1
|
Fang WJ, Tang SN, Liang RY, Zheng QT, Yao DQ, Hu JX, Song M, Zheng GP, Rosenthal A, Tartakovsky M, Lu PX, Wáng YXJ. Differences in pulmonary nodular consolidation and pulmonary cavity among drug-sensitive, rifampicin-resistant and multi-drug resistant tuberculosis patients: the Guangzhou computerized tomography study. Quant Imaging Med Surg 2024; 14:1010-1021. [PMID: 38223080 PMCID: PMC10783999 DOI: 10.21037/qims-23-694] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/18/2023] [Accepted: 10/28/2023] [Indexed: 01/16/2024]
Abstract
Background Pulmonary nodular consolidation (PN) and pulmonary cavity (PC) may represent the two most promising imaging signs in differentiating multidrug-resistant (MDR)-pulmonary tuberculosis (PTB) from drug-sensitive (DS)-PTB. However, there have been concerns that literature described radiological feature differences between DS-PTB and MDR-PTB were confounded by that MDR-PTB cases tend to have a longer history. This study seeks to further clarify this point. Methods All cases were from the Guangzhou Chest Hospital, Guangzhou, China. We retrieved data of consecutive new MDR cases [n=46, inclusive of rifampicin-resistant (RR) cases] treated during the period of July 2020 and December 2021, and according to the electronic case archiving system records, the main PTB-related symptoms/signs history was ≤3 months till the first computed tomography (CT) scan in Guangzhou Chest Hospital was taken. To pair the MDR-PTB cases with assumed equal disease history length, we additionally retrieved data of 46 cases of DS-PTB patients. Twenty-two of the DS patients and 30 of the MDR patients were from rural communities. The first CT in Guangzhou Chest Hospital was analysed in this study. When the CT was taken, most cases had anti-TB drug treatment for less than 2 weeks, and none had been treated for more than 3 weeks. Results Apparent CT signs associated with chronicity were noted in 10 cases in the DS group (10/46) and 9 cases in the MDR group (10/46). Thus, the overall disease history would have been longer than the assumed <3 months. Still, the history length difference between DS patients and MDR patients in the current study might not be substantial. The lung volume involvement was 11.3%±8.3% for DS cases and 8.4%±6.6% for MDR cases (P=0.022). There was no statistical difference between DS cases and MDR cases both in PN prevalence and in PC prevalence. For positive cases, MDR cases had more PN number (mean of positive cases: 2.63 vs. 2.28, P=0.38) and PC number (mean of positive cases: 2.14 vs. 1.38, P=0.001) than DS cases. Receiver operating characteristic curve analysis shows, PN ≥4 and PC ≥3 had a specificity of 86% (sensitivity 25%) and 93% (sensitivity 36%), respectively, in suggesting the patient being a MDR cases. Conclusions A combination of PN and PC features allows statistical separation of DS and MDR cases.
Collapse
Affiliation(s)
- Wei-Jun Fang
- Department of Radiology, Guangzhou Chest Hospital, State Key Laboratory of Respiratory Disease, Guangzhou, China
| | - Sheng-Nan Tang
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong SAR, China
| | - Rui-Yun Liang
- Department of Radiology, Guangzhou Chest Hospital, State Key Laboratory of Respiratory Disease, Guangzhou, China
| | - Qiu-Ting Zheng
- Department of Medical Imaging, Shenzhen Center for Chronic Disease Control, Shenzhen, China
| | - Dian-Qi Yao
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong SAR, China
| | - Jin-Xing Hu
- Department of Tuberculosis, Guangzhou Chest Hospital, State Key Laboratory of Respiratory Disease, Guangzhou, China
| | - Min Song
- Department of Radiology, Guangzhou Chest Hospital, State Key Laboratory of Respiratory Disease, Guangzhou, China
| | - Guang-Ping Zheng
- Department of Radiology, The Third People’s Hospital of Shenzhen, Shenzhen, China
| | - Alex Rosenthal
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, Bethesda, MD, USA
| | - Michael Tartakovsky
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, Bethesda, MD, USA
| | - Pu-Xuan Lu
