1
|
Harris M, Qi A, Jeagal L, Torabi N, Menzies D, Korobitsyn A, Pai M, Nathavitharana RR, Ahmad Khan F. A systematic review of the diagnostic accuracy of artificial intelligence-based computer programs to analyze chest x-rays for pulmonary tuberculosis. PLoS One 2019; 14:e0221339. [PMID: 31479448 PMCID: PMC6719854 DOI: 10.1371/journal.pone.0221339] [Citation(s) in RCA: 74] [Impact Index Per Article: 14.8] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/10/2019] [Accepted: 08/05/2019] [Indexed: 12/11/2022] Open
Abstract
We undertook a systematic review of the diagnostic accuracy of artificial intelligence-based software for identification of radiologic abnormalities (computer-aided detection, or CAD) compatible with pulmonary tuberculosis on chest x-rays (CXRs). We searched four databases for articles published between January 2005-February 2019. We summarized data on CAD type, study design, and diagnostic accuracy. We assessed risk of bias with QUADAS-2. We included 53 of the 4712 articles reviewed: 40 focused on CAD design methods (“Development” studies) and 13 focused on evaluation of CAD (“Clinical” studies). Meta-analyses were not performed due to methodological differences. Development studies were more likely to use CXR databases with greater potential for bias as compared to Clinical studies. Areas under the receiver operating characteristic curve (median AUC [IQR]) were significantly higher: in Development studies AUC: 0.88 [0.82–0.90]) versus Clinical studies (0.75 [0.66–0.87]; p-value 0.004); and with deep-learning (0.91 [0.88–0.99]) versus machine-learning (0.82 [0.75–0.89]; p = 0.001). We conclude that CAD programs are promising, but the majority of work thus far has been on development rather than clinical evaluation. We provide concrete suggestions on what study design elements should be improved.
Collapse
Affiliation(s)
- Miriam Harris
- Department of Epidemiology and Biostatistics, McGill University, Montreal, Canada
- Department of Medicine, McGill University Health Centre, Montreal, Canada
- Department of Medicine, Boston University–Boston Medical Center, Boston, Massachusetts, United States of America
- * E-mail:
| | - Amy Qi
- Department of Medicine, McGill University Health Centre, Montreal, Canada
- Respiratory Epidemiology and Clinical Research Unit, Montreal Chest Institute & Research Institute of the McGill University Health Centre, Montreal, Canada
| | - Luke Jeagal
- Respiratory Epidemiology and Clinical Research Unit, Montreal Chest Institute & Research Institute of the McGill University Health Centre, Montreal, Canada
| | - Nazi Torabi
- St. Michael's Hospital, Li Ka Shing International Healthcare Education Centre, Toronto, Canada
| | - Dick Menzies
- Department of Epidemiology and Biostatistics, McGill University, Montreal, Canada
- Respiratory Epidemiology and Clinical Research Unit, Montreal Chest Institute & Research Institute of the McGill University Health Centre, Montreal, Canada
- McGill International TB Centre, Montreal, Canada
| | - Alexei Korobitsyn
- Laboratories, Diagnostics & Drug Resistance Global TB Programme WHO, Geneva, Switzerland
| | - Madhukar Pai
- Department of Epidemiology and Biostatistics, McGill University, Montreal, Canada
- Respiratory Epidemiology and Clinical Research Unit, Montreal Chest Institute & Research Institute of the McGill University Health Centre, Montreal, Canada
- McGill International TB Centre, Montreal, Canada
| | - Ruvandhi R. Nathavitharana
- Division of Infectious Diseases, Beth Israel Deaconess Medical Center, Boston, Massachusetts, United States of America
| | - Faiz Ahmad Khan
- Department of Epidemiology and Biostatistics, McGill University, Montreal, Canada
- Respiratory Epidemiology and Clinical Research Unit, Montreal Chest Institute & Research Institute of the McGill University Health Centre, Montreal, Canada
- McGill International TB Centre, Montreal, Canada
| |
Collapse
|
2
|
