Automated Capillary Analysis Using Deep Learning and Optical Flow
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for analyzing capillaries in microvascular videos are time-consuming, labor-intensive, and prone to errors, limiting their integration into routine clinical practice and restricting the analysis of parameters such as capillary density, velocity, and morphology.
Innovation Solution
An automated method using deep learning and rule-based algorithms to detect capillaries and classify erythrocyte velocity in microvascular videos, involving the generation of capillary candidate maps, optical flow determination, and extraction of capillary parameters through neural networks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual analysis of capillaries is performed by trained researchers, then measurement precision is improved, but productivity deteriorates
Solution Approach 1:
The patent replaces the manual mechanical analysis process with an automated computer-based system using image processing and machine learning algorithms. The system automatically detects capillaries, tracks red blood cells, and calculates flow velocity parameters, eliminating the need for manual observation and measurement while maintaining accuracy through trained neural networks.
Solution Approach 2:
The system enables self-service by allowing the computer to autonomously perform capillary analysis without continuous human intervention. The automated algorithm independently identifies capillaries, tracks blood flow, and generates measurements, making the system self-sufficient in performing tasks that previously required trained researchers.
2Productivity
If automated methods are used for capillary analysis, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent replaces manual mechanical analysis with automated computer-based image processing and machine learning algorithms that can rapidly analyze capillary networks while maintaining measurement accuracy through trained neural networks and sophisticated detection algorithms.
Solution Approach 2:
The system achieves self-service capability by autonomously performing capillary detection, red blood cell tracking, and flow velocity calculation without human intervention, enabling rapid automated analysis that maintains precision through algorithmic accuracy.
3Adaptability or versatility
If multiple capillary parameters are analyzed simultaneously, then adaptability is improved, but device complexity deteriorates
Solution Approach 1:
The patent implements a universal automated system that can simultaneously analyze multiple capillary parameters including density, flow velocity, and morphology. The single integrated platform performs diverse functions through unified image processing and machine learning algorithms, eliminating the need for separate specialized tools for each parameter.
Solution Approach 2:
The system merges multiple analysis functions into a single integrated automated platform. By combining capillary detection, red blood cell tracking, and flow velocity calculation into one unified system, the patent reduces the complexity that would arise from using separate tools while enhancing adaptability to analyze various parameters simultaneously.
Data Source
AI summary
An automated method for analysing capillaries in a plurality of images acquired from a subject. The method comprising the steps of: a) acquiring the plurality of images; b) generating a plurality of capillary candidate maps for each of said images, each capillary candidate map comprising one or more regions of interest for each of said images, wherein for each image, each of the respective capillary candidate maps is generated by comparing said image to a different criterion; c) combining said capillary candidate maps to generate a combined capillary candidate map; d) using a first neural network to determine a respective location of one or more detected capillaries in said combined capillary candidate map; e) using a second neural network to determine an optical flow of said detected capillaries; and f) extracting one or more capillary parameters using said detected capillaries and/or said determined flow.