- Department of Medical Imaging, Shenzhen Center for Chronic Disease Control, Shenzhen, China
| | - Yì Xiáng J. Wáng
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong SAR, China
| |
Collapse
|
2
|
Xu CJ, Lu PX, Li CH, He YL, Fang WJ, Xie RM, Jin GQ, Lu YB, Zheng QT, Zheng GP, Lv SX, Huang H, Li L, Ren M, Shi YX, Wen XN, Li L, Wei FJ, Hou DL, Lv Y, Shan F, Wu ZC, Hu ZL, Zhang XR, Liu DX, Shi WY, Li HR, Zhang N, Song M, Zhang X, Deng YY, Li J, Liu Q, Li D, Zhao L, Chen BD, Shi YB, Jiang FL, Tang X, Wu LJ, Ma W, Xu XY, Li HJ. Chinese expert consensus on imaging diagnosis of drug-resistant pulmonary tuberculosis. Quant Imaging Med Surg 2024; 14:1039-1060. [PMID: 38223121 PMCID: PMC10784038 DOI: 10.21037/qims-23-1223] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/27/2023] [Accepted: 09/23/2023] [Indexed: 01/16/2024]
Abstract
Tuberculosis (TB) remains one of the major infectious diseases in the world with a high incidence rate. Drug-resistant tuberculosis (DR-TB) is a key and difficult challenge in the prevention and treatment of TB. Early, rapid, and accurate diagnosis of DR-TB is essential for selecting appropriate and personalized treatment and is an important means of reducing disease transmission and mortality. In recent years, imaging diagnosis of DR-TB has developed rapidly, but there is a lack of consistent understanding. To this end, the Infectious Disease Imaging Group, Infectious Disease Branch, Chinese Research Hospital Association; Infectious Diseases Group of Chinese Medical Association of Radiology; Digital Health Committee of China Association for the Promotion of Science and Technology Industrialization, and other organizations, formed a group of TB experts across China. The conglomerate then considered the Chinese and international diagnosis and treatment status of DR-TB, China's clinical practice, and evidence-based medicine on the methodological requirements of guidelines and standards. After repeated discussion, the expert consensus of imaging diagnosis of DR-PB was proposed. This consensus includes clinical diagnosis and classification of DR-TB, selection of etiology and imaging examination [mainly X-ray and computed tomography (CT)], imaging manifestations, diagnosis, and differential diagnosis. This expert consensus is expected to improve the understanding of the imaging changes of DR-TB, as a starting point for timely detection of suspected DR-TB patients, and can effectively improve the efficiency of clinical diagnosis and achieve the purpose of early diagnosis and treatment of DR-TB.
Collapse
Affiliation(s)
- Chuan-Jun Xu
- Department of Radiology, The Second Hospital of Nanjing, Nanjing University of Chinese Medicine, Nanjing, China
| | - Pu-Xuan Lu
- Department of Medical Imaging, Shenzhen Center for Chronic Disease Control, Shenzhen, China
| | - Chun-Hua Li
- Department of Radiology, Chongqing Public Health Medical Center, Chongqing, China
| | - Yu-Lin He
- Department of Radiology, The First Affiliated Hospital of Nanchang University, Nanchang, China
| | - Wei-Jun Fang
- Department of Radiology, Guangzhou Chest Hospital, Guangzhou, China
| | - Ru-Ming Xie
- Department of Radiology, Beijing Ditan Hospital, Capital Medical University, Beijing, China
| | - Guan-Qiao Jin
- Department of Radiology, The Affiliated Cancer Hospital of Guangxi Medical University, Nanning, China
| | - Yi-Bo Lu
- Department of Radiology, The Fourth People’s Hospital of Nanning, Nanning, China
| | - Qiu-Ting Zheng
- Department of Medical Imaging, Shenzhen Center for Chronic Disease Control, Shenzhen, China
| | - Guang-Ping Zheng
- Department of Radiology, The Third People’s Hospital of Shenzhen, Shenzhen, China
| | - Sheng-Xiu Lv
- Department of Radiology, Chongqing Public Health Medical Center, Chongqing, China
| | - Hua Huang
- Department of Radiology, The Third People’s Hospital of Shenzhen, Shenzhen, China
| | - Li Li
- Department of Radiology, Beijing You’an Hospital, Capital Medical University, Beijing, China
| | - Meiji Ren
- Department of Radiology, Beijing You’an Hospital, Capital Medical University, Beijing, China
