The overarching goal of research conducted in my lab aims at pushing the boundary of what image analyses and machine learning can do for quantifying biological data. Projects we are currently pursuing aim at understanding e.g. apical constriction during gastrulation in C.elegans, mitotic cluster formation in Drosophila and Tribolium, or morphological tissue changes in the green alga Volvox. The common denominator of such projects is the undisputable necessity to analyze large amounts of light microscopy data without causing impossible amounts of manual data curation and processing – often the one major bottleneck.

From a computational point of view we are interests in (i) object segmentation and tracking, e.g. cells or nuclei, that enables easy and powerful data curation workflows (‘leveraged editing’), (ii) building feature rich ‘downstream’ quantification pipelines, and (iii) casting those functionalities in reusable and modular software packages. More technical aspects of our daily work are: (i) developing and applying novel, scalable solvers for the graphical models that underlie the automated data processing, and (ii) enabling so called ‘deep learning’ techniques to be applied to biological data.

Results in the field show that the next generation of quantification algorithms and (semi-)automated pipelines will have a big impact on cell and developmental biology, enabling research that is to date simply not feasible. At the same time I am convinced that recent advances in image analysis and machine learning, once applicable to biologically relevant data, will allow us to make big strides towards gaining a deeper understanding of a wide range of cellular and developmental processes.

• Graphical Models (Bayesian Networks/ Factor Graphs)
• Deep Learning (Classification, Domain Adaptation, etc.)
• Discrete Optimization
• Efficient Algorithms
• Software Design and Development
  
  
  

(see all)