• Title/Summary/Keyword: Chest Radiography(CXR)

Search Result 10, Processing Time 0.024 seconds

Implementation of a Deep Learning-Based Computer-Aided Detection System for the Interpretation of Chest Radiographs in Patients Suspected for COVID-19

  • Eui Jin Hwang;Hyungjin Kim;Soon Ho Yoon;Jin Mo Goo;Chang Min Park
    • Korean Journal of Radiology
    • /
    • v.21 no.10
    • /
    • pp.1150-1160
    • /
    • 2020
  • Objective: To describe the experience of implementing a deep learning-based computer-aided detection (CAD) system for the interpretation of chest X-ray radiographs (CXR) of suspected coronavirus disease (COVID-19) patients and investigate the diagnostic performance of CXR interpretation with CAD assistance. Materials and Methods: In this single-center retrospective study, initial CXR of patients with suspected or confirmed COVID-19 were investigated. A commercialized deep learning-based CAD system that can identify various abnormalities on CXR was implemented for the interpretation of CXR in daily practice. The diagnostic performance of radiologists with CAD assistance were evaluated based on two different reference standards: 1) real-time reverse transcriptase-polymerase chain reaction (rRT-PCR) results for COVID-19 and 2) pulmonary abnormality suggesting pneumonia on chest CT. The turnaround times (TATs) of radiology reports for CXR and rRT-PCR results were also evaluated. Results: Among 332 patients (male:female, 173:159; mean age, 57 years) with available rRT-PCR results, 16 patients (4.8%) were diagnosed with COVID-19. Using CXR, radiologists with CAD assistance identified rRT-PCR positive COVID-19 patients with sensitivity and specificity of 68.8% and 66.7%, respectively. Among 119 patients (male:female, 75:44; mean age, 69 years) with available chest CTs, radiologists assisted by CAD reported pneumonia on CXR with a sensitivity of 81.5% and a specificity of 72.3%. The TATs of CXR reports were significantly shorter than those of rRT-PCR results (median 51 vs. 507 minutes; p < 0.001). Conclusion: Radiologists with CAD assistance could identify patients with rRT-PCR-positive COVID-19 or pneumonia on CXR with a reasonably acceptable performance. In patients suspected with COVID-19, CXR had much faster TATs than rRT-PCRs.

A Clinical Study on a 5 Decades Tuberculosis Screening Program Based on Chest Radiography(CXR) (흉부방사선영상(CXR)에 의한 폐결핵검진사업 50년의 임상적 고찰)

  • Kim, Ham-Gyum
    • Journal of radiological science and technology
    • /
    • v.32 no.2
    • /
    • pp.141-146
    • /
    • 2009
  • This study analyzed decade-based statistic data which had been collected from the reports of annual radiographic pulmonary tuberculosis screening program initiated by the Korean National Tuberculosis Association (KNTA) for last 5 decades (from 1956 to 2005). We analyzed only the content of annual statistic report to preserve the characteristic of statistic data and the contents of original copy by focusing on the analysis of tuberculosis cases where age and sex were excluded. The results of the disease-based analysis on the tuberculosis cases from cumulative subjects of chest radiography (CXR) from 1956 to 2005 are summarized as follows. 1. The cumulative number of subjects who were examined under annual chest radiography over last 5 decades totaled 54,938,875 persons. 2. The cumulative number of pulmonary tuberculosis cases during same period totaled 958,251 persons (1.74%). 3. The cumulative number of subjects treated during same period totaled 465,082 persons (0.85%). 4. The cumulative number of mild pulmonary tuberculosis cases during same period totaled 229,615 persons (0.42%). 5. The cumulative number of moderate pulmonary tuberculosis cases during same period totaled 144,247 persons (0.26%). 6. The cumulative number of severe pulmonary tuberculosis cases during same period totaled 74,066 persons (0.13%). 7. The cumulative number of exudative pleurisy cases during same period totaled 17,154 persons (0.03%). 8. The cumulative number of subjects under monitoring during same period totaled 493,169 persons (0.90%). 9. The cumulative number of uncertain activity cases during same period totaled 78,214 persons (0.14%). 10. The cumulative number of pseudo-pulmonary tuberculosis cases during same period totaled 272,349 persons (0.50%).