| | - Yu-Xin Shi
- Department of Radiology, Shanghai Public Health Clinical Center, Shanghai, China
| | - Xin-Nian Wen
- Department of Medical Imaging, Chest Hospital of Guangxi Zhuang Autonomous Region, Liuzhou, China
| | - Lin Li
- Department of Radiology, Linyi People’s Hospital, Linyi, China
| | - Fang-Jun Wei
- Department of Radiology, The Third People’s Hospital of Shenzhen, Shenzhen, China
| | - Dai-Lun Hou
- Department of Medical Imaging, Beijing Chest Hospital, Capital Medical University, Beijing, China
| | - Yan Lv
- Department of Medical Imaging, Beijing Chest Hospital, Capital Medical University, Beijing, China
| | - Fei Shan
- Department of Radiology, Shanghai Public Health Clinical Center, Shanghai, China
| | - Zheng-Can Wu
- Department of Radiology, Nanjing First Hospital, Nanjing Medical University, Nanjing, China
| | - Zhi-Liang Hu
- Department of Infectious Disease, The Second Hospital of Nanjing, Nanjing University of Chinese Medicine, Nanjing, China
| | - Xiang-Rong Zhang
- Department of Pulmonary Tuberculosis, The Second Hospital of Nanjing, Nanjing University of Chinese Medicine, Nanjing, China
| | - Du-Xian Liu
- Department of Pathology, The Second Hospital of Nanjing, Nanjing, China
| | - Wei-Ya Shi
- Department of Radiology, Shanghai Public Health Clinical Center, Shanghai, China
| | - Hui-Ru Li
- Department of Radiology, Guangzhou Chest Hospital, Guangzhou, China
| | - Na Zhang
- Department of Radiology, Public Health and Clinical Center of Chengdu, Chengdu, China
| | - Min Song
- Department of Radiology, Guangzhou Chest Hospital, Guangzhou, China
| | - Xin Zhang
- Department of Medical Imaging, The Fourth People’s Hospital of Huai’an, Huai’an, China
| | - Ying-Ying Deng
- Department of Radiology, Shenzhen Yantian District People’s Hospital, Shenzhen, China
| | - Jinlong Li
- Department of Laboratory Medicine, The Second Hospital of Nanjing, Nanjing University of Chinese Medicine, Nanjing, China
| | - Qiang Liu
- Department of Radiology, Shandong Provincial Hospital, Shandong First Medical University, Jinan, China
| | - Dechun Li
- Department of Radiology, Xuzhou Central Hospital, Xuzhou, China
| | - Lingling Zhao
- Department of Radiology, The Sixth Peoples Hospital of Zhengzhou, Zhengzhou, China
| | - Bu-Dong Chen
- Medical Imaging Quality Research Committee, China Quality Association for Pharmaceuticals, Beijing, China
| | - Yan-Bin Shi
- Department of Radiology, The Sixth Peoples Hospital of Zhengzhou, Zhengzhou, China
| | - Feng-Li Jiang
- Department of Radiology, Zhejiang Cancer Hospital, Hangzhou, China
| | - Xin Tang
- Department of Radiology, The First Affiliated Hospital of Nanchang University, Nanchang, China
| | - Li-Ji Wu
- Department of Imaging, Fourth Hospital of Inner Mongolia Autonomous, Hohhot, China
| | - Wei Ma
- Department of Radiology, The Third People’s Hospital of Longgang, Shenzhen, China
| | - Xin-Yue Xu
- The School of Radiation Medicine and Protection (SRMP) of Soochow University, Suzhou, China
| | - Hong-Jun Li
- Department of Radiology, Beijing You’an Hospital, Capital Medical University, Beijing, China
| |
Collapse
|
3
|
Oladimeji O, Adeniji-Sofoluwe AT, Othman Y, Adepoju VA, Oladimeji KE, Atiba BP, Anyiam FE, Odugbemi BA, Afolaranmi T, Zoakah AI. Chest X-ray Features in Drug-Resistant Tuberculosis Patients in Nigeria; a Retrospective Record Review. MEDICINES (BASEL, SWITZERLAND) 2022; 9:medicines9090046. [PMID: 36135827 PMCID: PMC9504772 DOI: 10.3390/medicines9090046] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 06/13/2022] [Revised: 08/22/2022] [Accepted: 08/25/2022] [Indexed: 11/16/2022]
Abstract
