The DCELLIQ platform is self-sufficient, making it possible to analyze large data sets with little need for attendance by the user. A fundamental requirement of our method is that the imaging frequency must be sufficient to capture at least three images during experiments . Our imaging software and hardware settings were optimized to keep Fulvestrant fluorescence light exposure to a minimum. These conditions produce no difference in mitotic or interphase duration compared to cells imaged by phase contrast microscopy . The fact that our method requires only a single exposure for each time point also helps limit overall exposure compared to another method that uses confocal microscopy and requires multiple planes to be acquired for each time point . Most automated image analysis platforms use SVM-based approaches to classify individual images . Most SVM approaches, however, do not exploit how individual features change as a function of time in the initial step of classifying objects, potentially losing valuable information. Instead, time information is used afterwards to reconstruct how class membership of a nucleus changes as a function of time, based on input of allowed phase transitions. Furthermore, accuracy of SVM approaches may require high-resolution images that limit the number of cells that can be analyzed in a given experiment or concomitant analysis of multiple markers, thereby increasing light exposure . In our own experience, we have found that extensive training is required for SVM approaches to function efficiently. Training sets are very dependent on similar experimental conditions and may not be easily transferred from one cell line to another. Because TSA analyzes how extracted features change as a function of time, the approach is more tolerant of initial variation in the GW786034 VEGFR/PDGFR inhibitor absolute values of these parameters between cells. For example, cells often vary in the amount of H2B-GFP protein that is expressed, leading to variation in fluorescence intensity. Because the time-series algorithm only measures how intensity changes as a function of time, the approach is less sensitive to variation in absolute levels of the reporter protein.