Development and Validation of a Deep Learning System for Detection of Active Pulmonary Tuberculosis on Chest Radiographs: Clinical and Technical Considerations

2018 ◽  
Vol 69 (5) ◽  
pp. 748-750 ◽  
Author(s):  
Daniel Shu Wei Ting ◽  
Tien-En Tan ◽  
C C Tchoyoson Lim
Keyword(s):  
Deep Learning ◽  
Pulmonary Tuberculosis ◽  
Chest Radiographs ◽  
Learning System ◽  
Active Pulmonary Tuberculosis ◽  
Development And Validation ◽  
Technical Considerations

scholarly journals Development and Validation of a Deep Learning–based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs

2018 ◽  
Vol 69 (5) ◽  
pp. 739-747 ◽  
Author(s):  
Eui Jin Hwang ◽  
Sunggyun Park ◽  
Kwang-Nam Jin ◽  
Jung Im Kim ◽  
So Young Choi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Pulmonary Tuberculosis ◽  
Performance Test ◽  
Automatic Detection ◽  
High Sensitivity ◽  
High Specificity ◽  
Roc Curves ◽  
Chest Radiographs ◽  
Automated System ◽  
Active Pulmonary Tuberculosis

Abstract Background Detection of active pulmonary tuberculosis on chest radiographs (CRs) is critical for the diagnosis and screening of tuberculosis. An automated system may help streamline the tuberculosis screening process and improve diagnostic performance. Methods We developed a deep learning–based automatic detection (DLAD) algorithm using 54c221 normal CRs and 6768 CRs with active pulmonary tuberculosis that were labeled and annotated by 13 board-certified radiologists. The performance of DLAD was validated using 6 external multicenter, multinational datasets. To compare the performances of DLAD with physicians, an observer performance test was conducted by 15 physicians including nonradiology physicians, board-certified radiologists, and thoracic radiologists. Image-wise classification and lesion-wise localization performances were measured using area under the receiver operating characteristic (ROC) curves and area under the alternative free-response ROC curves, respectively. Sensitivities and specificities of DLAD were calculated using 2 cutoffs (high sensitivity [98%] and high specificity [98%]) obtained through in-house validation. Results DLAD demonstrated classification performance of 0.977–1.000 and localization performance of 0.973–1.000. Sensitivities and specificities for classification were 94.3%–100% and 91.1%–100% using the high-sensitivity cutoff and 84.1%–99.0% and 99.1%–100% using the high-specificity cutoff. DLAD showed significantly higher performance in both classification (0.993 vs 0.746–0.971) and localization (0.993 vs 0.664–0.925) compared to all groups of physicians. Conclusions Our DLAD demonstrated excellent and consistent performance in the detection of active pulmonary tuberculosis on CR, outperforming physicians, including thoracic radiologists.

scholarly journals The Accuracy of Emergency Physicians’ Suspicions of Active Pulmonary Tuberculosis

2021 ◽  
Vol 10 (4) ◽  
pp. 860
Author(s):  
Shiang-Jin Chen ◽  
Chun-Yu Lin ◽  
Tzu-Ling Huang ◽  
Ying-Chi Hsu ◽  
Kuan-Ting Liu
Keyword(s):  
Chronic Kidney Disease ◽  
Weight Loss ◽  
Kidney Disease ◽  
Pulmonary Tuberculosis ◽  
General Ward ◽  
Chest Radiographs ◽  
Emergency Physicians ◽  
Active Pulmonary Tuberculosis ◽  
Factors Associated ◽  
History Of

