scholarly journals Generalization Challenges in Drug-Resistant Tuberculosis Detection from Chest X-rays

Diagnostics ◽  
2022 ◽  
Vol 12 (1) ◽  
pp. 188
Author(s):  
Manohar Karki ◽  
Karthik Kantipudi ◽  
Feng Yang ◽  
Hang Yu ◽  
Yi Xiang J. Wang ◽  
...  

Classification of drug-resistant tuberculosis (DR-TB) and drug-sensitive tuberculosis (DS-TB) from chest radiographs remains an open problem. Our previous cross validation performance on publicly available chest X-ray (CXR) data combined with image augmentation, the addition of synthetically generated and publicly available images achieved a performance of 85% AUC with a deep convolutional neural network (CNN). However, when we evaluated the CNN model trained to classify DR-TB and DS-TB on unseen data, significant performance degradation was observed (65% AUC). Hence, in this paper, we investigate the generalizability of our models on images from a held out country’s dataset. We explore the extent of the problem and the possible reasons behind the lack of good generalization. A comparison of radiologist-annotated lesion locations in the lung and the trained model’s localization of areas of interest, using GradCAM, did not show much overlap. Using the same network architecture, a multi-country classifier was able to identify the country of origin of the X-ray with high accuracy (86%), suggesting that image acquisition differences and the distribution of non-pathological and non-anatomical aspects of the images are affecting the generalization and localization of the drug resistance classification model as well. When CXR images were severely corrupted, the performance on the validation set was still better than 60% AUC. The model overfitted to the data from countries in the cross validation set but did not generalize to the held out country. Finally, we applied a multi-task based approach that uses prior TB lesions location information to guide the classifier network to focus its attention on improving the generalization performance on the held out set from another country to 68% AUC.

2021 ◽  
Vol 2071 (1) ◽  
pp. 012001
Author(s):  
J Ureta ◽  
A Shrestha

Abstract Tuberculosis(TB) is one of the top 10 causes of death worldwide, and drug-resistant TB is a major public health concern especially in resource-constrained countries. In such countries, molecular diagnosis of drug-resistant TB remains a challenge; and imaging tools such as X-rays, which are cheaply and widely available, can be a valuable supplemental resource for early detection and screening. This study uses a specialized convolutional neural network to perform binary classification of chest X-ray images to classify drug-resistant and drug-sensitive TB. The models were trained and validated using the TBPortals dataset which contains 2,973 labeled X-ray images from TB patients. The classifiers were able to identify the presence or absence of drug-resistant Tuberculosis with an AUROC between 0.66–0.67, which is an improvement over previous attempts using deep learning networks.


2020 ◽  
Vol 18 (2) ◽  
pp. 115-121
Author(s):  
Mariia V. Pavlova ◽  
Tatiana I. Vinogradova ◽  
Natalia V. Zabolotnykh ◽  
Elena S. Ershova ◽  
Nadezhda V. Sapozhnikova ◽  
...  

The aim of this work was an experimental and clinical study of increasing of the efficiency of therapy of respiratory tuberculosis with drug resistant pathogen by including a combination of anti-tuberculosis drugs of new generation Bedaquiline (Bq) and Thioureidoiminomethylpyridi-nium perchlorate (perchlozone, Tpp) into treatment regimens. The presented results were obtained: in the experiment on a model of tuberculosis infection in 103 male C57black / 6 mice, reproduced by inserting into the lateral vein of the tail of a suspense of clinical strains of Mycobacterium tuberculosis with multi-drug resistance (MDR) and with various combinations of mutations in the genes; in the clinic on the basis of examination and treatment of 148 patients with multidrug-resistant respiratory tuberculosis. It is shown that the application of the therapy schemes for multiple and broad drug resistant tuberculosis, containing Bq and Tpp in combination with drugs, selected taking into account the drug sensitivity of mycobacteria, allows to significantly reduce the periods of relief of symptoms of intoxication, regression of inflammatory changes, abacillation and achieving positive X-ray dynamics. Undesirable phenomena against the background of combined prescription of new generation drugs, as a rule, corresponded to light and moderate severity, and in frequency of development and expression did not differ from control groups of observation.


Sign in / Sign up

Export Citation Format

Share Document