Designing image analysis pipelines in light microscopy: A rational approach

Abstract

With the progress of microscopy techniques and the rapidly growing amounts of acquired imaging data, there is an increased need for automated image processing and analysis solutions in biological studies. Each new application requires the design of a specific image analysis pipeline, by assembling a series of image processing operations. Many commercial or free bioimage analysis software are now available and several textbooks and reviews have presented the mathematical and computational fundamentals of image processing and analysis. Tens, if not hundreds, of algorithms and methods have been developed and integrated into image analysis software, resulting in a combinatorial explosion of possible image processing sequences. This paper presents a general guideline methodology to rationally address the design of image processing and analysis pipelines. The originality of the proposed approach is to follow an iterative, backwards procedure from the target objectives of analysis. The proposed goal-oriented strategy should help biologists to better apprehend image analysis in the context of their research and should allow them to efficiently interact with image processing specialists.