de Oliveira M, Duarte SB, Giacomini G, Pereira PCM, de Souza LDR, Miranda JRDA, de Pina DR. A lung image reconstruction from computed radiography images as a tool to tuberculosis treatment control. J Venom Anim Toxins Incl Trop Dis 2019; 25:e144918. [PMID: 31130999 PMCID: PMC6483412 DOI: 10.1590/1678-9199-jvatitd-1449-18] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/11/2018] [Accepted: 10/08/2018] [Indexed: 11/21/2022] Open
Abstract
Background Background: Tuberculosis (TB) is an infectious lung disease with high worldwide incidence that severely compromises the quality of life in affected individuals. Clinical tests are currently employed to monitor pulmonary status and treatment progression. The present study aimed to apply a three-dimensional (3D) reconstruction method based on chest radiography to quantify lung-involvement volume of TB acute-phase patients before and after treatment. In addition, these results were compared with indices from conventional clinical exams to show the coincidence level. Methods A 3D lung reconstruction method using patient chest radiography was applied to quantify lung-involvement volume using retrospective examinations of 50 patients who were diagnosed with pulmonary TB and treated with two different drugs schemes. Twenty-five patients were treated with Scheme I (rifampicin, isoniazid, and pyrazinamide), whereas twenty-five patients were treated with Scheme II (rifampicin, isoniazid, pyrazinamide, and ethambutol). Acute-phase reaction: Serum exams included C-reactive protein levels, erythrocyte sedimentation rate, and albumin levels. Pulmonary function was tested posttreatment. Results We found strong agreement between lung involvement and serum indices pre- and posttreatment. Comparison of the functional severity degree with lung involvement based on 3D image quantification for both treatment schemes found a high correlation. Conclusions The present 3D reconstruction method produced a satisfactory agreement with the acute-phase reaction, most notably a higher significance level with the C-reactive protein. We also found a quite reasonable coincidence between the 3D reconstruction method and the degree of functional lung impairment posttreatment. The performance of the quantification method was satisfactory when comparing the two treatment schemes. Thus, the 3D reconstruction quantification method may be useful tools for monitoring TB treatment. The association with serum indices are not only inexpensive and sensitive but also may be incorporated into the assessment of patients during TB treatment.
Collapse
Affiliation(s)
- Marcela de Oliveira
- Universidade Estadual Paulista, Instituto de Biociências de Botucatu, Departamento de Física e Biofísica, Botucatu, SP, Brasil
| | - Sergio Barbosa Duarte
- Centro Brasileiro de Pesquisas Físicas, Departamento de Altas Energias, Rio de Janeiro, RJ, Brasil
| | - Guilherme Giacomini
- Universidade Estadual Paulista, Instituto de Biociências de Botucatu, Departamento de Física e Biofísica, Botucatu, SP, Brasil
| | - Paulo Câmara Marques Pereira
- Universidade Estadual Paulista, Faculdade de Medicina de Botucatu, Departamento de Doenças Tropicais e Diagnóstico por Imagem, Botucatu, SP, Brasil
| | - Lenice do Rosário de Souza
- Universidade Estadual Paulista, Faculdade de Medicina de Botucatu, Departamento de Doenças Tropicais e Diagnóstico por Imagem, Botucatu, SP, Brasil
| | - José Ricardo de Arruda Miranda
- Universidade Estadual Paulista, Instituto de Biociências de Botucatu, Departamento de Física e Biofísica, Botucatu, SP, Brasil
| | - Diana Rodrigues de Pina
- Universidade Estadual Paulista, Faculdade de Medicina de Botucatu, Departamento de Doenças Tropicais e Diagnóstico por Imagem, Botucatu, SP, Brasil
| |
Collapse
|
3
|
Wen H, Li P, Ma H, Lv G. Diagnostic accuracy of Xpert MTB/RIF assay for musculoskeletal tuberculosis: a meta-analysis. Infect Drug Resist 2017; 10:299-305. [PMID: 29026323 PMCID: PMC5627758 DOI: 10.2147/idr.s145843] [Citation(s) in RCA: 13] [Impact Index Per Article: 1.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/18/2022] Open
Abstract
