Identifying candidate cells using image analysis with intensity levels

• Main Authors: Shao, Hung-Jen, Wang, XiaoMing, Hsieh, Huangpin B, Lai, Jr-Ming, Mei, Rui, Wu, Jen-Chia
• Format: Patent
• Language: English
• Published: 09.01.2024
• Subjects: BEER; BIOCHEMISTRY; CALCULATING; CHEMISTRY; COMPOSITIONS OR TEST PAPERS THEREFOR; COMPUTING; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL ORENZYMOLOGICAL PROCESSES; COUNTING; ENZYMOLOGY; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIRCHEMICAL OR PHYSICAL PROPERTIES; MEASURING; MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEICACIDS OR MICROORGANISMS; METALLURGY; MICROBIOLOGY; MUTATION OR GENETIC ENGINEERING; PHYSICS; PROCESSES OF PREPARING SUCH COMPOSITIONS; SPIRITS; TESTING; VINEGAR; WINE

Techniques for identifying and enumerating candidate target cells within a biological fluid specimen are described. A digital image of the biological fluid specimen is received, and one or more candidate regions of pixels in the digital image are identified by identifying connected regions of pixels of a minimum intensity having a size between a minimum size and a maximum size and an aspect ratio that meets a threshold. For each candidate region of at least one of the one or more candidate region, whether the portion of the image corresponding to the candidate region includes more than a threshold number of intensity levels is determined. If the portion of the image corresponding to the candidate region includes more than the threshold number of intensity levels the portion of the image is continued to be treated as a candidate for classification.