Automatic Localization of Interest Points in Zebrafish Images with Tree-Based Methods

Abstract

Abstract. In many biological studies, scientists assess effects of experimental conditions by visual inspection of microscopy images. They are able to observe whether a protein is expressed or not, if cells are going through normal cell cycles, how organisms evolve in different experimental conditions, etc. But, with the large number of images acquired in high-throughput experiments, this manual inspection becomes lengthy, tedious and error-prone. In this paper, we propose to automatically detect specific interest points in microscopy images using machine learning methods with the aim of performing automatic morphometric measurements in the context of Zebrafish studies. We systematically evaluate variants of ensembles of classification and regression trees on four datasets corresponding to different imaging modalities and experimental conditions. Our results show that all variants are effective, with a slight advantage for multiple output methods, which are more robust to parameter choices.

Evaluation and Comparison of Anatomical Landmark Detection Methods for Cephalometric X-Ray Images: A Grand Challenge

“…Following [41], Vandaele et al address the problem as 19 separate binary pixel classification problems, one for each landmark. One pixel belongs to the positive class if its distance to the landmark position is less than R, where R is a method parameter.…”

Towards generic image classification using tree-based learning: An extensive empirical study

“…Despite its conceptual simplicity and its rather low run-time complexity, it yielded interesting results on a few datasets. Subsequently, variants of the method were proposed in (Moosmann et al, 2008;Marée et al, 2009;Dumont et al, 2009;Stern et al, 2011) for object categorization, image segmentation, interest point detection, and content-based image retrieval.…”

Automated Multimodal Volume Registration based on Supervised 3D Anatomical Landmark Detection

“…They have been shown recently to outperform global landmark matching algorithms in various applications (Fanelli et al, 2013;Wang et al, 2015). Here, we extended the 2D landmark detection method of (Stern et al, 2011) to 3D imaging.…”

“…As in (Stern et al, 2011), we propose and compare two approaches: in the first, a classification model is trained for each landmark to predict if a voxel corresponds to the landmark position. In the second, a regression model is trained to predict the euclidean distance between a voxel and the landmark position.…”