Chest X-ray (CXR) characteristics of patients with drug-resistant tuberculosis (DR-TB) depend on a variety of factors, and therefore, identifying the influence of these factors on the appearance of DR-TB in chest X-rays can help physicians improve diagnosis and clinical suspicion. Our aim was to describe the CXR presentation of patients with DR-TB and its association with clinical and demographic factors. A retrospective analysis of the CXRs of DR-TB patients in Nigeria between 2010 and 2016 was performed, reviewing features of chest radiographs, such as cavitation, opacity and effusion, infiltration and lung destruction. The association of these abnormal CXR findings with clinical and demographic characteristics was evaluated using bivariate and multivariate models, and a p-value < 0.05 was considered statistically significant with a 95% confidence interval. A total of 2555 DR-TB patients were studied, the majority (66.9%) were male, aged 29−38 years (36.8%), previously treated (77%), from the South West treatment zone (43.5%), HIV negative (76.7%) and bacteriologically diagnosed (89%). X-ray findings were abnormal in 97% of the participants, with cavitation being the most common (41.5%). Cavitation, effusion, fibrosis, and infiltration were higher in patients presenting in the South West zone and in those previously treated for DR-TB, while lung destruction was significantly higher in patients who are from the South South zone, and in those previously treated for DR-TB. Patients from the South East zone (AOR: 6.667, 95% CI: 1.383−32.138, p = 0.018), the North East zone (AOR: 6.667, 95% CI: 1.179−37.682, p = 0.032) and the North West zone (AOR: 6.30, 95% CI: 1.332−29.787, p = 0.020) had a significantly increased likelihood of abnormal chest X-ray findings, and prior TB treatment predisposed the patient to an increased likelihood of abnormal chest X-ray findings compared to new patients (AOR: 8.256, 95% CI: 3.718−18.330, p = 0.001). The finding of a significantly higher incidence of cavities, effusions and fibrosis in DR-TB patients previously treated could indicate late detection or presentation with advanced DR-TB disease, which may require a more individualized regimen or surgical intervention.
Collapse
Affiliation(s)
- Olanrewaju Oladimeji
- Department of Public Health, Faculty of Health Sciences, Walter Sisulu University, Mthatha 5099, South Africa
- Faculty of Health Sciences, Durban University of Technology, Durban 4001, South Africa
- Department of Community Medicine, University of Jos, Jos 930105, Nigeria
- Correspondence:
| | | | - Yasir Othman
- Department of Medicine, Hull University Teaching Hospitals NHS Trust, Hall University, Hull HU3 2JZ, UK
| | - Victor Abiola Adepoju
- Department of HIV and Infectious Diseases, Jhpiego (An Affiliate of John Hopkins University), Abuja 900271, Nigeria
| | - Kelechi Elizabeth Oladimeji
- Department of Public Health, Faculty of Health Sciences, Walter Sisulu University, Mthatha 5099, South Africa
| | - Bamidele Paul Atiba
- Faculty of Health Sciences, Durban University of Technology, Durban 4001, South Africa
| | - Felix Emeka Anyiam
- Faculty of Health Sciences, Durban University of Technology, Durban 4001, South Africa
| | - Babatunde A. Odugbemi
- Departments of Community Health & Primary Health Care, Lagos State University College of Medicine, Ikeja 102212, Nigeria
| | | | | |
Collapse
|
4
|
Song WM, Li SJ, Liu JY, Fu Q, Xu TT, Tao NN, Zhang QY, Liu SQ, An QQ, Zhu XH, Liu Y, Yu CB, Li YF, Dong J, Li HC. Impact of alcohol drinking and tobacco smoking on the drug-resistance of newly diagnosed tuberculosis: a retrospective cohort study in Shandong, China, during 2004-2020. BMJ Open 2022; 12:e059149. [PMID: 35902191 PMCID: PMC9341182 DOI: 10.1136/bmjopen-2021-059149] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/17/2022] Open
Abstract
OBJECTIVES To investigate the independent and collective impact of alcohol drinking and tobacco smoking on the drug-resistance of newly diagnosed tuberculosis (TB). DESIGN This was a retrospective cohort study. SETTING Shandong, China. PARTICIPANTS Patients with newly diagnosed TB from 1 January 2004 to 31 December 2020 were collected. Exclusive criteria: retreated cases; extrapulmonary tuberculosis; without information on drug susceptibility testing results, smoking or drinking habits; bacteriological identification as non-tuberculous mycobacteria. PRIMARY AND SECONDARY OUTCOME MEASURES Patients were classified into four groups including smokers only (G1), drinker only (G2), smoker +drinker (G3), non-smoker +non-drinker group (G0). We described the drug-resistant profiles, clinical factors and calculated the ORs of different drug-resistance among G1, G2, G3, compared with G0 through univariate and multivariate logistics regression models. RESULTS Of the 7996 TB cases enrolled, the proportions of G1, G2, G3 and G0 were 8.25%, 3.89%, 16.46% and 71.40%, respectively. The rates of drug-resistant (DR)-TB, mono-resistant TB, multidrug resistant (MDR)-TB, polydrug resistant TB in G1, G2, G3 and G0 were 19.24%/16.4%/17.33%/19.08%, 11.52%/8.68%/10.94%/11.63%, 3.03%/2.57%/2.96%/3.66% and 4.70%/4.82%/3.34%/ 4.08%, respectively. G3 had a higher risk of MDR1: isoniazid +rifampin (adjusted OR (aOR)=1.91, 95% CI: 1.036 to 3.532), but had a lower risk of DR-TB (aOR=0.84, 95% CI: 0.71 to 0.99), rifampin-related resistance (aOR=0.68, 95% CI: 0.49 to 0.93), streptomycin-related resistance (aOR=0.82, 95% CI: 0.68 to 0.99), ethambutol-related resistance (aOR=0.57, 95% CI: 0.34 to 0.95), MDR3: isoniazid +rifampin+streptomycin (aOR=0.41, 95% CI: 0.19 to 0.85), any isoniazid +streptomycin resistance (aOR=0.85, 95% CI: 0.71 to 1.00). However, there were no significant differences between G1 and G0, G2 and G0 in all drug-resistant subtypes. Those patients with cavity had a higher risk of DR-TB among G3 (OR=1.35, 95% CI: 1.01 to 1.81). CONCLUSION Although we did not found an independent impact of alcohol drinking or tobacco smoking on TB drug-resistance, respectively, these two habits had a combined effect on TB drug-resistance.
Collapse
Affiliation(s)
- Wan-Mei Song
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
- Cheeloo College of Medicine, Shandong University, Jinan, Shandong, China
| | - Shi-Jin Li
- Department of Respiratory Medicine, Chengwu People's Hospital, Heze, Shandong, China
| | - Jin-Yue Liu
- Department of Medical Ultrasound, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Medicine and Health Key Laboratory of Abdominal Medical Imaging, Jinan, Shandong, China
| | - Qi Fu
- State Grid Shandong Electric Power Company, Jinan, Shandong, China
| | - Ting-Ting Xu
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
| | - Ning Ning Tao
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
| | - Qian-Yun Zhang
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
- Cheeloo College of Medicine, Shandong University, Jinan, Shandong, China
| | - Si-Qi Liu
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
- Cheeloo College of Medicine, Shandong University, Jinan, Shandong, China
| | - Qi-Qi An
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
- Cheeloo College of Medicine, Shandong University, Jinan, Shandong, China
| | - Xue-Han Zhu
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
| | - Yao Liu
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
| | - Chun-Bao Yu
- Katharine Hsu International Research Center of Human Infectious Diseases, Shandong Provincial Chest Hospital, Jinan, Shandong, China
| | - Yi-Fan Li
- Department of Respiratory and Critical Care Medicine, The Third Affiliated Hospital of Shandong First Medical University, Jinan, Shandong, China
| | - Jihua Dong
- Department of Respiratory Medicine, Heze Mudan People's Hospital, Heze, Shandong, China
| | - Huai-Chen Li
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital, Jinan, Shandong, China
- Department of Respiratory and Critical Care Medicine, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong, China
- College of Traditional Chinese Medicine, Shandong University of Traditional Chinese Medicine, Jinan, Shandong, China
| |
Collapse
|
5
|
Karki M, Kantipudi K, Yang F, Yu H, Wang YXJ, Yaniv Z, Jaeger S. Generalization Challenges in Drug-Resistant Tuberculosis Detection from Chest X-rays. Diagnostics (Basel) 2022; 12:188. [PMID: 35054355 PMCID: PMC8775073 DOI: 10.3390/diagnostics12010188] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2021] [Revised: 12/23/2021] [Accepted: 01/05/2022] [Indexed: 11/23/2022] Open
Abstract
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.