Objective: To investigate factors associated with recognition and delayed isolation of pulmonary tuberculosis (PTB). Background: Precise identification of PTB in the emergency department (ED) remains challenging. Methods: Retrospectively reviewed PTB suspects admitted via the ED were divided into three groups based on the acid-fast bacilli culture report and whether they were isolated initially in the ED or general ward. Factors related to recognition and delayed isolation were statistically compared. Results: Only 24.94% (100/401) of PTB suspects were truly active PTB and 33.77% (51/151) of active PTB were unrecognized in the ED. Weight loss (p = 0.022), absence of dyspnea (p = 0.021), and left upper lobe field (p = 0.024) lesions on chest radiographs were related to truly active PTB. Malignancy (p = 0.015), chronic kidney disease (p = 0.047), absence of a history of PTB (p = 0.013), and lack of right upper lung (p ≤ 0.001) and left upper lung (p = 0.020) lesions were associated with PTB being missed in the ED. Conclusions: Weight loss, absence of dyspnea, and left upper lobe field lesions on chest radiographs were related to truly active PTB. Malignancy, chronic kidney disease, absence of a history of PTB, and absence of right and/or left upper lung lesions on chest radiography were associated with isolation delay.

scholarly journals Development and Validation of a Deep Learning–Based Automated Detection Algorithm for Major Thoracic Diseases on Chest Radiographs

2019 ◽  
Vol 2 (3) ◽  
pp. e191095 ◽  
Author(s):  
Eui Jin Hwang ◽  
Sunggyun Park ◽  
Kwang-Nam Jin ◽  
Jung Im Kim ◽  
So Young Choi ◽  
...  
Keyword(s):  
Deep Learning ◽  
Detection Algorithm ◽  
Automated Detection ◽  
Chest Radiographs ◽  
Thoracic Diseases ◽  
Development And Validation

scholarly journals Development and validation of a deep learning algorithm detecting 10 common abnormalities on chest radiographs

2020 ◽  
pp. 2003061
Author(s):  
Ju Gang Nam ◽  
Minchul Kim ◽  
Jongchan Park ◽  
Eui Jin Hwang ◽  
Jong Hyuk Lee ◽  
...  
Keyword(s):  
Emergency Department ◽  
Deep Learning ◽  
Learning Algorithm ◽  
External Validation ◽  
Chest Radiographs ◽  
Reading Tests ◽  
Deep Learning Algorithm ◽  
The Mean ◽  
Emergency Department Patients ◽  
Development And Validation

We aimed to develop a deep-learning algorithm detecting 10 common abnormalities (DLAD-10) on chest radiographs and to evaluate its impact in diagnostic accuracy, timeliness of reporting, and workflow efficacy.DLAD-10 was trained with 146 717 radiographs from 108 053 patients using a ResNet34-based neural network with lesion-specific channels for 10 common radiologic abnormalities (pneumothorax, mediastinal widening, pneumoperitoneum, nodule/mass, consolidation, pleural effusion, linear atelectasis, fibrosis, calcification, and cardiomegaly). For external validation, the performance of DLAD-10 on a same-day CT-confirmed dataset (normal:abnormal, 53:147) and an open-source dataset (PadChest; normal:abnormal, 339:334) was compared to that of three radiologists. Separate simulated reading tests were conducted on another dataset adjusted to real-world disease prevalence in the emergency department, consisting of four critical, 52 urgent, and 146 non-urgent cases. Six radiologists participated in the simulated reading sessions with and without DLAD-10.DLAD-10 exhibited areas under the receiver-operating characteristic curves (AUROCs) of 0.895–1.00 in the CT-confirmed dataset and 0.913–0.997 in the PadChest dataset. DLAD-10 correctly classified significantly more critical abnormalities (95.0% [57/60]) than pooled radiologists (84.4% [152/180]; p=0.01). In simulated reading tests for emergency department patients, pooled readers detected significantly more critical (70.8% [17/24] versus 29.2% [7/24]; p=0.006) and urgent (82.7% [258/312] versus 78.2% [244/312]; p=0.04) abnormalities when aided by DLAD-10. DLAD-10 assistance shortened the mean time-to-report critical and urgent radiographs (640.5±466.3 versus 3371.0±1352.5 s and 1840.3±1141.1 versus 2127.1±1468.2, respectively; p-values<0.01) and reduced the mean interpretation time (20.5±22.8 versus 23.5±23.7 s; p<0.001).DLAD-10 showed excellent performance, improving radiologists' performance and shortening the reporting time for critical and urgent cases.

Sign in / Sign up

Export Citation Format

Share Document