Paper Info
Title
A Clinical TB Detection Method Based on Molecular Typing Technique with Quality Control.
Abstract
The gold standard for diagnosing pulmonary Mycobacterium tuberculosis (TB) is the detection of tubercle bacillus in patient sputum samples. However, current methods either require long waiting times to culture the bacteria or have a risk of getting false-positive results due to cross-contamination. In this study, a method to detect tubercle bacillus based on the molecular typing technique is presented. This method can detect genetic units, variable number of tandem repeat (VNTR), which are the characteristic of tuberculosis (TB), and performs quality control using a mathematical model, ensuring the reliability of the results. Compared to other methods, the proposed method was able to process and diagnose a large volume of samples in a run time of six hours, with high sensitivity and specificity. Our method is also in the pipeline for implementation in clinical testing. Reliable and confirmed results are stored into a database, and these data are used to further refine the model. As the volume of data processed from reliable samples increases, the diagnostic power of the model improves. In addition to improving the quality control scheme, the collected data can be also used to support other TB research, such as that regarding the evolution of the tubercle bacillus.
Year
DOI
Venue
2019
10.1155/2019/9872425
COMPUTATIONAL AND MATHEMATICAL METHODS IN MEDICINE
Field
DocType
Volume
Tandem repeat,Tubercle,Mycobacterium tuberculosis,Pattern recognition,Computer science,Typing,Sputum,Artificial intelligence,Tuberculosis,Machine learning
Journal
2019
ISSN
Citations 
PageRank 
1748-670X
0
0.34
References 
Authors
0
6
Name
Order
Citations
PageRank
Tienan Feng100.34
Yan Cheng254.17
Suwen Yu300.34
Feng Jiang42810.30
Min Su500.34
Jin Chen600.34