Nucleus detection on propidium iodide stained digital slides

Framework for Comparison and Evaluation of Image Segmentation Algorithms for Medical Imaging

“…Images taken with a d igital microscope of tissue samples provide a convenient way of evaluation that is reproducible and more robust compared to the legacy microscopy where the samples deteriorate over time making the co mparison of the evolution of the samples over time harder. Hav ing the slides containing the t issue samples digitalized at a high resolution also makes it possible to use sophisticated image p rocessing algorith ms of d ifferent sorts, some help diagnose illnesses by finding malignous cells [15] [16], some [11] just aid the pathologist by doing a pre-filtering that eliminates parts of the image that do not hold significant data so that the expert can focus on the parts which truly require their expertise. Image segmentation is usually a key element in any such algorith ms [3].…”

Semi-Automated Quantitative Validation Tool for Medical Image Processing Algorithm Development

“…Image cytometry is the sub-field where the software tools are used to process the digitized sample, enabling its users to analyze vastly more objects than the traditional visualmanual method using microscopes for inspection and clickers for counting. The project we are working on aims to reproduce a flow cytometry analysis by image processing means [2] [3]. Is the ICM approach viable as a diagnostic approach?…”