  • PDF

Bone Suppression on Chest Radiographs for Pulmonary Nodule Detection: Comparison between a Generative Adversarial Network and Dual-Energy Subtraction

  • Kyungsoo Bae;Dong Yul Oh;Il Dong Yun;Kyung Nyeo Jeon
    • Korean Journal of Radiology
    • /
    • v.23 no.1
    • /
    • pp.139-149
    • /
    • 2022
  • Objective: To compare the effects of bone suppression imaging using deep learning (BSp-DL) based on a generative adversarial network (GAN) and bone subtraction imaging using a dual energy technique (BSt-DE) on radiologists' performance for pulmonary nodule detection on chest radiographs (CXRs). Materials and Methods: A total of 111 adults, including 49 patients with 83 pulmonary nodules, who underwent both CXR using the dual energy technique and chest CT, were enrolled. Using CT as a reference, two independent radiologists evaluated CXR images for the presence or absence of pulmonary nodules in three reading sessions (standard CXR, BSt-DE CXR, and BSp-DL CXR). Person-wise and nodule-wise performances were assessed using receiver-operating characteristic (ROC) and alternative free-response ROC (AFROC) curve analyses, respectively. Subgroup analyses based on nodule size, location, and the presence of overlapping bones were performed. Results: BSt-DE with an area under the AFROC curve (AUAFROC) of 0.996 and 0.976 for readers 1 and 2, respectively, and BSp-DL with AUAFROC of 0.981 and 0.958, respectively, showed better nodule-wise performance than standard CXR (AUAFROC of 0.907 and 0.808, respectively; p ≤ 0.005). In the person-wise analysis, BSp-DL with an area under the ROC curve (AUROC) of 0.984 and 0.931 for readers 1 and 2, respectively, showed better performance than standard CXR (AUROC of 0.915 and 0.798, respectively; p ≤ 0.011) and comparable performance to BSt-DE (AUROC of 0.988 and 0.974; p ≥ 0.064). BSt-DE and BSp-DL were superior to standard CXR for detecting nodules overlapping with bones (p < 0.017) or in the upper/middle lung zone (p < 0.017). BSt-DE was superior (p < 0.017) to BSp-DL in detecting peripheral and sub-centimeter nodules. Conclusion: BSp-DL (GAN-based bone suppression) showed comparable performance to BSt-DE and can improve radiologists' performance in detecting pulmonary nodules on CXRs. Nevertheless, for better delineation of small and peripheral nodules, further technical improvements are required.

Evaluation of Deep-Learning Feature Based COVID-19 Classifier in Various Neural Network (코로나바이러스 감염증19 데이터베이스에 기반을 둔 인공신경망 모델의 특성 평가)

  • Hong, Jun-Yong;Jung, Young-Jin
    • Journal of radiological science and technology
    • /
    • v.43 no.5
    • /
    • pp.397-404
    • /
    • 2020
  • Coronavirus disease(COVID-19) is highly infectious disease that directly affects the lungs. To observe the clinical findings from these lungs, the Chest Radiography(CXR) can be used in a fast manner. However, the diagnostic performance via CXR needs to be improved, since the identifying these findings are highly time-consuming and prone to human error. Therefore, Artificial Intelligence(AI) based tool may be useful to aid the diagnosis of COVID-19 via CXR. In this study, we explored various Deep learning(DL) approach to classify COVID-19, other viral pneumonia and normal. For the original dataset and lung-segmented dataset, the pre-trained AlexNet, SqueezeNet, ResNet18, DenseNet201 were transfer-trained and validated for 3 class - COVID-19, viral pneumonia, normal. In the results, AlexNet showed the highest mean accuracy of 99.15±2.69% and fastest training time of 1.61±0.56 min among 4 pre-trained neural networks. In this study, we demonstrated the performance of 4 pre-trained neural networks in COVID-19 diagnosis with CXR images. Further, we plotted the class activation map(CAM) of each network and demonstrated that the lung-segmentation pre-processing improve the performance of COVID-19 classifier with CXR images by excluding background features.

Missed Lung Cancers on Chest Radiograph: An Illustrative Review of Common Blind Spots on Chest Radiograph with Emphasis on Various Radiologic Presentations of Lung Cancers (놓치기 쉬운 폐암: 흉부 X선 진단의 함정에 대한 이해와 다양한 폐암 영상 소견의 중요성)

  • Goun Choi;Bo Da Nam;Jung Hwa Hwang;Ki-Up Kim;Hyun Jo Kim;Dong Won Kim
    • Journal of the Korean Society of Radiology
    • /
    • v.81 no.2
    • /
    • pp.351-364
    • /
    • 2020
  • Missed lung cancers on chest radiograph (CXR) may delay the diagnosis and affect the prognosis. CXR is the primary imaging modality to evaluate the lungs and mediastinum in daily practice. The purpose of this article is to review chest radiographs for common blind spots and highlight the importance of various radiologic presentations in primary lung cancer to avoid significant diagnostic errors on CXR.

