A deep learning classification task for brain navigation in rodents using micro-Doppler ultrasound imaging.

Autor:	Lambert T; Neuro-Electronics Research Flanders, Leuven, Belgium.; VIB, Leuven, Belgium.; Imec, Leuven, Belgium.; Department of Neuroscience, Faculty of Medicine, KU Leuven, Leuven, Belgium., Brunner C; Neuro-Electronics Research Flanders, Leuven, Belgium.; VIB, Leuven, Belgium.; Imec, Leuven, Belgium.; Department of Neuroscience, Faculty of Medicine, KU Leuven, Leuven, Belgium., Kil D; Neuro-Electronics Research Flanders, Leuven, Belgium.; VIB, Leuven, Belgium.; Imec, Leuven, Belgium.; Department of Neuroscience, Faculty of Medicine, KU Leuven, Leuven, Belgium., Wuyts R; Imec, Leuven, Belgium., D'Hondt E; Imec, Leuven, Belgium., Montaldo G; Neuro-Electronics Research Flanders, Leuven, Belgium.; VIB, Leuven, Belgium.; Imec, Leuven, Belgium.; Department of Neuroscience, Faculty of Medicine, KU Leuven, Leuven, Belgium., Urban A; Neuro-Electronics Research Flanders, Leuven, Belgium.; VIB, Leuven, Belgium.; Imec, Leuven, Belgium.; Department of Neuroscience, Faculty of Medicine, KU Leuven, Leuven, Belgium.
Jazyk:	angličtina
Zdroj:	Heliyon [Heliyon] 2024 Mar 06; Vol. 10 (5), pp. e27432. Date of Electronic Publication: 2024 Mar 06 (Print Publication: 2024).
DOI:	10.1016/j.heliyon.2024.e27432
Abstrakt:	Positioning and navigation are essential components of neuroimaging as they improve the quality and reliability of data acquisition, leading to advances in diagnosis, treatment outcomes, and fundamental understanding of the brain. Functional ultrasound imaging is an emerging technology providing high-resolution images of the brain vasculature, allowing for the monitoring of brain activity. However, as the technology is relatively new, there is no standardized tool for inferring the position in the brain from the vascular images. In this study, we present a deep learning-based framework designed to address this challenge. Our approach uses an image classification task coupled with a regression on the resulting probabilities to determine the position of a single image. To evaluate its performance, we conducted experiments using a dataset of 51 rat brain scans. The training positions were extracted at intervals of 375 μm, resulting in a positioning error of 176 μm. Further GradCAM analysis revealed that the predictions were primarily driven by subcortical vascular structures. Finally, we assessed the robustness of our method in a cortical stroke where the brain vasculature is severely impaired. Remarkably, no specific increase in the number of misclassifications was observed, confirming the method's reliability in challenging conditions. Overall, our framework provides accurate and flexible positioning, not relying on a pre-registered reference but rather on conserved vascular patterns. Competing Interests: The authors of this publication declare that they have no financial or non-financial conflicts of interest to disclose. (© 2024 The Authors.)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu