|
Vol.11, No.3, August 2022. ISSN: 2217-8309 eISSN: 2217-8333
TEM Journal
TECHNOLOGY, EDUCATION, MANAGEMENT, INFORMATICS Association for Information Communication Technology Education and Science |
Convolutional Neural Network as an Image Processing Technique for Classification of Bacilli Tuberculosis Extra Pulmonary (TBEP) Disease
Bob Subhan Riza, Jufriadif Na’am, S Sumijan
© 2022 Bob Subhan Riza, published by UIKTEN. This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. (CC BY-NC-ND 4.0)
Citation Information: TEM Journal. Volume 11, Issue 3, Pages 1331-1340, ISSN 2217-8309, DOI: 10.18421/TEM113-43, August 2022.
Received: 24 April 2022. Revised: 08 August 2022.
Abstract:
Tuberculosis Extra Pulmonary (TBEP) is an infectious disease caused by the bacterium Mycobacterium tuberculosis and can cause death. Patients suffering from this disease must be treated quickly without waiting for a long time. Biopsy is one of the techniques used to take the patient's lung fluid and given Ziehl Neelsen chemical dye and then observed using a microscope to determine this TBEP disease. In this research, the TBEP detection process was developed using a classification method, namely CNN with feature extraction and feature selection. The feature uses 5 features where these features are a combination of shape features and texture features with the highest information gain value. From the results of research conducted through the training and testing stages of the classification method using feature selection, the accuracy rate is higher than not using feature selection with a comparison of the feature selection stage increasing 0.6536% for the training process, and 0.8942% for the testing process.
Keywords – Tuberculosis Extra Pulmonary, Otsu Thresholding, Hue Saturation Value (HSV), feature extraction, feature selection, Convolutional Neural Network. |
|
----------------------------------------------------------------------------------------------------------- ----------------------------------------------------------------------------------------------------------- |