Abstract—Angiogenesis is an interactive process between tumor, endothelial and stromal cells in order to create a network for oxygen and nutrients supply, necessary for tumor growth. According to this, angiogenic activity is considered a suitable method for both tumor growth or inhibition detection. The angiogenic potential is usually estimated by counting the number of blood vessels in particular sections. One of the most popular assay tissues to study the angiogenesis phenomenon is the developing chick embryo and its chorioallantoic membrane (CAM). In this paper we present an automated image analysis method and the corresponding tool that gives an unbiased quantification of the micro-vessel density and growth in angiogenic CAM images. Two experiments have been conducted using the developed tool; a) Tumor growth has been detected and quantified at different stages of embryonic development, b) the effect of dexamethasone (i.e. an inhibitor of the angiogenesis phenomenon) has been validated over a series of CAM samples. Experimental results presented in this work indicate the efficiency of the automated angiogenesis quantification method in both tumor growth and inhibition detection. I. INTRODUCTION ngiogenesis is a regulated process that is essential for cell development and that has also been implicated in physiological, as well as pathological phenomena in developed organisms [1]. More specifically angiogenesis is the physiological process involving the growth of new blood vessels from pre-existing vessels. Thus angiogenesis is a normal process in growth and development, as well as in wound healing. However, this is also a fundamental step in the transition of tumors from a dormant state to a malignant state. Thus, an excessive and deregulated angiogenic response is thought to contribute to cancer, and other diseases [2]. More precisely, the angiogenetic formation of blood vessels in tumors, is an interactive process between tumor, endothelial and stromal cells in order to create a network for oxygen and nutrients supply, necessary for tumor growth. Cancer cells are cells that have lost control of their ability Manuscript received June 30, 2006. Charalampos N. Doukas is with the University of the Aegean, Dep. of Information & Communication Systems Engineering, Samos, Greece. (phone: +30-2106010198, email: doukas@ieee.org) Ilias Maglogiannis is with the University of the Aegean, Dep. of Information & Communication Systems Engineering, Samos, Greece (phone: +30-22730-82239, email: imaglo@aegean.gr) Aristotle Chatziioannou is with the Institute of Biological Research and Biotechnology, National Hellenic Research Foundation, Greece. Heleni Loutrari is with the “Marianthi Simou” Laboratory, Evangelismos Hospital, University of Athens Medical School, Athens, Greece. to divide in a controlled fashion. A tumor consists of a population of rapidly dividing and growing cancer cells. Tumors induce blood vessel growth (angiogenesis) by the secretion of various growth factors (e.g. Vascular Endothelial Growth Factor or VEGF). Growth factors, such as VEGF can induce capillary growth into the tumor, supplying required nutrients and allowing for tumor expansion. Thus angiogenesis is a necessary and required step for transition from a small harmless cluster of cells, to a large tumor. Angiogenesis is also related to the spread of a tumor, or metastasis. Single cancer cells can break away from an established solid tumor, enter the blood vessel, and be carried to a distant site, where they can implant and begin the growth of a secondary tumor. Evidence now suggests that the blood vessel in a given solid tumor may in fact be mosaic vessels, comprised of endothelial cells and tumor cells, caused by the angiogenesis phenomenon. The subsequent growth of metastases will also require a supply of nutrients and oxygen by the angiogenetic vessel network. One of the most popular techniques assay tissues to study the angiogenesis phenomenon is the assessment of tissues developing in chick embryo and its chorioallantoic membrane (CAM). CAM is a highly vascular structure lining the inner surface of the egg shell. The study of the formation of its vessels, at different stages of development, is efficiently indicative for the phenomenon ([7]-[12]). The aim of this study was twofold; firstly, to develop and evaluate an automated image analysis method that would give an unbiased quantification of the micro-vessel density. Secondly, to conduct experiments regarding the quantification of both tumor growth and inhibition over the lapse of time. In this context, the developed tool uses advanced image processing techniques for exploiting vessel characteristics, such us vessel length, density, vessel branching points and textural information that assist in the assessment and quantification of the angiogenesis level. The presented method has been validated by comparing automated results to manual expert assessment over a series of digital chick embryo, indicating the high accuracy and reproducibility of the tool. In addition, the first set of the conducted experiments, indicates the average increase of the angiogenesis over time, whereas the second set demonstrates the significant efficiency of dexamethasone as an tumor inhibitor. Both validation and experimental results are presented in this work. The rest of the paper is organized as follows; Section II presents related work regarding CAM development and automated angiogenesis quantification. A Computer Based Tool for Tumor Growth and Inhibition Detection using Angiogenesis Quantification Charalampos N. Doukas, Ilias Maglogiannis, Aristotle Chatzioannou, Heleni Loutrari A