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

Ossama Mahmoud,Barry G H Janssen,Mahmoud El-Sakka

doi:10.1038/s41598-021-89469-w

Ossama Mahmoud, Barry G H Janssen + Show 1 more

Open Access

https://doi.org/10.1038/s41598-021-89469-w

Copy DOI

Abstract

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.

Highlights

Microvascular blood flow is crucial for tissue and organ function and is often severely affected by diseases
One set (392 videos containing 27,809 data points; 20,382 positive and 7427 negative samples) was used for 3D-Convolutional Neural Network (CNN) training only, while a different collection of 66 videos (4404 labeled data points; 3362 positive and 1042 negative samples) which had not been used for any previous training, was used for testing
Our results suggest that the 3D Convoluted Neural Network (3D-CNN) can efficiently predict and identify which vessel segments in the microscopic Field of View (FOV) during Intravital Video Microscopy (IVM) are perfused with blood flow

Summary

Introduction

Microvascular blood flow is crucial for tissue and organ function and is often severely affected by diseases. Analysis of the 3D-CNN performance on the 4404 data points shows that the network has an overall accuracy of 90% in determining whether blood flow in a vessel segment is flowing or not.

Results

Conclusion