Portable Ultrasound System for Automated Detection of Abdominal Free-Fluid

点击翻译按钮获取中文摘要

PROJECT SUMMARY / ABSTRACT The Focused Assessment with Sonography in Trauma (FAST) exam is the standard of care for rapid detection of abdominal free fluid in emergency medicine and trauma critical care. It is a point-of-care ultrasound exam that incorporates four views of the abdomen. In trauma, abdominal free fluid is assumed to be blood until proven otherwise. Timely detection is critical because if untreated abdominal hemorrhage can rapidly lead to hemorrhagic shock and death. Performing and interpreting a FAST exam requires appropriate training and practice. In community hospitals the FAST exam is underutilized due to limited access to physicians who are able to perform the exam (typically an emergency physician or trauma surgeon). In addition, the low cost and portability of ultrasound make it an ideal triage tool for pre-hospital settings. In these settings, detection of abdominal free fluid would impact medical transport prioritization and aid in the distribution of limited resources. BioSensics, in collaboration with Boston University School of Medicine, proposes to develop a portable ultrasound system to enable a minimally trained operator to perform a FAST exam. The system will guide the operator through the image acquisition process and automatically compute a probability for abdominal free fluid. A fundamental innovation is that ultrasound probe position and orientation, with respect to the patient, will be measured and used to inform both the image processing system and the operator guidance system. Feasibility of the image processing component of the system has been demonstrated through two retrospective studies of perihepatic (e.g., right upper quadrant) abdominal ultrasound views. In the first study (prior to our Phase I SBIR) abdominal free fluid was identified with 100% sensitivity and 90% specificity in a cohort of 20 FAST exams mostly from medical causes of abdominal free fluid, such as ascites. In the second study (during Phase I of this project) abdominal free fluid was identified with 81% sensitivity and 88% specificity in a cohort of 47 cases of hemoperitoneum and 51 negative controls (classification of hemoperitoneum, especially from trauma, is more difficult because the free fluid volume is often smaller). In addition, during Phase I we built an ultrasound system that records the position and orientation of the probe during the image acquisition process. The objectives of this Phase II SBIR project are the following: 1) develop validate image processing algorithms to compute a probability for the presence of abdominal free fluid in all four FAST exam views by expanding on the prior work done for the perihepatic view; and 2) develop and test software to enable a minimally trained user to perform a FAST exam by providing real-time guidance based on probe position and orientation. The proposed technology has significant commercialization potential. Our first target market is emergency medical services (EMS) including civilian ambulance companies and the US military. Our second target market is emergency departments with a particular focus on rural and community hospitals. More broadly, the technology could aid in many settings where non-trauma assessment of abdominal free fluid is necessary.