Rapid determination of microbial growth and antimicrobial susceptibility
Inventors
Shamsheyeva, Alena • Howson, David C. • Metzger, Steven W.
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11603550-B2
Publication Date
2023-03-14
Expiration Date
Abstract
Systems and methods for rapid determination of microorganism growth and antimicrobial agent susceptibility and/or resistance are disclosed.
Core Innovation
The invention relates to a computer-based system that detects and analyzes immobilized microorganisms in gel medium images by using event-driven capture and image-derived signal analysis. The system detects signals, spots, or singularities in a first image of a sample and determines a first subset that is greater than a predetermined threshold by wavelet transformation, generating wavelet coefficients representing the image on a plurality of levels of resolution to create a stack of image planes.
The system determines a first value associated with an attribute of the immobilized microorganism based on first information associated with the threshold-exceeding subset. The attribute comprises at least one of microorganism size, shape, mass, intrinsic fluorescence, cell surface feature membrane integrity, genomic DNA, ribosomal RNA subunits, or a bound marker used in indirect immunofluorescence microscopy. The system then detects a second set of signals, spots, or singularities in a second image after an occurrence of an event and determines a second threshold-exceeding subset using a second wavelet transformation that creates a stack of second image planes.
Using the first value and the second value, the system determines a growth rate of the immobilized microorganism and compares the growth rate to a control growth rate. The system determines a clone signal intensity curve shape likelihood and a tracking error likelihood, and calculates a growth likelihood value based on the clone signal intensity curve shape likelihood and the tracking error likelihood. From the growth likelihood value, the system determines microorganism susceptibility.
Claims Coverage
Independent claim clm-00001 provides 5 inventive features covering wavelet-threshold image decomposition, attribute-value computation, event-driven before/after capture, growth-rate determination versus control, and likelihood-based growth likelihood calculation using clone curve shape and tracking error.
Wavelet threshold detection on multiple resolution levels for immobilized microorganisms
Detecting signals, spots, or singularities in a first image and determining that a first subset is greater than a predetermined threshold by employing a first wavelet transformation to decompose the first set into wavelet functions, applying the wavelet functions, and generating wavelet coefficients representing the first image on a plurality of levels of resolution to create a stack of image planes
Attribute value determination from threshold-exceeding image information
Determining a first value associated with an attribute of the immobilized microorganism based on first information associated with the first subset, wherein the attribute comprises at least one of microorganism size, shape, mass, intrinsic fluorescence, cell surface feature membrane integrity, genomic DNA, ribosomal RNA subunits, or bound marker used in indirect immunofluorescence microscopy
Event-driven second image detection and wavelet-threshold decomposition
Detecting a second set of signals, spots, or singularities in a second image after an occurrence of an event and determining that a second subset is greater than a predetermined threshold by employing a second wavelet transformation to decompose the second set into wavelet functions, applying the wavelet functions, and generating wavelet coefficients representing the second image on a plurality of levels of resolution to create a stack of second image planes
Growth rate determination and comparison to a control growth rate
Determining a growth rate of the immobilized microorganism associated with the first value and the second value, comparing the growth rate to a control growth rate, and using likelihood measures including clone signal intensity curve shape likelihood and tracking error likelihood
Growth likelihood calculation from clone curve shape likelihood and tracking error likelihood
Calculating a growth likelihood value based on the clone signal intensity curve shape likelihood and the tracking error likelihood
The independent claim centers on wavelet-transformed, threshold-based detection of image signals, spots, or singularities at multiple resolution levels, mapping threshold-exceeding information to attribute values for immobilized microorganisms, computing growth rate from before and after event images, and deriving growth likelihood using clone signal intensity curve shape likelihood and tracking error likelihood for susceptibility determination.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?