You are here

Development of a Clinical Decision Support Tool for Early-Detection of Directed Energy-Induced Injury to the Retina



OBJECTIVE: To develop a clinical decision support tool that can be utilized by medics/physicians at Role 3/4 to automatically identify retinal damage after laser exposure. 

DESCRIPTION: Laser injury to the retina does not cause immediate pain therefore it may be undetected until long-term damage to the retina causes vision loss. Currently Role 3 and Role 4 medics and physicians do not have the ability to quickly detect retinal damage after laser exposure. This deficiency reduces quality of care and increases the time to return to duty. A clinical decision support tool to detect laser-induced retinal injury will enable decisions to be made at the Role 3 level and allow physicians to determine the best treatment plan. Lasers are extensively used by the military in designators, rangefinders and guidance systems. Many of these devices operate at wavelengths that are absorbed by the human eye which can produce harmful effects. In addition, high energy laser directed energy weapons have been progressively evolving and there is the possibility that anti-eye laser weapons are being developed. Their use would cause new types of combat casualty which have not yet been extensively experienced, but which will require accurate diagnosis to ensure effective medical solutions. The development of accurate and smart ocular diagnostic technology will expand the capability of clinicians to diagnose and treat ocular injuries induced by laser exposure at the point-of-injury as well as point-of-care. The proposed technology will provide improved field-care capabilities, reduce recovery time of injured warfighters, and help minimize complications of wound healing after trauma or surgery. This test if developed would be a valuable tool in the hands of eye care providers worldwide to assist in the evaluation of laser induced retinal injuries. Payoff: Soldiers will receive appropriate care and will return to duty more quickly. Quality of care will be improved for soldiers who suffer laser-induced retinal injury that may not be detected immediately after exposure without a rapid portable diagnostic tool.  

PHASE I: Develop clinical decision support tool algorithms for rapid analysis of fundus images of the retina before and after laser injury. The retinal images will first go through a rigorous image enhancement phase, image data augmentation, and pixel normalization. Following the enhancement and augmentation processes, algorithms will be developed to identify retinal abnormalities associated with laser damage. Optimize and validate the laser –induced injury detection technology in human fundus images. The end product must achieve high sensitivity and accuracy > 95%. Create protection plan for intellectual property. 

PHASE II: Transition algorithms developed in Phase I. Develop a fully functional prototype. Define the parameters including ease of use, sensitivity and specificity. Develop strategy to acquire FDA approval as a clinical decision support tool. Identify commercial and clinical partners for Phase III. Develop a detailed business plan outlining monetary return on investment within two years of completion of Phase II ( sales, licensing agreements, venture capital, non-SBIR grants). 

PHASE III: Perform experiments as necessary to prepare for FDA review of an IND application. Conduct market analysis. Initiate a Phase I clinical trial to validate utility of a clinical decision support tool to detect laser-induced retinal injury. The diagnostic methodology once developed will be commercialized and made available to the military including forward deployed medics, FSTs and Combat Support Hospitals (CSH). Demand for this device in emergency rooms, ophthalmology and optometry practices worldwide is expected to be high. The portable rapid retinal damage detection tool will also be highly useful to non-military health care providers due to increased use of lasers in the civilian sector, ie fiberoptics and industry. Coordinate with Vision Center of Excellence to facilitate FDA approval and dissemination of the product to military clinicians. 


1: Mainster MA, Stuck BE and Brown JJr. 2004. Assessment of alleged retinal laser injuries. Arch Ophthalmol. 122:1210-1217.

2:  Harris MD1, Lincoln AE, Amoroso PJ, Stuck B, Sliney D. 2003 .Laser eye injuries in military occupations. Aviat Space Environ Med. Sep

3: 74(9):947-52.

4:  Marrugo, A. G., & Millan, M. S. (2011). Retinal image analysis: preprocessing and feature extraction. In Journal of Physics: Conference Series (Vol. 274, No. 1, p. 012039). IOP Publishing.

5:  Grewal, P. S., Oloumi, F., Rubin, U., & Tennant, M. T. (2018). Deep learning in ophthalmology: a review. Canadian Journal of Ophthalmology.

6:  Razzak, M. I., Naz, S., & Zaib, A. (2018). Deep Learning for Medical Image Processing: Overview, Challenges and the Future. In Classification in BioApps (pp. 323-350). Springer, Cham.

KEYWORDS: Laser, Retina, Machine-learning, Deep Learning Architecture 

US Flag An Official Website of the United States Government