Detection method and system for qualitatively and quantitatively detecting different wound bacteria based on machine learning

Abstract

The invention relates to a method for qualitatively and quantitatively detecting bacteria on different wounds based on machine learning, which comprises the following steps of: S1, acquiring a prepared agar plate, and acquiring a plate bacterial colony image; s2, adopting an absdiff difference algorithm of OpenCV for the plate bacterial colony image to obtain a difference result between each subsequent time point of the plate bacterial colony image and the background image, and performing preprocessing to obtain a training set; step S3, constructing and training a Desection Net binary classification detection network; and S4, inputting a plate bacterial colony image of a to-be-detected wound surface sample into the trained classification detection network, and determining different types of bacteria in the sample. The method has accurate qualitative and quantitative ability to culture early bacteria, and the detection efficiency is effectively improved.

Description

technical field [0001] The invention belongs to the field of biological detection, and in particular relates to a detection method and system for qualitatively and quantitatively different wound bacteria based on machine learning. Background technique [0002] The most common harmful pathogens on wounds include Pseudomonas aeruginosa (PA), Acinetobacter baumannii (AB), Staphylococcus (SA) and Escherichia coli (EC) . At present, there are three commonly used methods for the quantitative detection of pathogenic bacteria in clinical practice, namely plate culture counting method, molecular biology detection method and immunological detection method. Polymerase chain reaction and immunoassay of specific antigen-antibody interaction have greatly improved the detection sensitivity, but false positive results cannot be avoided, and the detection cost is expensive. Bacterial culture counting is the gold standard for clinical detection of bacteria, which is cost-effective, but this...

AI Technical Summary

Problems solved by technology

Polymerase chain reaction and immunoassays for specific antigen-antibody interactions have greatly improved detection sensitivity, but false positive results cannot be avoided, and the detection cost is expensive

Bacterial culture counting is the gold standard for clinical detection of bacteria, which is cost-effective, but this traditional detection method based on bacterial culture requires bacterial colonies to grow to a certain size visible to the naked eye, so it takes a long time to detect bacteria

Images