Evaluation of topographical variation in ocular surface temperature by functional infrared thermography

The objective of this work is to evaluate topographical variation in the ocular surface temperature (OST) among the young, elderly and the subjects wearing contact lens using thermographic methodology. We recorded thermographic sequence lasting of 25 s for each eye. The ocular region in each of the thermal images in the sequence was identified and warped into a standard form. Then, the warped sequence was divided into a number of sub-sequences. A differential image which is an image matrix was obtained from each of these sub-sequences, by subtracting thermal images within the sub-sequence. And the histogram of the differential image was modelled by Gaussian mixture model to discriminate eyelashes from the ocular surface for every thermal image in the sub-sequence. Later, thermal data of eyelashes were eliminated in every thermal image and statistical analysis was performed on the sequences. Finally, topographical profile of each subject group was approximated by equations and illustrated using various temperature profiles. The ocular surface of the young subject was observed to be the warmest, and tear film was determined to play a major role in the topographical and temporal variations in OST. Significant topographical variation was observed among subject groups. Based on our compiled average OST profile (AOSTP), the maximum predictability of the bioheat simulation on ocular model can reach up to 90%.