Chest X-ray For Detection Research Articles

Intersection of Performance, Interpretability, and Fairness in Neural Prototype Tree for Chest X-Ray Pathology Detection: Algorithm Development and Validation Study.

While deep learning classifiers have shown remarkable results in detecting chest X-ray (CXR) pathologies, their adoption in clinical settings is often hampered by the lack of transparency. To bridge this gap, this study introduces the neural prototype tree (NPT), an interpretable image classifier that combines the diagnostic capability of deep learning models and the interpretability of the decision tree for CXR pathology detection. This study aimed to investigate the utility of the NPT classifier in 3 dimensions, including performance, interpretability, and fairness, and subsequently examined the complex interaction between these dimensions. We highlight both local and global explanations of the NPT classifier and discuss its potential utility in clinical settings. This study used CXRs from the publicly available Chest X-ray 14, CheXpert, and MIMIC-CXR datasets. We trained 6 separate classifiers for each CXR pathology in all datasets, 1 baseline residual neural network (ResNet)-152, and 5 NPT classifiers with varying levels of interpretability. Performance, interpretability, and fairness were measured using the area under the receiver operating characteristic curve (ROC AUC), interpretation complexity (IC), and mean true positive rate (TPR) disparity, respectively. Linear regression analyses were performed to investigate the relationship between IC and ROC AUC, as well as between IC and mean TPR disparity. The performance of the NPT classifier improved as the IC level increased, surpassing that of ResNet-152 at IC level 15 for the Chest X-ray 14 dataset and IC level 31 for the CheXpert and MIMIC-CXR datasets. The NPT classifier at IC level 1 exhibited the highest degree of unfairness, as indicated by the mean TPR disparity. The magnitude of unfairness, as measured by the mean TPR disparity, was more pronounced in groups differentiated by age (chest X-ray 14 0.112, SD 0.015; CheXpert 0.097, SD 0.010; MIMIC 0.093, SD 0.017) compared to sex (chest X-ray 14 0.054 SD 0.012; CheXpert 0.062, SD 0.008; MIMIC 0.066, SD 0.013). A significant positive relationship between interpretability (ie, IC level) and performance (ie, ROC AUC) was observed across all CXR pathologies (P<.001). Furthermore, linear regression analysis revealed a significant negative relationship between interpretability and fairness (ie, mean TPR disparity) across age and sex subgroups (P<.001). By illuminating the intricate relationship between performance, interpretability, and fairness of the NPT classifier, this research offers insightful perspectives that could guide future developments in effective, interpretable, and equitable deep learning classifiers for CXR pathology detection.

The Role of Conventional Chest X-Ray in Detection and Evaluation of Long COVID-19

A 52-year-old woman came to ER with shortness of breath, fever, and cough with oxygen saturation of 80% on room air. The COVID-19 PCR test was positive. CXR showed it infiltrates ground-glass opacities in both lungs and cardiomegaly. Chest X-ray at discharge from the hospital showed decreased infiltrate and fibrosis in two lung. She came to the ER 4 days later with worsened shortness of breath. CXR showed diffuse infiltrates in both lungs and cardiomegaly. Discussion: COVID-19 patients can develop long-term sequelae and complications (long COVID-19). Although CT is now considered the primary investigation for COVID-19, CXR is still valid for detecting and monitoring its progression. Conclusion: CXR is useful for detecting and monitoring the rapid advancement of lung abnormalities in long COVID-19.&#x0D; Wanita 52 tahun datang ke IGD dengan sesak napas, demam, batuk; dengan saturasi oksigen 80% udara ruangan. Tes PCR COVID-19 positif. Rontgen toraks menunjukkan infiltrat di kedua lapangan paru, ground glass opacities, dan kardiomegali. Rontgen toraks saat keluar rumah sakit menunjukkan penurunan gambaran infiltrat dan fibrosis pada dua lapangan paru. Empat hari setelah keluar rumah sakit pasien kembali sesak napas. Rontgen toraks menunjukkan infiltrat difus dikedua paru dan kardiomegali. Diskusi: Pasien COVID-19 dapat mengalami sekuele jangka panjang dan komplikasi, disebut long COVID-19. Walaupun CT-scan menjadi pendeteksi utama COVID-19, rontgen toraks masih bermanfaat untuk memantau progresi sindrom tersebut. Simpulan: Rontgen toraks berguna untuk deteksi dan memantau kelainan paru pada long COVID-19.

