Filters

2019 - 201912020 - 20231

More filters

2020 - 20242
Reset filters

Settings

Author: Dustin, Michael L. × Start date: 2011 ×

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Evolutionary design of explainable algorithms for biomedical image segmentation
    ([London] : Nature Publishing Group UK, 2023) Cortacero, Kévin; McKenzie, Brienne; Müller, Sabina; Khazen, Roxana; Lafouresse, Fanny; Corsaut, Gaëlle; Van Acker, Nathalie; Frenois, François-Xavier; Lamant, Laurence; Meyer, Nicolas; Vergier, Béatrice; Wilson, Dennis G.; Luga, Hervé; Staufer, Oskar; Dustin, Michael L.; Valitutti, Salvatore; Cussat-Blanc, Sylvain
    An unresolved issue in contemporary biomedicine is the overwhelming number and diversity of complex images that require annotation, analysis and interpretation. Recent advances in Deep Learning have revolutionized the field of computer vision, creating algorithms that compete with human experts in image segmentation tasks. However, these frameworks require large human-annotated datasets for training and the resulting “black box” models are difficult to interpret. In this study, we introduce Kartezio, a modular Cartesian Genetic Programming-based computational strategy that generates fully transparent and easily interpretable image processing pipelines by iteratively assembling and parameterizing computer vision functions. The pipelines thus generated exhibit comparable precision to state-of-the-art Deep Learning approaches on instance segmentation tasks, while requiring drastically smaller training datasets. This Few-Shot Learning method confers tremendous flexibility, speed, and functionality to this approach. We then deploy Kartezio to solve a series of semantic and instance segmentation problems, and demonstrate its utility across diverse images ranging from multiplexed tissue histopathology images to high resolution microscopy images. While the flexibility, robustness and practical utility of Kartezio make this fully explicable evolutionary designer a potential game-changer in the field of biomedical image processing, Kartezio remains complementary and potentially auxiliary to mainstream Deep Learning approaches.
  • Thumbnail Image
    Item
    Reconstitution of immune cell interactions in free-standing membranes
    (Cambridge : Company of Biologists, 2019) Jenkins, Edward; Santos, Ana M.; Felce, James H; Hatherley, Deborah; Dustin, Michael L.; Davis, Simon J.; Eggeling, Christian; Sezgin, Erdinc
    The spatiotemporal regulation of signalling proteins at the contacts formed between immune cells and their targets determines how and when immune responses begin and end. Therapeutic control of immune responses therefore relies on thorough elucidation of the molecular processes occurring at these interfaces. However, the detailed investigation of each component's contribution to the formation and regulation of the contact is hampered by the complexities of cell composition and architecture. Moreover, the transient nature of these interactions creates additional challenges, especially in the use of advanced imaging technology. One approach that circumvents these problems is to establish in vitro systems that faithfully mimic immune cell interactions, but allow complexity to be ‘dialled-in’ as needed. Here, we present an in vitro system that makes use of synthetic vesicles that mimic important aspects of immune cell surfaces. Using this system, we began to explore the spatial distribution of signalling molecules (receptors, kinases and phosphatases) and how this changes during the initiation of signalling. The GUV/cell system presented here is expected to be widely applicable.The spatiotemporal regulation of signalling proteins at the contacts formed between immune cells and their targets determines how and when immune responses begin and end. Therapeutic control of immune responses therefore relies on thorough elucidation of the molecular processes occurring at these interfaces. However, the detailed investigation of each component's contribution to the formation and regulation of the contact is hampered by the complexities of cell composition and architecture. Moreover, the transient nature of these interactions creates additional challenges, especially in the use of advanced imaging technology. One approach that circumvents these problems is to establish in vitro systems that faithfully mimic immune cell interactions, but allow complexity to be ‘dialled-in’ as needed. Here, we present an in vitro system that makes use of synthetic vesicles that mimic important aspects of immune cell surfaces. Using this system, we began to explore the spatial distribution of signalling molecules (receptors, kinases and phosphatases) and how this changes during the initiation of signalling. The GUV/cell system presented here is expected to be widely applicable.