Digital image processing is widely used in the field of time-lapse microscopy and biological research to provide statistical data of cellular dynamics. The data can provide more comprehensive understanding of the molecular phenomenon. Further, digital image processing enables rapid and consistent quantification of qualitative observations. The image processing model examined here provides a study to identify structures called retraction fibers (RFs) that are formed during epithelial-mesenchymal transition (EMT) [1], an important developmental process which also occurs during cancer metastasis. Quantifying RF formation is an important task for biologists studying cellular regulation of EMT. This thesis work uses digital image processing and computer vision algorithms to detect and track each RF in image sequences of cells undergoing EMT that are captured using time-lapse microscopy. The algorithms isolate the RFs with reasonable precision. Statistical information is generated about these automatically detected RFs, such as the number formed during a particular time window, lifetime of each, and their geometric dimension. This information can in turn be used by biologists to quantitatively measure the extent of EMT under different test conditions. Biologists feel that the information thus obtained may help clarify the molecular interactions of cell migration and will aid in developing methods of preventing cancer metastasis. Experimental results show that this methodology has significant potential in helping biologists determine RF behavior during EMT.



College and Department

Ira A. Fulton College of Engineering and Technology; Electrical and Computer Engineering



Date Submitted


Document Type





Epithelial-mesenchymal transition, digital image processing, cancer metastasis, retraction fibers