The Role of Conventional Chest X-Ray in Detection and Evaluation of Long COVID-19

&lt;p&gt;A 52-year-old woman came to ER with shortness of breath, fever, and cough with oxygen saturation of 80% on room air. The COVID-19 PCR test was positive. CXR showed it infiltrates ground-glass opacities in both lungs and cardiomegaly. Chest X-ray at discharge from the hospital showed decreased infiltrate and fibrosis in two lung. She came to the ER 4 days later with worsened shortness of breath. CXR showed diffuse infiltrates in both lungs and cardiomegaly. Discussion: COVID-19 patients can develop long-term sequelae and complications (long COVID-19). Although CT is now considered the primary investigation for COVID-19, CXR is still valid for detecting and monitoring its progression. Conclusion: CXR is useful for detecting and monitoring the rapid advancement of lung abnormalities in long COVID-19.&lt;/p&gt;&lt;p&gt;Wanita 52 tahun datang ke IGD dengan sesak napas, demam, batuk; dengan saturasi oksigen 80% udara ruangan. Tes PCR COVID-19 positif. Rontgen toraks menunjukkan infiltrat di kedua lapangan paru, ground glass opacities, dan kardiomegali. Rontgen toraks saat keluar rumah sakit menunjukkan penurunan gambaran infiltrat dan fibrosis pada dua lapangan paru. Empat hari setelah keluar rumah sakit pasien kembali sesak napas. Rontgen toraks menunjukkan infiltrat difus dikedua paru dan kardiomegali. Diskusi: Pasien COVID-19 dapat mengalami sekuele jangka panjang dan komplikasi, disebut long COVID-19. Walaupun CT-scan menjadi pendeteksi utama COVID-19, rontgen toraks masih bermanfaat untuk memantau progresi sindrom tersebut.&lt;/p&gt;&lt;p&gt; &lt;/p&gt;

Diagnostic Performance of Standard and Inverted Grey-Scale CXR in Detection of Lung Lesions in COVID-19 Patients: A Single Institute Study in the Region of Abu Dhabi

Purpose: This study aimed at evaluating the diagnostic performance of standard greyscale and inverted greyscale Chest X-ray (CXR) using Computed Tomography (CT) scan as a gold standard. Methods: In this retrospective study, electronic medical records of 120 patients who had valid CXR and High-resolution CT (HRCT) within less than 24 hours after having a positive COVID-19 RT-PCR test during the period from May 19th to May 23rd, 2020, in a single tertiary care center were reviewed. PA chest radiographs were presented on 2 occasions to 5 radiologists to evaluate the role and appropriateness of standard greyscale and inverted greyscale chest radiographs (CXR) when images are viewed on high-specification viewing systems using a primary display monitor and compared to computed tomography (CT) findings for screening and management of suspected or confirmed COVID-19 patients. Results: Ninety-six (80%) patients had positive CT findings, 81 (67.5%) had positive grey scale CXR lesions, and 25 (20.8%) had better detection in the inverted grey scale CXR. The CXR sensitivity for COVID-19 pneumonia was 93.8% (95% CI (86.2% - 98.0%) and the specificity was 48.7% (95% CI (32.4% - 65.2%). The CXR sensitivity of detecting lung lesions was slightly higher in male (95.1% (95% CI (86.3% - 99.0%)) than female (90.0% (95% CI (68.3% - 98.8%)), while the specificity was 48.0% (95% CI (27.8% - 68.7%) and 50.0% (95% CI (23.0% - 77.0%) in males and females, respectively. However, no significant difference was detected in ROC area between men and women. Conclusion: The sensitivity of detecting lung lesions of CXR was relatively high, particularly in men. The results of the study support the idea of considering conventional radiographs as an important diagnostic tool in suspected COVID-19 patients, especially in healthcare facilities where there is no access to HRCT scans. CXR shows high sensitivity for detecting lung lesions in HRCT confirmed COVID-19 patients. Better detection of lesions was noted in the inverted greyscale CXR in 20.8% of cases, with positive findings in standard greyscale CXR. Conventional radiographs can be used as diagnostic tools in suspected COVID-19 patients, especially in healthcare facilities where there is no access to HRCT scans.

