The 3D fluorescence distribution and provide accurate quantification as in previous studies. Therefore the combination of CTA, BLI and FMT enabled us to improve the accuracy of Endostar anti-cancer efficacy. Our future work direction is to further optimize the resolution of our Micro-CT. The conventional histology of tumor vessels requires sacrificing large number of animals at different time points, then takes time for slicing and staining tissues to identify the location and the density of the vessels. Therefore, if CTA is able to detect miniscule vascular changes, it may become a real-time method to estimate the in vivo anti-angiogenesis effect continuously, which is an effective way to reduce costs and save time. At the same time we will improve the reconstruction accuracy to better solve the problem of high absorption of light in deep tissue and integrate different molecular imaging modalities. Due to the phase dependent nature of chemotherapy drugs and radiotherapy knowledge of how the cells progress through the different phases is crucial. There have been a number of mathematical models developed for populations of cells progressing round the cell cycle. Systems of ordinary differential equations may be used to model the growth kinetics of populations of cells however these are too simplistic to capture the intrinsic properties of the cell cycle, but are often an invaluable first step in understanding the kinetics of a population of cells. To adequately model crucial properties of a population of cells such as age, mass or DNA distribution a system of partial differential equations is needed. Many 
partial differential equation models share the same fundamental population balance structure as detailed in, and. These models may broadly be grouped in terms of which property of the cell is used to structure the model, the main properties used being DNA, age and mass. There are advantages of using a DNA or mass structured model in as much that these quantities may be easily determined experimentally, however such a model contains no information about the age of a particular cell and as such it is possible for cells to remain in the cycle for an infinite amount of time. By use of an age-structured model it is possible to control the length of time a cell may remain in the cell cycle, in particular the G1 phase. Another advantage of age structuring is that, if growth rates and nutrient uptake rates for a given cell line are known, it is possible to determine the mass and DNA Cryptochlorogenic-acid content of a cell from its age, however given the cells DNA content or mass it is not possible to determine a cell’s age as there is not a one-to-one mapping between age and DNA or mass. Analysis has been undertaken to determine the existence and stability of steady size/DNA distributions which may occur under specific circumstances using an age structured model. Population balance models have been used not only on healthy, unperturbed cell lines but also to model the effects of various treatments to cancer cell populations,, and. In this paper, an age structured cell cycle model is considered together with two different functions governing the movement between the G1 and S phases. Whilst, different functions have been used in the past, and little has been done to study the effects of different functions on the phase distributions of cells. It is shown that it is possible to obtain very similar growth curves using different Anemarsaponin-BIII transition functions with the fundamental difference being in the phase distributions for the cells. Although the differences in the phase distributions lie within the range of experimental error for many techniques such as conventional flow cytometry it may be significant in terms of treatment optimisation.