Deep Learning Detection of Active Pulmonary Tuberculosis at Chest Radiography Matched the Clinical Performance of Radiologists

Kazemzadeh, Sahar; Jin, Yu; Jamshy, Shahar; Pilgrim, Rory; Nabulsi, Zaid; Chen, Christina; Beladia, Neeral; Lau, Charles T.; McKinney, Scott Mayer; Hughes, T. A.; Kiraly, Atilla P.; Kalidindi, Sreenivasa Raju; Muyoyeta, Monde; Malemela, Jameson; Shih, Ting Yu; Corrado, Greg S.; Peng, Lily; Chou, Katherine; Chen, Po-Hsuan Cameron; Liu, Yun; Eswaran, Krishnan; Tse, Daniel; Shetty, Shravya; Prabhakara, Shruthi

doi:10.1148/radiol.212213

Cited by 34 publications

(18 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 25 An accurate CAD algorithm used as a triage test prior to molecular TB testing can provide 40–80% cost savings at a TB prevalence of 1–10% among the screened population. 26 In a randomised trial among adults presenting with cough in Malawi, incorporating HIV testing in addition to CAD-based CXR analysis to triage for Xpert testing reduced time to TB treatment and untreated or undiagnosed HIV. 27 However, no economic benefit was found.…”

Section: Economic Aspectsmentioning

confidence: 99%

The rise of artificial intelligence reading of chest X-rays for enhanced TB diagnosis and elimination

Geric

Qin

Denkinger

et al. 2023

int j tuberc lung dis

View full text Add to dashboard Cite

We provide an overview of the latest evidence on computer-aided detection (CAD) software for automated interpretation of chest radiographs (CXRs) for TB detection. CAD is a useful tool that can assist in rapid and consistent CXR interpretation for TB. CAD can achieve high sensitivity TB detection among people seeking care with symptoms of TB and in population-based screening, has accuracy on-par with human readers. However, implementation challenges remain. Due to diagnostic heterogeneity between settings and sub-populations, users need to select threshold scores rather than use pre-specified ones, but some sites may lack the resources and data to do so. Efficient standardisation is further complicated by frequent updates and new CAD versions, which also challenges implementation and comparison. CAD has not been validated for TB diagnosis in children and its accuracy for identifying non-TB abnormalities remains to be evaluated. A number of economic and political issues also remain to be addressed through regulation for CAD to avoid furthering health inequities. Although CAD-based CXR analysis has proven remarkably accurate for TB detection in adults, the above issues need to be addressed to ensure that the technology meets the needs of high-burden settings and vulnerable sub-populations.

show abstract

Section: Economic Aspectsmentioning

confidence: 99%

The rise of artificial intelligence reading of chest X-rays for enhanced TB diagnosis and elimination

Geric

Qin

Denkinger

et al. 2023

int j tuberc lung dis

View full text Add to dashboard Cite

show abstract

“…Deep learning, a major branch of AI, 36 is experiencing an era of explosive growth, constituting a breakthrough in medical image classification tasks by mining the association between raw input visual data and desired output, generating decisions ranging from macroscopic disease diagnosis and outcome prediction to microscopic gene mutation status 37–49 . Along with the vigorous evolution of AI techniques, medical image‐based computational approaches have been proposed for TB, showing comparable performance to radiologists in disease assessment 15–20 . And efficiency of software for automated TB diagnosis has been validated, offering novel insights for TB management 50,51 .…”

Section: Discussionmentioning

confidence: 99%

“…14 Currently, image-based artificial intelligence (AI) systems have been proposed for the detection and activity assessment of TB, achieving, or even surpassing the performance of human physicians. [15][16][17][18][19][20] For automated diagnosis of DR-TB, chest X-ray (CXR), a two-dimensional projection consisting of overlapped anatomical structures, was commonly utilized, whereas CT imaging, a three-dimensional (3D) reconstruction presenting higher diagnostic reliability, has been seldom reported. 21 Moreover, moderate performance and small-scale datasets hinder the practicality and generality of those models.…”

Section: Introductionmentioning

confidence: 99%

Deep learning for precise diagnosis and subtype triage of drug‐resistant tuberculosis on chest computed tomography

