Optical Coherence Tomography

About

Background: Many OCT devices are commercially available. As each machine exhibits its own characteristics on the OCT signal, it is not possible to have one image processing software to handle all the OCT data generated by different devices in a consistent universal fashion. In other words, image appearance and quantitative measurement results may be different among multiple OCT images scanned on the same subject but obtained using different devices, which may lead to less reliable and consistent diagnosis and clinical decision making.   Technology: In order to solve this problem, investigators have developed a way to normalize OCT data so that OCT images scanned on the same subject should look the same and provide similar quantitative measurements (within the expected measurement variability range). Signal characteristics (histogram pattern and parameters) are automatically analyzed image by image and signals are normalized to the standard histogram. With this normalization, one universal image processing software can be used for all the visualization and quantitative measurement tasks regardless the origin of the data (different devices). This will enable sharing OCT data globally (in the medical community) without worrying about the systematic differences among different devices.    Advantages: 1. Normalizes OCT signal without losing information and distorting images 2. Standardizes OCT signal data range so that universal image viewer software can be developed without worrying about variable raw data range (conventionally preprocessing the data is required to accommodate widely variable data range) 3. With the standardized OCT signal data, universal image processing algorithms (e.g. segmentation) can be employed without worrying about the systematic measurement differences among OCT data obtained using different devices (common problem for data sharing/comparison)