Automated analysis of blood vessel morphology using angiographic optical coherence tomography
In collaboration with Aarhus University Hospital, this project automates OCTA-based vascular analysis to support skin disease diagnostics using a custom, multi-step MATLAB workflow.
Quantifying microscopic droplets in colloidal systems through machine learning-based image analysis
Segmenting and quantifying droplets of varying sizes in complex food colloids is a major challenge for image analysis. MIDAS, a machine learning-based tool built on Cellpose, tackles this challenge with multi-scale segmentation and advanced shape analysis, optimized for speed and scalability.
Deep Learning-assisted 3D Segmentation for Monitoring Cartilage Regeneration in Knee MRI Scans
This collaborative project with Aarhus University Hospital supports a clinical trial investigating the regenerative effects of stem cell injections for treating knee osteoarthritis. An AI-assisted tool is being developed to enable accurate 3D segmentation of cartilage and surrounding bone in MRI scans, facilitating the quantification of structural changes over time.
Quantification of Microglia Morphology from whole Piglet Brain Brightfield Images
This project investigates how gut inflammation affects brain health by analyzing changes in microglia structure in piglet hippocampus tissue using a custom QuPath–ImageJ pipeline.
Classification of Centralized Nuclei from Muscle Fibers
In healthy muscle fibers, the nuclei is positioned at the periphery of the cell; abnormal nuclear positioning, where the nuclei has moved to a more central location, is a common marker for myopathies. In this project, we identify muscle cells and nuclei from fluorescent images of muscle tissue sections, and then classify each cell based on the absence or presence of nuclei that have “detached“ from the periphery, in order to count the number of affected cells in the tissue.
Estimation of Myonuclear Domains from the Assisted Segmentation of Muscle Fiber Nuclei
The tips of human muscle fibres were probed with RNAscope and imaged with 3D microscopy, to characterise the arrangement of cells and myonuclei at the myotendinous junction. In this collaboration, we developed a Imaris XT Matlab module to estimate the myonucleus density along human muscle fibers and to characterize the myonuclei domains (MND) from semi-automatically segmented data.
Protein Expression in Kidney Epithelial Cell Primary Cilium
The IACF created a fully automated MATLAB script to report the mean fluorescence intensity of different proteins of interest inside subregions of the primary cilium in kidney epithelial cells.