Characterization of Human Torso Vascular Morphometry in Normotensive and Hypotensive Trauma Patients

Abstract

Non-compressible torso hemorrhage remains the leading cause of preventable death on the battlefield and a leading cause of death in civilian centers. The purpose of this project is to provide basic morphometric understanding of aortic and vena caval anatomy as it relates to the rest of the body with the goal of enabling the use of various occlusion catheters that can control bleeding in non-compressible torso injuries in the field. One year into the project, we have identified and processed the base morphomics and much of the aortic and vena caval measurements for the 2000 civilian CT scans. Additionally, we have developed a machine learning algorithm for aortic identification that will speed processing and reduce inter- and intra-user variability. Regarding the military scans, we have secured HRPO IRB approval and are actively working on a process for receiving military CT scans. Finally, we have begun planning our data analysis and have executed pilot assessments to validate the approach.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 2016
Accession Number
AD1016799

Entities

People

  • Stewart C. Wang

Organizations

  • University of Michigan

Tags

Communities of Interest

  • Autonomy
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Arteries
  • Cardiovascular Diseases
  • Cardiovascular Physiological Phenomena
  • Data Analysis
  • Data Mining
  • Data Science
  • Databases
  • Health Services
  • Hemorrhage
  • Identification
  • Information Science
  • Institutional Review Board
  • Machine Learning
  • Measurement
  • Medical Personnel
  • X-Ray Computed Tomography

Fields of Study

  • Medicine

Readers

  • Clinical Trial Research.
  • Trauma Surgery or Emergency Medicine.

Technology Areas

  • AI & ML