Automated 3D bio-imaging analysis of nuclear organization by NucleusJ 2.0

NucleusJ 1.0, an ImageJ plugin, is a useful tool to analyze nuclear morphology and chromatin organization in plant and animal cells. NucleusJ 2.0 is a new release of NucleusJ, in which image processing is achieved more quickly using a command-lineuser 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 are compared for their accuracy by using three types of datasets available to the community at https://www.brookes.ac.uk/indepth/images/ . Finally, NucleusJ 2.0 was evaluated using original plant genetic material by assessing its efficiency on nuclei stained with DNA dyes or after 3D-DNA Fluorescence in situ hybridization. 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.

Fast simulation of stent deployment with plastic beam elements

Coronary stent deployment is a reference cardiology intervention, used to treat atherosclerosis and prevent heart attacks. The outcomes of the intervention highly depend on the accuracy of the stent apposition, which could benefit from per-operative prediction tools. In this paper, we propose a fast and mechanically realistic 3D simulation of a coronary stent expansion. Our simulation relies on the finite element method and involves serially linked beam elements to model the slender geometry of a stent. The elements are implemented with a non-linear elasto-plastic behavior, describing realistically the complex deformation of a balloon-expandable stent. As a proof of concept, we simulated the free expansion of a coronary stent. The simulation output was compared with micro-CT data, acquired experimentally during the device expansion. Results show that the plastic beam model is able to reproduce successfully the final geometry of the stent. In addition, the use of 1D elements allows to achieve a significantly lower computational time than for equivalent literature simulations, based on 3D elements. This preliminary work highlights the compatibility of our method with clinical routine in terms of execution time. Further developments include the application of the method to more advanced simulation scenarios, with the addition of a personalized artery model.

[Physical principles of optical biopsy]

The purpose of this presentation is to explain the physical principles underlying the three main methods used to obtain images of living tissues at the cellular scale. In confocal microscopy, the tissue of interest is illuminated and scanned through a confocal aperture; a lateral resolution close to 1 microm can be obtained with high numerical aperture. Multiphoton microscopy uses a high-power short-pulse laser with instantaneous irradiance sufficient to excite fluorescence in a very small focal volume. The concentration of natural tissue fluorophores is too low to obtain an adequate signal, so exogenous fluorophores have to be added, either locally or through the body. These fluorophores can be conjugated to a variety of biomolecules that target specific disease processes, thereby increasing diagnostic specificity. Finally, OCT (optical coherence tomography) provides high-resolution images of entire tissue volumes by using a broadband source and an interferometric configuration; the depth of field and lateral resolution are both on the micrometer scale. These methods allow images to be obtained at the cellular level, but image contrast and stability require specific adjustment for their significance to be established with respect to conventional biopsy methods.