Automated analysis of blood vessel morphology using angiographic optical coherence tomography

Collaboration

N. Moustgaard Knudsen and M. Pedersen, Comparative Medicine Lab, Aarhus University Hospital

Research Background

Non-invasive imaging techniques like Laser Speckle Contrast Imaging (LSCI) and Optical Coherence Tomography Angiography (OCTA) offer valuable insights into skin vascularization, blood perfusion and micro-morphology in patients with atopic dermatitis and psoriasis. This study investigates the correlation between quantitative imaging measures, such as vascular density, depth, and blood perfusion, and clinical findings.

Method

Development of a multi-component pipeline in MATLAB. The pipeline consists of three scripts designed to automate the following processing steps:

  1. Preprocessing: Autocropping z-stacks to eliminate noisy image signals, removing motion artifacts using a wavelet-FFT filter, and detecting the air-skin border to align the entire stack at this border.

  2. Depth Detection: Identifying the Capillary Loop Depth (CLD) and Superficial Plexus Depth (SPD) from the z-aligned image stacks.

  3. Morphological Analysis: Calculating morphological metrics at the depth of the SPD, including mean vessel diameter, branch length, vessel density, and fractal dimension.

Next
Next

Deep Learning-assisted 3D Segmentation for Monitoring Cartilage Regeneration in Knee MRI Scans