Towards an Image Analysis Toolbox for High-Throughput Drosophila Embryo Rnai Screens

Abstract

We build an image analysis toolbox for high-throughput Drosophila embryo RNAi screens. The goal is to tag the embryo as normal, developmentally delayed or abnormal based on the ventral furrow formation. We break the problem into two parts: in the first, we detect the developmental stage based on the progress of the ventral furrow formation, and in the second, we tag the embryo as normal/developmentally delayed/abnormal based on the stage detected and the elapsed time. The crux of the algorithm is the multiresolution classifier, and we show that, by classifying in multiresolution spaces, we obtain better results than by classifying the embryo image alone. The final 2D accuracy obtained was 93.17%, while by using 3D information, it increased to 98.35%.

Multiresolution classification with semi-supervised learning for indirect bridge structural health monitoring

“…MRC is a supervised classification framework (see Figure 2), originally proposed for bioimaging applications [4,7,8]. It decomposes images into S localized space-frequency subbands using wavelet packets, a data-adaptive MR technique [9].…”

Exploring indirect vehicle-bridge interaction for bridge SHM

“…Redundancy often leads to increased accuracy, as has been found in a host of bioimaging problems (see Chebira & Kovačević 2008a and references therein). One possible example of the power of multiresolution techniques in pattern classification is that developed for the classification of Drosophila embryo development (Kellogg et al 2007). Using a highly accurate multiresolution classification algorithm developed by Kovačević and her group, the process is now automated and reproducible, with accuracy greater than 98% (Kellogg et al 2007).…”

Multiresolution identification of germ layer components in teratomas derived from human and nonhuman primate embryonic stem cells

“…2). This system has proven accurate for various classification tasks [5,4,6,7]; we now briefly describe it. Multiresolution Block.…”

“…In our previous work [5,4,6,7], we found that classifying in multiresolution (MR) subspaces adds to the discriminative power of a classification system. Moreover, based on our previous experience, modified Haralick texture features [4] (our improvement of the known Haralick texture features), were typically sufficient for high classification accuracies on those data sets (in the mid to upper 90s).…”