Accurate and Versatile 3D Segmentation of Plant Tissues at Cellular Resolution

Abstract

Quantitative analysis of plant and animal morphogenesis requires accurate segmentation of individual cells in volumetric images of growing organs. In the last years, deep learning has provided robust automated algorithms that approach human performance, with applications to bio-image analysis now starting to emerge. Here, we present PlantSeg, a pipeline for volumetric segmentation of plant tissues into cells. PlantSeg employs a convolutional neural network to predict cell boundaries and graph partitioning to segment cells based on the neural network predictions. PlantSeg was trained on fixed and live plant organs imaged with confocal and light sheet microscopes. PlantSeg delivers accurate results and generalizes well across different tissues, scales, and acquisition settings. We present results of PlantSeg applications in diverse developmental contexts. PlantSeg is free and open-source, with both a command line and a user-friendly graphical interface.Recently an approach combining the output of two neural networks and watershed to detect individual cells showed promising results on segmentation of cells in 2D [19]. Although this method can in principle be generalized to 3D images, the necessity to train two separate networks poses additional difficulty for non-experts.While deep learning-based methods define the state-of-the-art for all image segmentation problems, only a handful of software packages strives to make them accessible to non-expert users in biology (reviewed in [20]). Notably, the U-Net segmentation plugin for ImageJ [21] conveniently exposes U-Net predictions and computes the final segmentation from simple thresholding of the probability maps. CDeep3M [22] and DeepCell [23] enable, via the command-line, the thresholding of the probability maps given by the network, and DeepCell allows instance segmentation as described in [19]. More advanced post-processing methods are provided by the ilastik Multicut workflow [24], however, these are not integrated with CNN-based prediction.Here, we present PlantSeg, a deep learning-based pipeline for volumetric instance segmentation of dense plant tissues at single-cell resolution. PlantSeg processes the output from the microscope with a CNN to produce an accurate prediction of cell boundaries. Building on the insights from previous work on cell segmentation in electron microscopy volumes of neural tissue [13,15], the second step of the pipeline delivers an accurate segmentation by solving a graph partitioning problem. We trained PlantSeg on 3D confocal images of fixed Arabidopsis thaliana ovules and 3D+t light sheet microscope images of developing lateral roots, two standard imaging modalities in the studies of plant morphogenesis. We investigated a range of network architectures and graph partitioning algorithms and selected the ones which performed best with regard to extensive manually annotated groundtruth. We benchmarked PlantSeg on a variety of datasets covering a range of plant organs and image resolutions. Overall, PlantSeg delivers excellent results on ...

BigDataProcessor2: A free and open-source Fiji plugin for inspection and processing of TB sized image data

“…As big image data is becoming increasingly prevalent we are positive that the already existing user base (e.g. Alladin et al 2020, Villani et al 2019, Wolny et al 2020 will grow even further in the future. Supplementary Figure 4: Screenshot of the Open menu.…”

Microscopy-based assay for semi-quantitative detection of SARS-CoV-2 specific antibodies in human sera

“…Then, images were processed to correct for the uneven illumination profile in each channel. Next, we segmented individual cells with a seeded watershed algorithm (21), using nuclei segmented via StarDist (22) as seeds and boundary predictions from a U-Net (23,24) as a heightmap. We evaluated this approach using leave-oneimage-out cross-validation on the manual annotations and measured an average precision (25) of 0.77 +-0.08 (i.e., on average 77% of segmented cells are matched correctly to the corresponding cell in the annotations).…”

“…In addition, we predict per pixel probabilities for the boundaries between cells and for the foreground (i.e. whether a given pixel is part of a cell) using a 2D U-Net (23) based on the implementation of Wolny et al 2020 (24). This method was trained using the 9 annotated images, see above.…”

“…The processing steps are parallelized over the images of a plate if possible. We use efficient implementations for the U-Net(24), StarDist(22) and the watershed algorithm (http://ukoethe.github.io/vigra/) as well as other image processing algorithms(49). We use pytorch (https://pytorch.org/) to implement GPU-accelerated cell feature extraction.…”

Integration of Cell Growth and Asymmetric Division During Lateral Root Initiation In Arabidopsis thaliana

“…Second, we used PlantSeg (Wolny et al 2020) to segment all cells of the three lateral root primordia and overlying tissues. The segmented cells were visualized using the software Blender and segmentation errors (under-segmentation requiring to split cells and over-segmentation requiring to merge cells) were corrected (see methods).…”