research team members simulating clinical interactions. Two independent bilingual evaluators assessed interpretation quality using a 5-point Likert scale across fluency, adequacy, meaning preservation, and severity of errors. Clarity and cultural appropriateness were also rated. Percentage agreement was used to assess interrater agreement. AI-TransLATE achieved acceptable composite scores (≥16/20) across all languages. Spanish and Turkish performed consistently well; Chinese and Arabic showed variability due to omissions and terminology errors. AI-TransLATE shows promise as a clinical communication tool, but further evaluation in real-world, unscripted ICU settings is needed.

ABSTRACT Understanding clinical trajectories of sepsis patients is crucial for prognostication, resource planning, and to inform digital twin models of critical illness. This study aims to identify common clinical trajectories based on dynamic assessment of cardiorespiratory support using a validated electronic health record data that covers retrospective cohort of 19,177 patients with sepsis admitted to intensive care units (ICUs) of Mayo Clinic Hospitals over 8-year period. Patient trajectories were modeled from ICU admission up to 14 days using an unsupervised machine learning two-stage clustering method based on cardiorespiratory support in ICU and hospital discharge status. Of 19,177 patients, 42% were female with a median age of 65 (interquartile range [IQR], 55–76) years, The Acute Physiology, Age, and Chronic Health Evaluation III score of 70 (IQR, 56–87), hospital length of stay (LOS) of 7 (IQR, 4–12) days, and ICU LOS of 2 (IQR, 1–4) days. Four distinct trajectories were identified: fast recovery (27% with a mortality rate of 3.5% and median hospital LOS of 3 (IQR, 2–15) days), slow recovery (62% with a mortality rate of 3.6% and hospital LOS of 8 (IQR, 6–13) days), fast decline (4% with a mortality rate of 99.7% and hospital LOS of 1 (IQR, 0–1) day), and delayed decline (7% with a mortality rate of 97.9% and hospital LOS of 5 (IQR, 3–8) days). Distinct trajectories remained robust and were distinguished by Charlson Comorbidity Index, The Acute Physiology, Age, and Chronic Health Evaluation III scores, as well as day 1 and day 3 SOFA ( P < 0.001 ANOVA). These findings provide a foundation for developing prediction models and digital twin decision support tools, improving both shared decision making and resource planning.

Background and Aim: Distal radius fractures are the most frequent upper extremity fractures in the whole population. The incidence of distal radius fractures is 195.2/100,000, and the rate is increasing. It can occur with low-energy trauma in the elderly associated with osteoporosis. The purpose of this study is to compare the two different treatment groups (1. Sedation analgesia, 2. Supracondylar radial nerve block) in terms of reduction in Visual Analogue Scale (VAS) score, the length of hospital stay, complications, side effects, patient-physician satisfaction, and the ease of application. Materials and Methods: Patients diagnosed with distal radius fracture were enrolled prospectively, to whom the study’s exclusion criteria do not apply. The patients included in the study were randomly separated into two groups. Ketofol was administered to the sedation and analgesia group’s patients at a dose of 0.5 mg/kg (ketamine 0.5 mg/kg, propofol 0.5 mg/kg) intravenously, and supracondylar radial nerve block was performed on the other group’s patients under ultrasound guidance. Results: There were 20 patients (6 male, 14 female) in the sedation and analgesia group and 21 patients (11 male, 10 female) in the supracondylar radial nerve block group. There was no significant difference between the groups regarding patient-physician satisfaction and complications. The length of hospital stay was significantly shorter in the supracondylar radial nerve block group (49.25 ± 4.05) compared with the sedation and analgesia group (125 ± 5.85) (p < 0.01). The mean reduction VAS score in the block group was observed to be lower than the admission VAS score (p < 0.01). Conclusion: Consequently, ultrasound-guided supracondylar radial nerve block would be a valuable alternative to procedural sedation analgesia and could become routine clinical practice in treating distal radius fractures.