Computer vision for sequential non-invasive microscopy imaging cytometry with applications in embryology

Molder, Anna (2016) Computer vision for sequential non-invasive microscopy imaging cytometry with applications in embryology. Doctoral thesis (PhD), Manchester Metropolitan University.

Preview

Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (34MB) | Preview

Abstract

Many in vitro cytometric methods requires the sample to be destroyed in the process. Using image analysis of non-invasive microscopy techniques it is possible to monitor samples undisturbed in their natural environment, providing new insights into cell development, morphology and health. As the effect on the sample is minimized, imaging can be sustained for long un-interrupted periods of time, making it possible to study temporal events as well as individual cells over time. These methods are applicable in a number of fields, and are of particular importance in embryological studies, where no sample interference is acceptable. Using long term image capture and digital image cytometry of growing embryos it is possible to perform morphokinetic screening, automated analysis and annotation using proper software tools. By literature reference, one such framework is suggested and the required methods are developed and evaluated. Results are shown in tracking embryos, embryo cell segmentation, analysis of internal cell structures and profiling of cell growth and activity. Two related extensions of the framework into three dimensional embryo analysis and adherent cell monitoring are described.