Abstract Preview

Oral Presentation
 
Format: jpg / png / jpeg / pdf
Abstract Details
Scientific Research Abstract
Multidisciplinary
N/A
Author & Affiliation
1
Hung-Bin Tsai hbtsai37@gmail.com National Taiwan University Hospital Division of Hospital Medicine Taipei Taiwan *
 
 
 
 
 
 
 
 
 
Presenting Author
Hung-Bin
Tsai
hbtsai37@gmail.com
Taiwan
Abstract Content
E-FASH-Guided POCUS at Home for AKI and UTI Assessment in Hospital-at-Home Care
Acute kidney injury (AKI) and urinary tract infection (UTI) are common challenges in older and medically complex patients receiving Hospital-at-Home (HaH) care. Timely assessment of volume status, urinary retention, and kidney–urinary tract complications is essential for safe home-based acute care. This study aimed to develop and preliminarily evaluate the Extended Focused Assessment with Sonography for Home Care (E-FASH) protocol as an AI-enhanced, teleconsultation-assisted POCUS-at-Home framework for AKI and UTI patients.
This mixed-methods study included two phases. From December 2022 to December 2024, the Taiwan Society of Home Health Care collaborated with one medical center, one regional hospital, and the Dulan Home Clinic team to review HaH cases and convene expert meetings. Discussions focused on clinical indications, workflows, and standardized POCUS-at-Home applications for AKI and UTI assessment. From January to July 2025, the team partnered with a Taiwanese ultrasound manufacturer to design and validate the E-FASH protocol. Core components included inferior vena cava assessment for volume status, residual bladder volume measurement for urinary retention, and kidney–urinary tract screening. AI-assisted imaging analysis and voice-recognition synchronization were evaluated during home-based teleconsultation.
Seven expert meetings were conducted, and 18 HaH cases were reviewed. The findings were compiled into an educational monograph on POCUS at Home, including clinical indications, representative cases, symptom-based assessments, ultrasound-assisted procedures, and E-FASH workflows. The AI model detected inspiratory and expiratory inferior vena cava changes with precision of 99.97% and recall of 99.71%. For residual bladder volume measurement, accuracy was 93.2%, recall 100%, and F1-score 88.1. Voice recognition showed sensitivity above 95%, trigger latency below 500 ms, and false activation per hour below 0.5. Four physicians achieved a 100% match rate between voice commands and ultrasound image segments.
E-FASH-guided POCUS at Home appears feasible for AKI and UTI assessment in HaH care. Integrating volume assessment, bladder evaluation, urinary tract screening, AI-assisted imaging, and voice-synchronized documentation may support safer decision-making, earlier escalation, and competency-based POCUS training.
Acute Kidney Injury; Urinary Tract Infection; Point-of-Care Ultrasound; Hospital-at-Home
https://storage.unitedwebnetwork.com/files/1367/1245359-37821-Real-time AI Augmentation for Teleconsultation_0630'26_405038.jpg
* JPG only
 
2358
Submitted