METHOD OF AUTOMATED TUBERCULOSIS EXAMINATION OF A SPUTUM SMEAR IMAGE

Disclosed is a method of conducting a tuberculosis examination. The method includes a specimen image storage step of storing a sputum smear image photographed using a microscope, a specimen feature data extraction step of extracting feature data from the stored sputum smear image, a prelearned model...

Full description

Saved in:
Bibliographic Details
Main Authors KIM, Moon Gi, CHOI, Myung Jin, KIM, Tae Young, PARK, Hyun Woo
Format Patent
LanguageEnglish
French
German
Published 13.12.2017
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:Disclosed is a method of conducting a tuberculosis examination. The method includes a specimen image storage step of storing a sputum smear image photographed using a microscope, a specimen feature data extraction step of extracting feature data from the stored sputum smear image, a prelearned model adoption classification step of performing learning using an artificial neural network using a sample image in which the presence or the absence of tuberculosis bacilli is known in advance, and a step of displaying whether or not a specimen has tuberculosis by inputting the extracted feature data of the specimen into a pre-learned model to identify whether or not the specimen has tuberculosis.
Bibliography:Application Number: EP20170172214