Rxivist logo

Anatomical modeling of brain vasculature in two-photon microscopy by generalizable deep learning

By Waleed Tahir, Sreekanth Kura, Jiabei Zhu, Xiaojun Cheng, Rafat Damseh, Fetsum Tadesse, Alex Seibel, Blaire S. Lee, Frederic Lesage, Sava Sakadžié, David A. Boas, Lei Tian

Posted 10 Aug 2020
bioRxiv DOI: 10.1101/2020.08.09.243394

Segmentation of blood vessels from two-photon microscopy (2PM) angiograms of brains has important applications in hemodynamic analysis and disease diagnosis. Here we develop a generalizable deep-learning technique for accurate 2PM vascular segmentation of sizable regions in mouse brains acquired from multiple 2PM-setups. The technique is computationally efficient, thus ideal for large-scale neurovascular analysis. Vascular segmentation from 2PM angiograms is an important first-step in hemodynamic modeling of brain vasculature. Existing segmentation methods based on deep-learning either lack the ability to generalize to different imaging systems, or are computationally infeasible for large-scale angiograms. In this work, we overcome both these limitations by a method that is generalizable to various imaging systems, and is able to segment large-scale angiograms. We employ a computationally efficient deep-learning framework with a loss function that incorporates a balanced binary-cross-entropy loss and a total-variation regularization on the network's output. Its effectiveness is demonstrated on experimentally acquired in-vivo angiograms from mouse brains of dimensions up to 808x808x702 micrometers. To demonstrate the superior generalizability of our framework, we train on data from only one 2PM-microscope, and demonstrate high-quality segmentation on data from a different microscope without any network-tuning. Overall, our method demonstrates 10x faster computation in terms of voxels-segmented-per-second and 3x larger depth compared to the state-of-the-art. Our work provides a generalizable and computationally efficient anatomical modeling framework for brain vasculature, which consists of deep-learning based vascular segmentation followed by graphing. It paves the way for future modeling and analysis of hemodynamic response at much greater scales that were inaccessible before. ### Competing Interest Statement The authors have declared no competing interest.

Download data

  • Downloaded 578 times
  • Download rankings, all-time:
    • Site-wide: 64,221
    • In neuroscience: 9,197
  • Year to date:
    • Site-wide: 152,527
  • Since beginning of last month:
    • Site-wide: 120,186

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide