System For Detecting Infectious Agents Using Computer-Controlled Automated Image Analysis

Abstract

A method for providing quantitative information regarding the extent of infection of host cells by an infectious agent. A microscope image of a specimen of a bodily fluid is analyzed using image processing techniques to quantify the percentage of the area of the specimen that is infected.

What is claimed: 
     
         1 . A computer software product, comprising a computer-readable storage medium having fixed therein a sequence of instructions which, when executed by a computer, direct the performance of a method which comprises:
 a) acquiring a microscope image of an optical field of a substrate having fixed thereon a monolayer of animal cells which either produce or are treated to produce a first signal specific to an animal cell of interest and treated to produce a second signal specific to an infectious agent of interest, if present, and transferring the image to an ROB image;   b) transferring the Red component of the RGB image to a new monochrome grey image;   c) transforming the grey level image to a binary image using a cut off point set to a value indicative of the expected size of the animal cells of interest;   d) operating on the binary image to remove noise and fill holes;   e) measuring size of the image of step d), selecting and recording areas representative of the animal cells of interest;   f) transferring the Red component of the original ROB image to a second binary image using an expected value indicative of the infectious agent of interest;   g) transferring the Green component of the original RGB image to grey level;   h) forming a new grey level image using all pixels having a value equal to a set value M and any grey level value in the Green component in the original RGB image equal to a set value N;   i) transforming the new grey level images to a third binary image;   j) operating on the third binary image to remove noise and fill holes;   k) recording the area of the third binary image after step j); and   l) determining whether or not, each identified animal cell is occupied by an infectious agent of interest.   
     
     
         2 . The computer software product in accordance with  claim 1 , wherein the method further comprises calculating the extent of each animal cell occupied by said infectious agent. 
     
     
         3 . The computer software product in accordance with  claim 1 , further comprising a computer-readable storage medium having fixed therein a sequence of instructions which, when executed by a computer, direct the performance of a method which comprises:
 a) acquiring a microscope image of an optical field of a substrate having fixed thereon a sample of a body fluid or tissue from an animal patient suspected of containing an infectious agent of interest, which sample either produces or is treated to produce a first signal specific to an infectious agent of interest, if present, and transferring the image to an RGB image;   b) transferring the RGB image to an HLS image and transforming the HIS image to a new monochrome grey level image;   c) transforming the grey level image to a binary image;   d) operating on the binary image using at least one filter to identify blobs; and   e) selecting blobs based on the expected size of area of an infectious agent to determining whether or not an infectious agent of interest is present in the sample.