Collapse
Affiliation(s)
- Manohar Karki
- Lister Hill National Center for Biomedical Communications, U.S. National Library of Medicine, Bethesda, MD 20894, USA; (F.Y.); (H.Y.); (Y.X.J.W.)
| | - Karthik Kantipudi
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, Bethesda, MD 20894, USA;
| | - Feng Yang
- Lister Hill National Center for Biomedical Communications, U.S. National Library of Medicine, Bethesda, MD 20894, USA; (F.Y.); (H.Y.); (Y.X.J.W.)
| | - Hang Yu
- Lister Hill National Center for Biomedical Communications, U.S. National Library of Medicine, Bethesda, MD 20894, USA; (F.Y.); (H.Y.); (Y.X.J.W.)
| | - Yi Xiang J. Wang
- Lister Hill National Center for Biomedical Communications, U.S. National Library of Medicine, Bethesda, MD 20894, USA; (F.Y.); (H.Y.); (Y.X.J.W.)
- Department of Imaging and Interventional Radiology, Faculty of Medicine, The Chinese University of Hong Kong, Prince of Wales Hospital, New Territories, Hong Kong
| | - Ziv Yaniv
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, Bethesda, MD 20894, USA;
| | - Stefan Jaeger
- Lister Hill National Center for Biomedical Communications, U.S. National Library of Medicine, Bethesda, MD 20894, USA; (F.Y.); (H.Y.); (Y.X.J.W.)
| |
Collapse
|
6
|
Yang F, Yu H, Kantipudi K, Karki M, Kassim YM, Rosenthal A, Hurt DE, Yaniv Z, Jaeger S. Differentiating between drug-sensitive and drug-resistant tuberculosis with machine learning for clinical and radiological features. Quant Imaging Med Surg 2022; 12:675-687. [PMID: 34993110 DOI: 10.21037/qims-21-290] [Citation(s) in RCA: 11] [Impact Index Per Article: 5.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/16/2021] [Accepted: 07/23/2021] [Indexed: 12/12/2022]
Abstract
Background Tuberculosis (TB) drug resistance is a worldwide public health problem that threatens progress made in TB care and control. Early detection of drug resistance is important for disease control, with discrimination between drug-resistant TB (DR-TB) and drug-sensitive TB (DS-TB) still being an open problem. The objective of this work is to investigate the relevance of readily available clinical data and data derived from chest X-rays (CXRs) in DR-TB prediction and to investigate the possibility of applying machine learning techniques to selected clinical and radiological features for discrimination between DR-TB and DS-TB. We hypothesize that the number of sextants affected by abnormalities such as nodule, cavity, collapse and infiltrate may serve as a radiological feature for DR-TB identification, and that both clinical and radiological features are important factors for machine classification of DR-TB and DS-TB. Methods We use data from the NIAID TB Portals program (https://tbportals.niaid.nih.gov), 1,455 DR-TB cases and 782 DS-TB cases from 11 countries. We first select three clinical features and 26 radiological features from the dataset. Then, we perform Pearson's chi-squared test to analyze the significance of the selected clinical and radiological features. Finally, we train machine classifiers based on different features and evaluate their ability to differentiate between DR-TB and DS-TB. Results Pearson's chi-squared test shows that two clinical features and 23 radiological features are statistically significant regarding DR-TB vs. DS-TB. A ten-fold cross-validation using a support vector machine shows that automatic discrimination between DR-TB and DS-TB achieves an average accuracy of 72.34% and an average AUC value of 78.42%, when combing all 25 statistically significant features. Conclusions Our study suggests that the number of affected lung sextants can be used for predicting DR-TB, and that automatic discrimination between DR-TB and DS-TB is possible, with a combination of clinical features and radiological features providing the best performance.
