NucleusJ 1.0, an ImageJ plugin, has been shown to be a useful tool to analyse nuclear morphology and chromatin organisation in plant and animal cells. However, technological improvements of confocal microscopy have speeded-up image acquisition, highlighting the bottleneck in 3D image analysis caused by manual steps in NucleusJ 1.0 and limiting its use for big data analysis. NucleusJ 2.0 is a new release of NucleusJ, in which image processing is achieved more quickly using a command-line user interface. Starting with large collection of 3D nuclei, segmentation can be performed by the previously developed Otsu-modified method or by a new 3D gift-wrapping method, taking better account of nuclear indentations and unstained nucleoli. These two complementary methods based on threshold and edgebased detection, are compared for their accuracy by using three types of datasets-digitised spheres, microspheres and plant nuclei-available to the community through an open source repository accessible at https://www.brookes.ac.uk/indepth/images/. A discrete geometric method was introduced to improve the surface area calculation, a key parameter when studying nuclear morphology, replacing an ImageJ default tool by a new one that includes pixel context information. Finally, NucleusJ 2.0 was evaluated using original plant genetic material, in which nuclear morphology and chromatin organisation are strongly affected and by assessing its efficiency on nuclei stained with DNA dyes or after 3D-DNA Fluorescence in situ hybridisation. With these improvements, NucleusJ 2.0 permits the generation of large user-curated datasets that will be useful for software benchmarking or to train convolution neural networks.

PurposeQuantitative and automatic analysis of intracoronary optical coherence tomography images is useful and time-saving to assess cardiovascular risk in the clinical arena.MethodsFirst, the interfaces of the intima, media, and adventitia layers are segmented, by means of an original front propagation scheme, running in a 4D multi-parametric space, to simultaneously extract three non-crossing contours in the initial cross-sectional image. Second, information resulting from the tentative contours is exploited by a machine learning approach to identify healthy and diseased regions of the arterial wall. The framework is fully automatic.ResultsThe method was applied to 40 patients from two different medical centers. The framework was trained on 140 images and validated on 260 other images. For the contour segmentation method, the average segmentation errors were for the intima–media interface, for the media–adventitia interface, and for the adventitia–periadventitia interface. The classification method demonstrated a good accuracy, with a median Dice coefficient equal to 0.93 and an interquartile range of (0.78–0.98).ConclusionThe proposed framework demonstrated promising offline performances and could potentially be translated into a reliable tool for various clinical applications, such as quantification of tissue layer thickness and global summarization of healthy regions in entire pullbacks.

No abstract