Two-step machine learning method for the rapid analysis of microvascular flow in intravital video microscopy

Microvascular blood flow is crucial for tissue and organ function and is often severely affected by diseases. Therefore, investigating the microvasculature under different pathological circumstances is essential to understand the role of the microcirculation in health and sickness. Microvascular blood flow is generally investigated with Intravital Video Microscopy (IVM), and the captured images are stored on a computer for later off-line analysis. The analysis of these images is a manual and challenging process, evaluating experiments very time consuming and susceptible to human error. Since more advanced digital cameras are used in IVM, the experimental data volume will also increase significantly. This study presents a new two-step image processing algorithm that uses a trained Convolutional Neural Network (CNN) to functionally analyze IVM microscopic images without the need for manual analysis. While the first step uses a modified vessel segmentation algorithm to extract the location of vessel-like structures, the second step uses a 3D-CNN to assess whether the vessel-like structures have blood flowing in it or not. We demonstrate that our two-step algorithm can efficiently analyze IVM image data with high accuracy (83%). To our knowledge, this is the first application of machine learning for the functional analysis of microvascular blood flow in vivo.

Modeling of Cerebral Oxygen Transport Based on In vivo Microscopic Imaging of Microvascular Network Structure, Blood Flow, and Oxygenation

Oxygen is delivered to brain tissue by a dense network of microvessels, which actively control cerebral blood flow (CBF) through vasodilation and contraction in response to changing levels of neural activity. Understanding these network-level processes is immediately relevant for (1) interpretation of functional Magnetic Resonance Imaging (fMRI) signals, and (2) investigation of neurological diseases in which a deterioration of neurovascular and neuro-metabolic physiology contributes to motor and cognitive decline. Experimental data on the structure, flow and oxygen levels of microvascular networks are needed, together with theoretical methods to integrate this information and predict physiologically relevant properties that are not directly measurable. Recent progress in optical imaging technologies for high-resolution in vivo measurement of the cerebral microvascular architecture, blood flow, and oxygenation enables construction of detailed computational models of cerebral hemodynamics and oxygen transport based on realistic three-dimensional microvascular networks. In this article, we review state-of-the-art optical microscopy technologies for quantitative in vivo imaging of cerebral microvascular structure, blood flow and oxygenation, and theoretical methods that utilize such data to generate spatially resolved models for blood flow and oxygen transport. These “bottom-up” models are essential for the understanding of the processes governing brain oxygenation in normal and disease states and for eventual translation of the lessons learned from animal studies to humans.