Portable Neural-Network-Based Burn Injury Analyzer Using UV, Visible, and Near-IR Spectroscopy

The objective of the proposed program is to develop a portable biomedical system for real-time automatic evaluation of burn injury. The system will comprise ultiple wavelength light sources emitting in the ultraviolet, visible and near-infrared ranges, optical detectors and a neural-network-based signal processing unit. It will allow physicians to objectively evaluate injured individuals, determine the burn depth and the structure of viable dermis and subcutaneous tissue, quanitfy the burn surface area and monitor healing. Using the same technqiue, the proposed neural-network-based system can be designed to monitor the vital signs of neonatal cerebral oxygenation, blood perfusion, kidney failure, and myocardial infarction. Since the proposed system is portable and operates in real-time, it can be carried by a person on the battlefield or to an emergency accident site to provide automatic diagnosis. Resulsts are produced in digital form, ready for transfer to a central command station. Application in the civilian sector would improve the delivery of emergency care, thereby reducing hospitalization time and health care costs. Physical Optics Corporation and Cedars-Sinai Medical Center have formed a uniquely qualified research team for the development of this biomedical sensor. Anticipated Benefits: The proposed ultraviolet, visible, and near-IR spectroscopy technology can be used widely in trauma care, under both military and civilian scenarious, to characterize biological tissues, tissue pH level, blood perfusion, oxygenation and arterial blood gas levels, and the anatomical structure of teeth. The neural network technology will provide reliable and sensitive identification of abnormal conditions.