COVID-19 Chest X-ray reading Technique based on Deep Learning (흉부 X-ray 사진 분석을 통한 코로나 판독)

  • Kim, Sung-Jung;Yoo, JaeChern
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2021.01a
    • /
    • pp.31-32
    • /
    • 2021
  • 신종 코로나바이러스 감염증(Coronavirus disease 2019; COVID-19)이 빠르게 확산됨에 따라 세계적인 전염병 대유행인 팬데믹(Pandemic)으로 선언되었다. 감염자들은 꾸준히 증가하고 있고 최근에는, 무증상 감염자들이 나타나고 있어 의심 환자를 조기에 판단하고 선별할 수 있는 기술이 필요하다. 본 논문에서는 흉부 방사선 검사(chest Radiography; CXR) 영상을 딥러닝(Deep Learning)하여 정상인, 폐렴 환자, 코로나바이러스 감염자를 분류할 수 있도록 한다.

  • PDF

Artificial Intelligence-Based Identification of Normal Chest Radiographs: A Simulation Study in a Multicenter Health Screening Cohort

  • Hyunsuk Yoo;Eun Young Kim;Hyungjin Kim;Ye Ra Choi;Moon Young Kim;Sung Ho Hwang;Young Joong Kim;Young Jun Cho;Kwang Nam Jin
    • Korean Journal of Radiology
    • /
    • v.23 no.10
    • /
    • pp.1009-1018
    • /
    • 2022
  • Objective: This study aimed to investigate the feasibility of using artificial intelligence (AI) to identify normal chest radiography (CXR) from the worklist of radiologists in a health-screening environment. Materials and Methods: This retrospective simulation study was conducted using the CXRs of 5887 adults (mean age ± standard deviation, 55.4 ± 11.8 years; male, 4329) from three health screening centers in South Korea using a commercial AI (Lunit INSIGHT CXR3, version 3.5.8.8). Three board-certified thoracic radiologists reviewed CXR images for referable thoracic abnormalities and grouped the images into those with visible referable abnormalities (identified as abnormal by at least one reader) and those with clearly visible referable abnormalities (identified as abnormal by at least two readers). With AI-based simulated exclusion of normal CXR images, the percentages of normal images sorted and abnormal images erroneously removed were analyzed. Additionally, in a random subsample of 480 patients, the ability to identify visible referable abnormalities was compared among AI-unassisted reading (i.e., all images read by human readers without AI), AI-assisted reading (i.e., all images read by human readers with AI assistance as concurrent readers), and reading with AI triage (i.e., human reading of only those rendered abnormal by AI). Results: Of 5887 CXR images, 405 (6.9%) and 227 (3.9%) contained visible and clearly visible abnormalities, respectively. With AI-based triage, 42.9% (2354/5482) of normal CXR images were removed at the cost of erroneous removal of 3.5% (14/405) and 1.8% (4/227) of CXR images with visible and clearly visible abnormalities, respectively. In the diagnostic performance study, AI triage removed 41.6% (188/452) of normal images from the worklist without missing visible abnormalities and increased the specificity for some readers without decreasing sensitivity. Conclusion: This study suggests the feasibility of sorting and removing normal CXRs using AI with a tailored cut-off to increase efficiency and reduce the workload of radiologists.

Measurement of Diaphragm in Normal Human (정상인의 가로막(diaphragm) 높이와 만곡도 계측)

  • Kim, Ham-Gyum;Ma, Sang-Chull
    • Journal of radiological science and technology
    • /
    • v.30 no.4
    • /
    • pp.335-341
    • /
    • 2007
  • General anatomy classifies diaphragm as muscle of boundary between chest and abdomen, while radiology divides it into right and left hemidiaphragm, because it is more advantageous in radiological diagnosis on chest and abdomen. Based on these anatomic characteristics of diaphragm, this study aimed to measure the height and curvature of right and left diaphragm in simple chest radiography. As a result, this study came to the following conclusions : 1. For all subjects who joined this study, it was found that their mean transverse diameter in internal diameter of thorax(ID) amounted to 293.3 mm(min. 221.0 mm, max 335.3 mm). 2. For the right and left height of diaphragm, it was found that 81.4% showed higher right diaphragm ; 16.2% showed equivalent height between right and left diaphragm ; and only 2.4% showed higher left diaphragm. 3. For higher right diaphragm, it was found that the mean height of right diaphragm amounted to 15.2 mm(min. height = 2.0 mm, max. height = 41.7 mm). 4. For higher left diaphragm, it was found that the mean height of left diaphragm amounted to 11.5 mm(min. height = 4.7 mm, max. height = 30.4 mm). 5. The mean curvature of right diaphragm amounted to 22.9 mm(min. curvature = 10.4 mm, max. curvature = 37.3 mm). 6. The mean curvature of left diaphragm amounted to 22.4 mm(min. curvature = 11.3 mm, max. curvature = 42.2 mm). 7. For possible associations between ID and right/left diaphragm curvature, it was noted that ID was in significantly positive correlations with right diaphragm curvature(r= .427, p<.001) and left diaphragm curvature(r= .425, p<.001) on statistical level. 8. For possible associations between right and left diaphragm curvature, it was found that right diaphragm curvature was in significantly positive correlations with left diaphragm curvature(r= .403, p<.001).