Collapse
Affiliation(s)
- Feng Yang
- Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| | - Hang Yu
- Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| | - Karthik Kantipudi
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA
| | - Manohar Karki
- Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| | - Yasmin M Kassim
- Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| | - Alex Rosenthal
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA
| | - Darrell E Hurt
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA
| | - Ziv Yaniv
- Office of Cyber Infrastructure and Computational Biology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA
| | - Stefan Jaeger
- Lister Hill National Center for Biomedical Communications, National Library of Medicine, National Institutes of Health, Bethesda, MD, USA
| |
Collapse
|
7
|
Karki M, Kantipudi K, Yu H, Yang F, Kassim YM, Yaniv Z, Jaeger S. Identifying Drug-Resistant Tuberculosis in Chest Radiographs: Evaluation of CNN Architectures and Training Strategies. ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY. ANNUAL INTERNATIONAL CONFERENCE 2021; 2021:2964-2967. [PMID: 34891867 DOI: 10.1109/embc46164.2021.9630189] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/14/2023]
Abstract
Tuberculosis (TB) is a serious infectious disease that mainly affects the lungs. Drug resistance to the disease makes it more challenging to control. Early diagnosis of drug resistance can help with decision making resulting in appropriate and successful treatment. Chest X-rays (CXRs) have been pivotal to identifying tuberculosis and are widely available. In this work, we utilize CXRs to distinguish between drug-resistant and drug-sensitive tuberculosis. We incorporate Convolutional Neural Network (CNN) based models to discriminate the two types of TB, and employ standard and deep learning based data augmentation methods to improve the classification. Using labeled data from NIAID TB Portals and additional non-labeled sources, we were able to achieve an Area Under the ROC Curve (AUC) of up to 85% using a pretrained InceptionV3 network.
Collapse
|
8
|
Salmanzadeh S, Karamian M, Alavi SM, Nashibi R. Evaluation of the frequency of resistance to 2 drugs (Isoniazid and Rifampin) by molecular investigation and it's risk factors in new cases of smear positive pulmonary tuberculosis in health centers under the cover of Jundishapur University of Medical Sciences in 2017. J Family Med Prim Care 2020; 9:1958-1962. [PMID: 32670947 PMCID: PMC7346914 DOI: 10.4103/jfmpc.jfmpc_983_19] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/06/2019] [Revised: 11/20/2019] [Accepted: 12/16/2019] [Indexed: 11/05/2022] Open
Abstract
Introduction: Despite the great efforts to control tuberculosis (TB), the disease is still one of the major health challenges throughout the world. The basic treatment for TB is drug therapy. Currently, the main anti-tuberculosis drugs with major use in the treatment and control of the disease are isoniazid, rifampin, pyrazinamide, ethambutol, and streptomycin. One of the serious crises in controlling TB epidemic is diagnosis and treatment of patients with Multidrug Resistant Tuberculosis (MDR-TB MDR). The purpose of the study was to examine and evaluate the resistance of mycobacterium TB strains isolated from specimens of newly diagnosed smear positive pulmonary TB to isoniazid and rifampin using molecular methods and their risk factors. Methods: Sputum samples of newly diagnosed smear positive pulmonary TB patients were prepared, collected, and sent to Reference Laboratory in Ahvaz. DNA of mycobacterium tuberculosis was prepared from the samples using Qiagen kit according to the instructions of the manufacturing company. Isoniazid resistance was evaluated using specific primers for inhA and KatG genes. Rifampin resistance was evaluated using MAS-PCR method with three specific alleles of rpobB codons and codons 516, 526 and 531. Results: Mycobacterium tuberculosis resistance to Isoniazid was 7.3%, to Rifampin 5.5% and to both drugs 1.8%. In our study, there were no association between drug resistance and gender, age, prison history, smoking, drug use, underlying disease, occupation, and HIV. Conclusion: According our findings that include prevalence of 7.3% Isoniazide resistance, 5.5% Rifampin resistance and 1.8% to both drugs, evaluating all newly diagnosed patients for resistance to standard anti-tuberculosis treatment seems rational.
Collapse
Affiliation(s)
- Shokrollah Salmanzadeh
- Health Research Institute, Infectious and Tropical Diseases Research Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
| | - Masoumeh Karamian
- Infectious Diseases and Tropical Medicine, Ahvaz Jundishapur University of Medical Sciences, Medical School, Razi Teaching Hospital, Ahvaz, Iran
| | - Seyed Mohammad Alavi
- Health Research Institute, Infectious and Tropical Diseases Research Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
| | - Roohangiz Nashibi
- Health Research Institute, Infectious and Tropical Diseases Research Center, Ahvaz Jundishapur University of Medical Sciences, Ahvaz, Iran
| |
Collapse
|