On the application of optical forward-scattering to bacterial identification in an automated clinical analysis perspective

Abstract

The Optical Forward Scattering (OFS) technique can be used to identify pathogens by direct observation of bacteria colonies growing on a culture plate. The identification is based on the acquisition of scattering images from isolated colonies and their subsequent comparison with reference images acquired from known bacteria. The technique has been mainly studied for the identification of pathogens in the food-safety field. This paper focuses on the possibility of extending the applicability of the technique to the field of clinical laboratory automation. This scenario requires that the paradigm of image acquisition at fixed colony-dimension, well established in the food-safety applications, should be substituted by an acquisition at fixed incubation time. As a consequence, the scatterometer must be adjustable in real-time for adapting to the actual features of the bacterial colony. The paper describes an OFS system prototype qualified by the possibility to tune both the laser beam diameter and the acquisition camera field of view.Preliminary experiments on bacteria cultures from pathogens causing infections of the urinary tract show that the proposed approach is promising for the development of an Submitted to Biosensor and Bioelectronics 2 automated bacteria identification station. The new OFS approach also involves an alternative method for building a reference image database for subsequent image analysis.

Automatic hemolysis identification on aligned dual-lighting images of cultured blood agar plates

“…Another processing and classification chain has been proposed in [30] for the classification of major urinary infections. Other works approach bacteria identification using hyperspectral datasets [31,32,33] or forward-scattering techniques [34,35]. All the above works concentrate on bacterial colonies without considering their possible interactions with the growing medium, while here we address for the first time the problem of automated hemolysis analysis, which involves the interaction between growing pathogens and blood agar which is of high diagnostic significance.…”

Development of a multispectral light‐scatter sensor for bacterial colonies

“…A label‐free, non‐destructive, automated detection technique called BARDOT (BActerial Rapid Detection using Optical scattering Technology) based on elastic‐light‐scatter (ELS) patterns of bacteria colonies was developed for rapid detection and classification of microbial organisms . The applicability of the technology were reported for different organisms using a single‐wavelength laser and a varying number of genera or species . The merit of this technology is that the interrogation photons interact with the whole volume of a colony, thus collecting better phenotypic characteristics than reflective imaging.…”

Reflected scatterometry for noninvasive interrogation of bacterial colonies

“…6 As the need for rapid identification and classification of microbial organisms increases in various fields such as food security, clinical studies, and biosurveillance, label-free optical diagnostics has been studied by several research groups owing to several merits of this technology. [7][8][9] To provide more robust and accurate screening of these patterns, machine-learning techniques such as support vector machines (SVMs) were applied to features extracted from scatter patterns. 10 In addition, the optical origin of the patterns was investigated based on elastic-lightscatter phenomena for a single wavelength, 11 multiwavelengths, 12 and a speckle analysis.…”