OBJECTIVES Xpert MTB/RIF assay, a rapid and automated real-time nucleic acid amplification test, has been reported for the diagnosis of musculoskeletal tuberculosis (TB) in current years. This meta-analysis aims to determine the diagnostic accuracy of Xpert for the detection of musculoskeletal TB and rifampicin (RIF) resistance. METHODS We searched PubMed, Embase, China National Knowledge Infrastructure, and Wanfang for original articles published up to 1st June 2017 to identify studies in which the Xpert assay was applied to diagnose musculoskeletal TB. Pooled estimates were calculated using a random-effects model or a fixed-effects model according to heterogeneity. Summary receiver operating characteristic curves and the area under the curve (AUC) were used to summarize overall diagnostic performance. Deeks' test was performed to evaluate potential publication bias. RESULTS Twelve studies were identified with a pooled sensitivity and specificity of respectively 0.81 (95% confidence interval [CI] 0.78-0.83) and 0.83 (95% CI 0.80-0.86) of Xpert for the diagnosis of musculoskeletal TB. Xpert was highly sensitive (0.89, 95% CI 0.79-0.95) and highly specific (0.96, 95% CI 0.92-0.98) in detecting RIF resistance. AUC (over 0.9) suggested a relatively high level of overall diagnostic accuracy of Xpert for detecting musculoskeletal TB and RIF resistance. Prevalence and reference standard were indicated to be sources of heterogeneity between studies. No publication bias was found. CONCLUSION This study provides available evidence of the rapid and effective role of Xpert in diagnosing musculoskeletal TB and detecting RIF resistance.
Collapse
Affiliation(s)
- Hai Wen
- Department of Spine Surgery, The Second Xiangya Hospital of Central South University, Changsha, People’s Republic of China
| | - Pengzhi Li
- Department of Spine Surgery, The Second Xiangya Hospital of Central South University, Changsha, People’s Republic of China
| | - Hong Ma
- Department of Spine Surgery, The Second Xiangya Hospital of Central South University, Changsha, People’s Republic of China
| | - Guohua Lv
- Department of Spine Surgery, The Second Xiangya Hospital of Central South University, Changsha, People’s Republic of China
| |
Collapse
|
4
|
de Pina DR, Alvarez M, Giacomini G, Pavan ALM, Guedes CIA, Cavalcante RDS, Mendes RP, Paniago AMM. Paracoccidioidomycosis: level of pulmonary sequelae in high resolution computed tomography images from patients of two endemic regions of Brazil. Quant Imaging Med Surg 2017; 7:318-325. [PMID: 28811998 DOI: 10.21037/qims.2017.06.04] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
Abstract
BACKGROUND The purpose of the present study was to evaluate differences in lung damage in patients with paracoccidioidomycosis (PCM) in two endemic regions of Brazil (Botucatu, SP, in the southeastern region and Campo Grande, MS, in the west central region). METHODS The study sought to objectively quantify fibrosis and emphysema treated patients using high-resolution computed tomography (HRCT) that was performed according to an established and validated computational method. The analysis was based on clinical data that were associated with objective quantifications of pulmonary sequelae. We performed a retrospective analysis of HRCT exams from 32 successfully treated patients with the chronic form of PCM. The two groups had similar characteristics with regard to age, symptom duration, smoking history, and titers on a diagnostic serologic test. RESULTS The statistical analysis revealed more severe cases and a higher percentage of emphysema in the Campo Grande group. The mean percentages of emphysema were 40.2% and 13.6% in the Campo Grande and Botucatu groups, respectively. The percentage of fibrosis was significantly higher in the Botucatu group (11.3%) than in the Campo Grande group (2.3%). CONCLUSIONS The present results may reflect differences in cryptic species of P. brasiliensis in these two geographic regions of Brazil. Further studies should be done to clarify the differences observed herein. Such findings may help unveil differences among cryptic species in terms of the pulmonary consequences that are caused by this disease, which would be highly beneficial to PCM patients.