Liang,

Xu,

Yang

et al. 2024

MedComm

View full text Add to dashboard Cite

Deep learning, transforming input data into target prediction through intricate network structures, has inspired novel exploration in automated diagnosis based on medical images. The distinct morphological characteristics of chest abnormalities between drug‐resistant tuberculosis (DR‐TB) and drug‐sensitive tuberculosis (DS‐TB) on chest computed tomography (CT) are of potential value in differential diagnosis, which is challenging in the clinic. Hence, based on 1176 chest CT volumes from the equal number of patients with tuberculosis (TB), we presented a Deep learning‐based system for TB drug resistance identification and subtype classification (DeepTB), which could automatically diagnose DR‐TB and classify crucial subtypes, including rifampicin‐resistant tuberculosis, multidrug‐resistant tuberculosis, and extensively drug‐resistant tuberculosis. Moreover, chest lesions were manually annotated to endow the model with robust power to assist radiologists in image interpretation and the Circos revealed the relationship between chest abnormalities and specific types of DR‐TB. Finally, DeepTB achieved an area under the curve (AUC) up to 0.930 for thoracic abnormality detection and 0.943 for DR‐TB diagnosis. Notably, the system demonstrated instructive value in DR‐TB subtype classification with AUCs ranging from 0.880 to 0.928. Meanwhile, class activation maps were generated to express a human‐understandable visual concept. Together, showing a prominent performance, DeepTB would be impactful in clinical decision‐making for DR‐TB.

show abstract

“…10,11 While AI prediction of NTM-LD nodule activity had not been previously studied, several authors reported the ability of DCNN models to determine the activity of pulmonary TB using CXRs. [15][16][17] One of these studies found AI performance (AUC, 0.89) to be comparable and noninferior to practicing radiologists. 17 Our DCNN model demonstrated good performance in identifying acute cases of NTM-LD that require prompt and appropriate management.…”

Section: Discussionmentioning

confidence: 99%

“…[15][16][17] One of these studies found AI performance (AUC, 0.89) to be comparable and noninferior to practicing radiologists. 17 Our DCNN model demonstrated good performance in identifying acute cases of NTM-LD that require prompt and appropriate management. It is important to emphasize that the model's predictions should be considered as just one piece of radiologic evidence, among others.…”

Section: Discussionmentioning

confidence: 99%

Utilizing Deep Learning and Computed Tomography to Determine Pulmonary Nodule Activity in Patients With Nontuberculous Mycobacterial-Lung Disease

Lancaster,

Cardin,

Nguyen

et al. 2023

Journal of Thoracic Imaging

View full text Add to dashboard Cite

Purpose: To develop and evaluate a deep convolutional neural network (DCNN) model for the classification of acute and chronic lung nodules from nontuberculous mycobacterial-lung disease (NTM-LD) on computed tomography (CT). Materials and Methods: We collected a data set of 650 nodules (316 acute and 334 chronic) from the CT scans of 110 patients with NTM-LD. The data set was divided into training, validation, and test sets in a ratio of 4:1:1. Bounding boxes were used to crop the 2D CT images down to the area of interest. A DCNN model was built using 11 convolutional layers and trained on these images. The performance of the model was evaluated on the hold-out test set and compared with that of 3 radiologists who independently reviewed the images. Results: The DCNN model achieved an area under the receiver operating characteristic curve of 0.806 for differentiating acute and chronic NTM-LD nodules, corresponding to sensitivity, specificity, and accuracy of 76%, 68%, and 72%, respectively. The performance of the model was comparable to that of the 3 radiologists, who had area under the receiver operating characteristic curve, sensitivity, specificity, and accuracy of 0.693 to 0.771, 61% to 82%, 59% to 73%, and 60% to 73%, respectively. Conclusions: This study demonstrated the feasibility of using a DCNN model for the classification of the activity of NTM-LD nodules on chest CT. The model performance was comparable to that of radiologists. This approach can potentially and efficiently improve the diagnosis and management of NTM-LD.

show abstract

Deep Learning Detection of Active Pulmonary Tuberculosis at Chest Radiography Matched the Clinical Performance of Radiologists

Cited by 34 publications

References 26 publications

The rise of artificial intelligence reading of chest X-rays for enhanced TB diagnosis and elimination

The rise of artificial intelligence reading of chest X-rays for enhanced TB diagnosis and elimination

Deep learning for precise diagnosis and subtype triage of drug‐resistant tuberculosis on chest computed tomography

Utilizing Deep Learning and Computed Tomography to Determine Pulmonary Nodule Activity in Patients With Nontuberculous Mycobacterial-Lung Disease

Contact Info

Product

Resources

About