Preparation of 2D sequences of corneal images for 3D model building

Abstract

A confocal microscope provides a sequence of images, at incremental depths, of the various corneal layers and structures. From these, medical practioners can extract clinical information on the state of health of the patient's cornea. In this work we are addressing problems associated with capturing and processing these images including blurring, non-uniform illumination and noise, as well as the displacement of images laterally and in the anterior posterior direction caused by subject movement. The latter may cause some of the captured images to be out of sequence in terms of depth. In this paper we introduce automated algorithms for classification, reordering, registration and segmentation to solve these problems. The successful implementation of these algorithms could open the door for another interesting development, which is the 3D modelling of these sequences.

Medical image classification based on artificial intelligence approaches: A practical study on normal and abnormal confocal corneal images

“…The work presented in [12] describes several approaches to prepare corneal images for 3D volume visualisation. Image displacements laterally and in the anterior-posterior directions, caused by subject movement, were addressed, in the former case, using the speeded-up robust features (SURF) algorithm, as well as the scale invariant feature transform (SIFT).…”

“…For the trained system, the pre-processing, feature extraction and Table 9 Comparison of statistical features between normal and abnormal data sets where INC indicates the abnormal value is higher than normal and DEC indicates it is lower. For comparison with the system proposed in [4,12], recognition ratios for the three main corneal layers obtained using the developed system were calculated in a way similar and for the same number of images. The results are shown in Table 10.…”

An efficient system for preprocessing confocal corneal images for subsequent analysis

“…The work presented in [8] established and discussed the main approaches to prepare the confocal microscope corneal images for subsequent analysis. It addressed the problems associated with capturing and processing these images including blurring, non-uniform illumination and noise, as well as the displacement of images laterally and in the anteriorposterior direction caused by subject movement due to respiration, cardiac pulse, or other factors.…”

A fully automatic nerve segmentation and morphometric parameter quantification system for early diagnosis of diabetic neuropathy in corneal images

“…CCM has also been used in the assessment of peripheral neuropathies [10,11,12,13,14,15,16,17,18,19]. The development of automated imaging algorithms for the processing of CCM images [20,21,22,23] is a necessary accompaniment to such work. Diabetes is the main cause of neuropathy complications and in the present work patients with a known history of cancer, chemotherapy, alcoholism, celiac disease or a deficiency of vitamin B12 or folate, abnormality in ANA or immunoglobulins were excluded to remove other causes of peripheral neuropathy from the input data.…”