  • PDF

A Study on Findings from Simple Chest Radiographes without Any Clinical Symptoms (임상적 증상이 없는 흉부 단순X선영상 소견에 대한 분석)

  • Kim, Ham-Gyum
    • Journal of radiological science and technology
    • /
    • v.30 no.2
    • /
    • pp.95-104
    • /
    • 2007
  • In this study, the analysis on findings from simple chest radiography(CXR) test with total 1,669 subjects without any special clinical symptom came to the following conclusions : 1. In terms of the general characteristics of subjects hereof, male and female group accounted for 55.2% and 44.8% respectively out of all 1,669 people. 2. Pulmonary disease cases amounted to 249 persons(14.9%) out of all subjects. 3. In the analysis on prevalence rate by age distribution, it was noted that the older age led to the more number of diseases, which was demonstrated by age 34 or younger(6.1%), age $35{\sim}39(9.7%)$, age $40{\sim}49(13.3\;%)$, and age 50 or older(30.8%). 4. In regard of pulmonary disease alone, the region of onset was represented primarily by right upper lobe, which was followed by both upper lobe and left upper lobe, respectively. 5. In terms of disease types, it was found that most cases were represented by pulmonary nodule(55.0%), which was followed by cardiomegaly(24.5%), CP angle blunting(4.8%), scoliosis(4.6%), tortuous aorta(2.8%), bronchial luminal dilatation(2.4%), and pleural thickening(2.0%). However, dextrocardia, cystic dilation of bronchus, cavitary lesion, and lung collapse accounted for relatively low rate(0.4% respectively). 6. In terms of disease types by sex, it was found that male group accounted for higher percentage of having pulmonary nodule than female group, while the latter accounted for higher percentage of having cardiomegaly, tortuous aorta and scoliosis than the former. 7. In terms of disease types by age distribution, it was noted that age 34 or younger group accounted for higher percentage of scoliosis than any other age groups, while age $40{\sim}49$ group, age $35{\sim}39$ group, and age 50 or older group represented the case of CP angle blunting, pulmonary nodule, and cardiomegaly/tortuous aorta, respectively.

  • PDF

Conventional Versus Artificial Intelligence-Assisted Interpretation of Chest Radiographs in Patients With Acute Respiratory Symptoms in Emergency Department: A Pragmatic Randomized Clinical Trial

  • Eui Jin Hwang;Jin Mo Goo;Ju Gang Nam;Chang Min Park;Ki Jeong Hong;Ki Hong Kim
    • Korean Journal of Radiology
    • /
    • v.24 no.3
    • /
    • pp.259-270
    • /
    • 2023
  • Objective: It is unknown whether artificial intelligence-based computer-aided detection (AI-CAD) can enhance the accuracy of chest radiograph (CR) interpretation in real-world clinical practice. We aimed to compare the accuracy of CR interpretation assisted by AI-CAD to that of conventional interpretation in patients who presented to the emergency department (ED) with acute respiratory symptoms using a pragmatic randomized controlled trial. Materials and Methods: Patients who underwent CRs for acute respiratory symptoms at the ED of a tertiary referral institution were randomly assigned to intervention group (with assistance from an AI-CAD for CR interpretation) or control group (without AI assistance). Using a commercial AI-CAD system (Lunit INSIGHT CXR, version 2.0.2.0; Lunit Inc.). Other clinical practices were consistent with standard procedures. Sensitivity and false-positive rates of CR interpretation by duty trainee radiologists for identifying acute thoracic diseases were the primary and secondary outcomes, respectively. The reference standards for acute thoracic disease were established based on a review of the patient's medical record at least 30 days after the ED visit. Results: We randomly assigned 3576 participants to either the intervention group (1761 participants; mean age ± standard deviation, 65 ± 17 years; 978 males; acute thoracic disease in 472 participants) or the control group (1815 participants; 64 ± 17 years; 988 males; acute thoracic disease in 491 participants). The sensitivity (67.2% [317/472] in the intervention group vs. 66.0% [324/491] in the control group; odds ratio, 1.02 [95% confidence interval, 0.70-1.49]; P = 0.917) and false-positive rate (19.3% [249/1289] vs. 18.5% [245/1324]; odds ratio, 1.00 [95% confidence interval, 0.79-1.26]; P = 0.985) of CR interpretation by duty radiologists were not associated with the use of AI-CAD. Conclusion: AI-CAD did not improve the sensitivity and false-positive rate of CR interpretation for diagnosing acute thoracic disease in patients with acute respiratory symptoms who presented to the ED.