Computer Vision News - May 2021

Congrats, Doctor! 12 Roxane Licandro completed the doctoral program in technical sciences in March 2021 at TU Wien with distinction in cooperation with the Medical University of Vienna (MUW) . The focus of her research lies in investigating and creating novel techniques to assess dynamic developmental patterns (DDP) in the human. In her work she proposes a concept for the spatio temporal modelling of these DDPs and demonstrates the applicability of it for different modalities (fMRI, fetal MRI, flowcytometry, whole body MRI), age ranges (fetus, children, adults), diseases (acute leukaemia, paediatric stroke, multiple myeloma) and applications (segmentation, prediction, exploration, classification). RoxaneLicandro is currentlyauniversityassistant at the Computer Vision Lab (TU Wien) and a postdoc researcher at the Computational Imaging Research Lab (MUW) . S patio Temporal Modelling is the process of estimating an optimal way on encoding trajectories in space and over time (e.g. of complex diseases, metabolic or developmental pathways) and plays an important role in personalizedmedicine, virtual clinical trials or drug target identification. It enables the assessment of an individual’s treatment response, the prediction of disease progression or the determination of a developmental status or its deviation from the model. D ynamic Developmental Patterns form the main challenge in modelling trajectories, constituted of the incompleteness and irregularity of observations, inter-patient variability and impairing factors like comorbidity, age or individual treatment response. In Figure 1 an example of DDPs in childhood leukaemia assessed by flowcytometry measurements of blood cells over three treatment time points of two patients are illustrated.Cancercellsaremarked inredandnon-cancercells inblue. Theheterogeneity of the blast cells’ appearance in the observation space as well as the diminishing count of these in later treatment time points form the main challenge in assessing the blast count and consequently the treatment response over time. Figure 1: Longitudinal flowcytometry measures of cancer (red) and non- cancer (blue) blood cells of two patients over three treatment time points. [European Project AutoFLOW] .

RkJQdWJsaXNoZXIy NTc3NzU=