Collapse
Affiliation(s)
- Diana Rodrigues de Pina
- Department of Tropical Diseases and Diagnostic Imaging, Botucatu Medical School, São Paulo State University, Botucatu, São Paulo, Brazil
| | - Matheus Alvarez
- Department of Physics and Biophysics, Biosciences Institute of Botucatu, São Paulo State University, Botucatu, São Paulo, Brazil
| | - Guilherme Giacomini
- Department of Physics and Biophysics, Biosciences Institute of Botucatu, São Paulo State University, Botucatu, São Paulo, Brazil
| | - Ana Luiza Menegatti Pavan
- Department of Physics and Biophysics, Biosciences Institute of Botucatu, São Paulo State University, Botucatu, São Paulo, Brazil
| | - Carlos Ivan Andrade Guedes
- Department of Internal Medicine, Federal University of Mato Grosso do Sul (UFMS), Campo Grande, Mato Grosso do Sul, Brazil
| | - Ricardo de Souza Cavalcante
- Department of Tropical Diseases and Diagnostic Imaging, Botucatu Medical School, São Paulo State University, Botucatu, São Paulo, Brazil
| | - Rinaldo Poncio Mendes
- Department of Tropical Diseases and Diagnostic Imaging, Botucatu Medical School, São Paulo State University, Botucatu, São Paulo, Brazil
| | - Anamaria Mello Miranda Paniago
- Department of Internal Medicine, Federal University of Mato Grosso do Sul (UFMS), Campo Grande, Mato Grosso do Sul, Brazil
| |
Collapse
|
5
|
Oshiro H, Czerniak BA, Sakamaki K, Tsuta K, Bondaruk J, Keyhani A, Dinney CP, Nagai T, Kamat AM. Comparison between whole mount tissue preparations and virtual tissue microarray samples for measuring Ki-67 and apoptosis indices in human bladder cancer: A cross-sectional study. Medicine (Baltimore) 2016; 95:e4500. [PMID: 27495099 PMCID: PMC4979853 DOI: 10.1097/md.0000000000004500] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/25/2022] Open
Abstract
Recent tissue microarray (TMA)-based studies have shown that cell proliferation- and apoptosis-related biomarkers are associated with clinical outcomes in patients with bladder urothelial carcinoma. However, little is known about the differences in these biomarker measurements between whole mount tissue preparations and TMAs. This study aimed to elucidate the discrepancy in the measurements of Ki-67 indices (KIs) and apoptosis indices (AIs) between whole mount tissue preparations and TMAs of bladder urothelial carcinoma samples.Whole mount tissue preparations for Ki-67 immunohistochemistry and terminal deoxynucleotidyl transferase dUTP nick end labeling were made from 30 patients who underwent transurethral resection of bladder urothelial carcinoma. Digital microscopy-assisted virtual TMAs, consisting of 3 small round areas (1 or 0.6 mm in diameter), were generated from the same whole mount tissue preparations. The measurement results in highly reactive areas of biomarkers were compared between the whole mount tissue preparation- and the TMA-based methods. Bland-Altman plot analysis, regression analysis, and Kendall τ were performed to investigate differences in the measurement results, systematic biases, and correlations between biomarkers.Although the Bland-Altman plot analysis demonstrated that almost all the plots were within the limits of agreement, fixed biases were detected in the 1- and 0.6-mm TMAs for the KI (0.181 and 0.222, respectively) and the AI (0.055 and 0.063, respectively). Proportional biases were also detected in the 1- and 0.6-mm TMAs for the AI (P < 0.001 and P < 0.001, respectively). Furthermore, positive correlations between KIs and AIs were observed in whole mount tissue preparations (r = 0.260, P = 0.044) and in the 1 mm TMAs (r = 0.375, P = 0.004); however, no such correlation was observed in the 0.6 mm TMAs.Our study suggests that the measurement results for certain biomarkers of bladder urothelial carcinoma obtained from TMA-based samples can be susceptible to systematic bias, and the lack of correlation between biomarkers cannot be avoided as it is in whole mount tissue preparations. Virtual TMAs can help identify systematic bias and establish a better sampling strategy prior to performing high-throughput TMAs for biomarker studies.
