Physiological Optics Laboratory

Dr. Michael Smolek’s laboratory is involved in several different areas of ophthalmology. The NIH-supported research program (EY014162) involves determining the diagnostic value of optical aberrations of the cornea and the cornea + lens optical system. It is believed that certain diseases and disorders that affect the optical system can be detected automatically from analysis of the aberration pattern, as well as by using other features, such as those based on the shape of the cornea. Dr. Smolek’s group determines the reliability of measuring these patterns with different conditions, the threshold between normal and abnormal conditions, and makes refinements in the algorithms used to extract and analyze the aberrations. A large part of this effort requires establishing standards for analyzing and viewing data, as well as determining potential standards for classifying and grading the severity of different disorders.

A second research program involves artificial intelligence analysis of fundus photography, including the automated detection of diabetic retinopathy. There are 3 goals in this program: 1) to develop a self-contained, inexpensive screening device that can be used at any clinical without the need for user training; 2) to eliminate the backlog of clinical data that must be interpreted by trained specialists at fundus reading centers by producing a reliable artificially intelligent system; and 3) to establish simpler and more objective standards for grading the severity of retinal disease.

A third research program is the interpretation of confocal microscopy images with artificial intelligence methods. Dr. Smolek’s group is currently collaborating with Dr. Richard W. Yee’s clinical research group at the Hermann Eye Center in Houston, Texas in this area.