Filters

2008 - 200912020 - 20231

More filters

2020 - 20242
Reset filters

Settings

Start date: 1984 × Type: Text × Subject: article × Subject: human × WGL Institute: INM ×

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
    Alterations in Event Related Potentials (ERP) associated with tinnitus distress and attention
    (Dordrecht [u.a.] : Springer Science + Business Media B.V, 2008) Delb, W.; Strauss, D.J.; Low, Y.F.; Seidler, H.; Rheinschmitt, A.; Wobrock, T.; D'Amelio, R.
    Tinnitus related distress corresponds to different degrees of attention paid to the tinnitus. Shifting attention to a signal other than the tinnitus is therefore particularly difficult for patients with high tinnitus related distress. As attention effects on Event Related Potentials (ERP) have been shown this should be reflected in ERP measurements (N100, phase locking). In order to prove this hypothesis single sweep ERP recordings were obtained in 41 tinnitus patients as well as 10 control subjects during a period of time when attention was shifted to a tone (attended) and during a second phase (unattended) when they did not focus attention to the tone. Whereas tinnitus patients with low distress showed a significant reduction in both N100 amplitude and phase locking when comparing the attended and unattended measurement condition a group of patients with high tinnitus related distress did not show such ERP alterations. Using single sweep ERP measurements the results of our study show, that attention in high tinnitus related distress patients is captured by their tinnitus significantly more than in low distress patients. Furthermore our results provide the basis for future neurofeedback based tinnitus therapies aiming at maximizing the ability to shift attention away from the tinnitus. © 2008 The Author(s).