Collapse
Affiliation(s)
- Hisashi Oshiro
- Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
- Department of Anatomic Pathology, Tokyo Medical University, Tokyo
- Department of Pathology, Jichi Medical University, Shimotsuke, Tochigi, Japan
| | - Bogdan A. Czerniak
- Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
| | - Kentaro Sakamaki
- Department of Biostatistics and Epidemiology, Yokohama City University Graduate School of Medicine, Yokohama, Kanagawa, Japan
| | - Koji Tsuta
- Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
- Department of Clinical Sciences and Laboratory Medicine, Kansai Medical University, Hirakata, Osaka
| | - Jolanta Bondaruk
- Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
| | - Afsaneh Keyhani
- Department of Urology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
| | - Colin P. Dinney
- Department of Urology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
| | - Takeshi Nagai
- Department of Anatomic Pathology, Tokyo Medical University, Tokyo
| | - Ashish M. Kamat
- Department of Urology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
- Correspondence: Ashish M. Kamat, Department of Urology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA (e-mail: )
| |
Collapse
|
6
|
Pang C, Wu Y, Wan C, Shen K, Hu Y, Yang T, Shen Y, Wen F. Accuracy of the Bronchoalveolar Lavage Enzyme-Linked Immunospot Assay for the Diagnosis of Pulmonary Tuberculosis: A Meta-analysis. Medicine (Baltimore) 2016; 95:e3183. [PMID: 27015211 PMCID: PMC4998406 DOI: 10.1097/md.0000000000003183] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/05/2023] Open
Abstract
Assessing of local immune response may improve the accuracy of pulmonary tuberculosis (PTB) diagnosis. Many studies have investigated diagnosing PTB based on enzyme-linked immunospot (ELISPOT) assay of bronchoalveolar lavage (BAL) fluid, but the results have been inconclusive. We meta-analyzed the available evidences on overall diagnostic performance of ELISPOT assay of BAL fluid for diagnosing PTB.A systematic literature search was performed using PubMed, Embase, Wangfang, Weipu, and CNKI. Data were pooled on sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR). Overall test performance was summarized using summary receiver operating characteristic curves and the area under the curve (AUC). Deeks test was used to test for potential publication bias.Seven publications with 814 subjects met our inclusion criteria and were included in this meta-analysis. The following pooled estimates for diagnostic parameters were obtained: sensitivity, 0.90 (95% CI: 0.85-0.94); specificity, 0.80 (95% CI: 0.77-0.84); PLR, 5.08 (95% CI: 2.70-9.57); NLR, 0.13 (95% CI: 0.06-0.28); DOR, 49.12 (95% CI: 12.97-186.00); and AUC, 0.96. No publication bias was identified.The available evidence suggests that ELISPOT assay of BAL fluid is a useful rapid diagnostic test for PTB. The results of this assay should be interpreted in parallel with clinical findings and the results of conventional tests.
Collapse
Affiliation(s)
- Caishuang Pang
- From the Department of Respiratory and Critical Care Medicine, West China Hospital of Sichuan University and Division of Pulmonary Diseases, State Key Laboratory of Biotherapy of China, Chengdu (CP, YW, CW, TY, YS, FW); Chongqing Cancer Hospital, Chongqing Cancer Institute, Chongqing (CP); Radiation Physics Center, Cancer Center and State Key Laboratory of Biotherapy, West China Hospital of Sichuan University (KS); and West China School of Medicine, Sichuan University, Chengdu, China (YH)
| | | | | | | | | | | | | | | |
Collapse
|