The Research And Development Of Surgical Blood Loss Monitoring System Based On Machine Vision

With the booming development of manufacturing industry,the maturity and development of machine vision technology,its application areas are becoming more and more extensive.Compared with the traditional manual processing methods,machine vision has high efficiency,high precision and l ong working hours in production.It has become an unstoppable trend to apply machine vision to surgical procedures to help doctors deal with problems more time ly and accurately.In this thesis,we study the blood loss assessment methods in surgery,and in response to the shortcomings of which such as large assessment errors and no timeliness,we study and design a surgical blood loss monitoring system based on machine vision technology,which can provide doctors with timely information about patients’ blood loss in order to make timely treatment plans.The research and design of the system is mainly described from the following aspects.First,according to the domestic and foreign development status of th e blood loss monitoring and evaluation system,combined with the actual situation in my country,a surgical blood loss monitoring system based on machine vision technology is proposed.After analyzing the needs of the blood loss monitoring system,design t he overall architecture of the system,and make analysis,research and selection of the system’s hemoglobin concentration monitoring,blood loss monitoring and data management technical solutions.Next,the image pre-processing and image recognition algori thms are studied and analyzed.The image of the mixed solution in the flushing solution collection bucket is the main basis for the system data analysis.The target localization segmentation is performed on the acquisition image and the color histogram and color moment features of the target image are extracted,the sample image that is most similar to the acquisition image is found from the sample library image using the weighted Euclidean distance and intersection method,and the color difference is calcu lated for calibration analysis,and the image data that meet the conditions are used as the data information of the blood loss situation.Then,for the machine vision technology,the industrial cameras and other equipment in the blood loss monitoring syste m are selected and the test environment is built,the C/S architecture is selected for the development of the system,and the SQL Server 2014 database is used as the data interaction center,and the call is installed in the Microsoft Visual Studio 2013 int egrated development platform Open CVSharp vision library is installed in Microsoft Visual Studio 2013 integrated development platform to complete the processing and analysis of images,and C#language is used to realize the system software development.The system includes user verification module,real-time monitoring module,image processing module,data management module,and alarm module,thus realizing the whole system functions.Finally,the system function,performance and blood loss were tested and analyzed,and the results showed that the system function basically achieved the expected goal,good performance,stable operation,and real-time accurate monitoring of blood loss within an allowable error range of ±5%.