3D convolutional neural networks for stalled brain capillary detection. (February 2022)

• Record Type: Journal Article
• Authors: Solovyev, Roman; Kalinin, Alexandr A.; Gabruseva, Tatiana
• Is Part Of: Computers in biology and medicine. Volume 141(2022)
• Publication Date: 2022-02
• Subjects: Alzheimer's disease -- Cerebral blood flow -- Computer vision -- Deep learning; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.105089
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Adequate blood supply is critical for normal brain function. Brain vasculature dysfunctions, including stalled blood flow in cerebral capillaries, are associated with cognitive decline and pathogenesis in Alzheimer's disease. Recent advances in imaging technology enabled generation of high-quality 3D images that can be used to visualize stalled blood vessels. However, localization of stalled vessels in 3D images is often required as the first step for downstream analysis. When performed manually, this process is tedious, time-consuming, and error-prone. Here, we describe a deep learning-based approach for automatic detection of stalled capillaries in brain images based on 3D convolutional neural networks. Our approach includes custom 3D data augmentations and a weights transfer method that re-uses weights from 2D models pre-trained on natural images for initialization of 3D networks. We used an ensemble of several 3D models to produce the winning submission to the "Clog Loss: Advance Alzheimer's Research with Stall Catchers" machine learning competition that challenged the participants with classifying blood vessels in 3D image stacks as stalled or flowing. In this setting, our approach outperformed other methods and demonstrated state-of-the-art results, achieving 85% Matthews correlation coefficient, 85% sensitivity, and 99.3% specificity. The source code for our solution is publicly available. Highlights: A state-of-the-art deep learning approach for detecting stalled capillaries in 3D microscopic brain images. Custom 3D image augmentations and 2D-to-3D model weight transfer substantially improve performance. Best model achieves 85% Matthews correlation coefficient, 85% sensitivity, and 99.3% specificity. 1st place in the "Clog Loss: Advance Alzheimer's Research with Stall Catchers" machine learning competition. 3D augmentations library, trained models, and source code are publicly available at GitHub.