Research On Rapid And Label-Free Bacterial Pathogens Identification System Based On Multi-angle Dynamic Light Scattering

The rapid detection and classification of bacterial pathogens have a good many medical utilities and widespread applications in clinical labs,intensive care units,and infectious disease departments.Failure in the detection of pathogenic contamination in food and water may not only affect the economic losses to the food industry but also have dramatic consequences on a healthy life.The bacterial contamination in food and drinking water may cause a fatal outbreak among people and can have a significant public health impact,especially in populated areas.The early detection of microbial pathogens helps in reducing food contamination and improving food quality.The prevention of pathogenic bacteria in hospitals can lower the risk of infections for patients,especially in intensive care units.Extensive research has been carried out for finding innovative techniques for the rapid identification of pathogens.The aim of research is to reduce the time duration,lower the cost,and make a very simple method for operation.The proposed system for rapid and label-free identification of pathogens is based on light scattering from the bacterial microbes.The prototype consists of three parts:laser beam,an assembly of photodetectors,and the data acquisition system.The bacterial testing sample is mixed with 10 m L of distilled water and placed inside the machine chamber.When the bacterial microbes pass by the laser beam,light is absorbed,refracted,and scattered by the sample.The surrounding photodetectors collected the scattered light from the sample and then converted the light signal into an electrical signal(voltage).The acquired signal is processed further for the classification of the sample by means of different algorithms.Different algorithms and methodologies were developed to classify the pathogens.The algorithm is based on the binned plots,and the peak values were acquired from the data for creating 3D histograms by pairing photodetectors.The 3D histograms are used for evaluating the frequency of occurrence.The algorithm of the system consists of two parts: Library files and the Comparator.Library files contain data of bacterial species in the form of binned plots,while comparator compares the data of test sample with library files.The classification of the sample depends on the maximum resemblance of the number of binned plots with library files.The classification of Enterococcus faecalis,Staphylococcus aureus,and Escherichia coli gives mean accuracies of 81.8 %,70.9 %,and 71.4 %,respectively.Machine learning approaches were utilized to increase the accuracy of the algorithm.Also,the number of surrounding photodetectors was reduced to twelve to minimize the computational power of the algorithm.The photodetectors were positioned at different locations according to MIE scattering theory to measure the intensity of scattered light.The waveform features were acquired using the power spectral characteristics,and the dimensionality of extracted features was reduced by means of minimal-redundancy-maximal-relevance criterion(m RMR).The selected power spectral features were used for training a Support Vector Machine(SVM)model for the classification of three different bacterial microbes.The resulting average identification accuracies of Enterococcus faecalis,Escherichia coli and Staphylococcus aureus were 99%,87%,and 94%,respectively.The overall experimental results yield higher identification accuracy of 93.6%,indicating that the given system has the potential for rapid and label-free identification of pathogens.The proposed microbial identification technique is rapid,label-free,and very simple to operate.Bacterial identification works on the detection of scattered laser light from the sample.In this research,a quick sample preparation method was used that utilizes the basic laboratory equipment before testing.The methods are conventional and easy to implement in a clinical setting.The algorithm of the system can be modified to add more data of new pathogen species.The proposed system can be applied to future real-time intelligent theranostic systems for diagnosis and treatment of pathogenic diseases.