Argue, David P.

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 Rationale: Ongoing healthcare staffing shortages have prompted new care paradigms that incorporate remote care delivery to support timely detection of patient physiological deterioration through mechanisms other than clinician bedside evaluations. In partnership with institutional Information Technology, our team has developed a unique clinical informatics environment, known as the Advanced Medical Practice (AMP) platform, that incorporates data from multiple sources and platforms and supports remote observation of patients from the electronic intensive care unit(eICU). AMP takes advantage of technologies outside of vendor-supported EHRs, and its advanced user-interface design reduces information overload while incorporating real-time data from novel devices and sensors. We partnered with the hospital practice to co-develop a score within the AMP platform(Clinical Events Detection and Response[CEDAR]) that incorporates heart rate, blood pressure, respiratory rate, and SpO2 to stratify patient risk and generate deterioration alerts for the electronic ICU(eICU) team. This formative evaluation employed an implementation science approach to refine CEDAR and ensure it is acceptable to users and can be feasibly implemented in practice. Methods: The study team conducted this pilot work in the eICU at Mayo Clinic (Rochester, MN). We observed the eICU workflow and documented feedback from eICU staff in fieldnotes. eICU staff were asked to log issues in a tracking file during shifts. Issues were reviewed daily and discussed at multidisciplinary team meetings weekly. Observation and tracking data were coded to constructs in the Consolidated Framework for Implementation Research(CFIR 2.0) framework and mapped to Expert Recommendations for Implementation Change(ERIC) strategies that targeted key determinants. Results: Between 9/23/24 and 10/25/24 we completed 25 days of observations. Most identified issues related to technical/design issues like unclear vital trigger alerts (CFIR: Innovation trialability and design); staff knowledge of CEDAR functionality and attitudes about functionality, complexity, or relative advantage compared to current systems (CFIR: Inner setting access to knowledge and information and relative advantage); and difficulties incorporating CEDAR into workflow(CFIR: Inner setting/compatibility). Implementation strategies mapped to CFIR determinants included promoting adaptability of customized interface features; staging implementation scale-up with iterative improvements; providing local technical assistance; assessing for readiness and identifying facilitators and barriers; and conducting educational outreach visits. Conclusions: Formative evaluation during pilot studies are a critical step in the process of intervention development and testing. User engagement and frameworks from implementation science can identify issues of feasibility and inform strategies to ensure acceptability prior to intervention scale-up in clinical practice and readiness for conducting larger implementation studies.

Purpose Hearing loss is a common impairment of the human senses with an estimated 48 million American adults reporting some trouble hearing; however, access to hearing health care is limited. Detection of hearing loss through a mobile, handheld tool can provide an important access point and potentially expedited access to the continuum of hearing health care. Here, we determined that GoAudio, a portable, automated hearing assessment tool, can be used to identify individuals who require additional hearing evaluation in a clinical workflow. Method This initial study included 24 adults, ages 18–65 years ( M = 50, SD = 12), tested with GoAudio versus “gold-standard” clinical audiometry for eight frequencies to evaluate “real-world” applications. Participants utilized noise-canceling headphones combined with a tablet-based application for the GoAudio assessment. Results The primary study outcome measurements were the comparison of hearing thresholds (dB HL) from clinical audiometry and GoAudio. Results suggest that GoAudio is comparable to clinical audiometry for the identification of hearing loss at most frequencies (except 1 kHz for both ears and 2 kHz in the right ear). Upon stratifying data based on age, we identified that GoAudio is capable of identifying suspected age-related hearing loss or hearing thresholds greater than 30 dB HL at higher frequencies in both ears. Conclusion The study results support that GoAudio can be used effectively in clinical practice workflows as a reliable hearing assessment tool for the identification of hearing loss at the majority of frequencies outside a sound-treated booth and can detect characteristics of age-related hearing loss. Supplemental Material https://doi.org/10.23641/asha.13087682

Background: The patient/caregiver experience during CAR-T therapy is stressful, overwhelming, terrifying, and often a patient’s last treatment option. The Mayo Clinic Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery Innovation and Design team has worked with the CAR-T therapy clinical team to develop a patient experience that provides patients with a sense of caring, supportive environment, timely knowledge, and realistic expectations. Using a human-centered design approach, the Innovation and Design team worked with patients and caregivers to understand latent and unspoken needs in order to develop an ideal CAR-T therapy patient journey. Methods: With qualitative interviewing techniques, patient observation, and low fidelity experimentation, 21 patients/caregiver pairs were interviewed throughout their CAR-T therapy experience in 2018. Patients were interviewed at several touch points as well as encouraged to reach out to the Innovation and Design team at any point with reflections on their experiences. Patients were recruited as they began their evaluation phase for CAR-T therapy. The interviews were unscripted to allow for a breadth of discovery by not constraining the conversations to previously developed themes. As themes emerged from patient/caregiver interviews, artifacts and interventions were designed to alleviate pain points and improve the patient/caregiver experience. These artifacts and interventions were integrated into the clinical processes in real time and patient/caregivers were interviewed to understand the impact of these activities. Results: Several themes emerged from qualitative interviews with patients and caregivers. From the themes, interventions were developed. We were able to demonstrate a qualitative improvement in patient/caregiver experience through these interventions (Figure 1). Conclusions: Patients/caregivers undergoing CAR-T therapy have unique issues surrounding the logistics of care, emotional burden, and physical effects of treatment. We implemented processes to address these issues and observed a qualitative improvement via patient interviews/feedback. Ongoing work includes optimizing remote monitoring, digital platforms for patient education, and a quantitative study looking at patient reported outcomes (PROs) in such patients. To our knowledge, this is the first report for care delivery optimization in real-world practice for this new therapy.

Transcription activator-like effectors (TALEs) are extremely effective, single-molecule DNA-targeting molecular cursors used for locus-specific genome science applications, including high-precision molecular medicine and other genome engineering applications. TALEs are used in genome engineering for locus-specific DNA editing and imaging, as artificial transcriptional activators and repressors, and for targeted epigenetic modification. TALEs as nucleases (TALENs) are effective editing tools and offer high binding specificity and fewer sequence constraints toward the targeted genome than other custom nuclease systems. One bottleneck of broader TALE use is reagent accessibility. For example, one commonly deployed method uses a multitube, 5-day assembly protocol. Here we describe FusX, a streamlined Golden Gate TALE assembly system that (1) is backward compatible with popular TALE backbones, (2) is functionalized as a single-tube 3-day TALE assembly process, (3) requires only commonly used basic molecular biology reagents, and (4) is cost-effective. More than 100 TALEN pairs have been successfully assembled using FusX, and 27 pairs were quantitatively tested in zebrafish, with each showing high somatic and germline activity. Furthermore, this assembly system is flexible and is compatible with standard molecular biology laboratory tools, but can be scaled with automated laboratory support. To demonstrate, we use a highly accessible and commercially available liquid-handling robot to rapidly and accurately assemble TALEs using the FusX TALE toolkit. Together, the FusX system accelerates TALE-based genomic science applications from basic science screening work for functional genomics testing and molecular medicine applications.