Lung Ultrasound Is More Sensitive for Hospitalized Consolidated Pneumonia Diagnosis Compared to CXR in Children.

Background: Pneumonia is the leading cause of death among children; thus, a correct early diagnosis would be ideal. The imagistic diagnosis still uses chest X-ray (CXR), but lung ultrasound (LUS) proves to be reliable for pneumonia diagnosis. The aim of our study was to evaluate the sensitivity and specificity of LUS compared to CXR in consolidated pneumonia. Methods: Children with clinical suspicion of bacterial pneumonia were screened by LUS for pneumonia, followed by CXR. The agreement relation between LUS and CXR regarding the detection of consolidation was evaluated by Cohen’s kappa test. Results: A total of 128 patients with clinical suspicion of pneumonia were evaluated; 74 of them were confirmed by imagery and biological inflammatory markers. The highest frequency of pneumonia was in the 0–3 years age group (37.83%). Statistical estimation of the agreement between LUS and CXR in detection of the consolidation found an almost perfect agreement, with a Cohen’s kappa coefficient of K = 0.89 ± 0.04 SD, p = 0.000. Sensitivity of LUS was superior to CXR in detection of consolidations. Conclusion: Lung ultrasound is a reliable method for the detection of pneumonia consolidation in hospitalized children, with sensitivity and specificity superior to CXR. LUS should be used for rapid and safe evaluation of child pneumonia.

Value of Chest Radiography and Electrocardiography in Diagnosis of Congenital Heart Diseases in Pediatrics in Comparison with Echocardiography

Background: Advances in echocardiographic technology have made a revolution in the quality of image and acquisition of data; however, it has also increased the cost of this diagnostic test. Objective: To determine if chest x ray (CXR) and electrocardiography (ECG) are still valuable in evaluation of new patients with heart murmurs in presence of echocardiography (ECHO). Patients and Methods: This was a six month, prospective clinical study undertaken in the Outpatient Pediatric Cardiology Unit, Zagazig University Children Hospital, between April 2018 and October 2018. The study included 44 patients (boys and girls) between the ages of 1 month and 14 years. All cases were subjected to detailed history, full clinical examination and imaging modalities as CXR, ECG and ECHO. Results: The highest accuracy of CXR was detection of right atrium (96.9%) followed by left ventricle (95.1%). The highest accuracy of ECG was detection of right atrium (98.5%) and left ventricle (95.5%). The agreement between CXR and ECHO in detection of pulmonary hypertension was 0.4 while that for ECG was 0.6. The agreement between CXR and ECHO in diagnosis of congenital heart diseases was 0.6 and that for ECG was 0.9. The accuracy of ECG in diagnosis of congenital heart disease was 95.4% while that for CXR was 70.4%. Conclusion: ECG is more accurate than CXR in detection of cardiac chamber enlargement and heart diseases in general but echocardiography still the most accurate method in this regard. The chest radiograph and electrocardiogram had limited values in the diagnosis of pulmonary pressure changes and echocardiography is the method to be used in this regard.

Value of Chest Radiography and Electrocardiography in Diagnosis of Congenital Heart Diseases in Pediatrics in Comparison with Echocardiography

Background: Advances in echocardiographic technology have made a revolution in the quality of image and acquisition of data; however, it has also increased the cost of this diagnostic test. Objective: To determine if chest x ray (CXR) and electrocardiography (ECG) are still valuable in evaluation of new patients with heart murmurs in presence of echocardiography (ECHO). Patients and Methods: This was a six month, prospective clinical study undertaken in the Outpatient Pediatric Cardiology Unit, Zagazig University Children Hospital, between April 2018 and October 2018. The study included 44 patients (boys and girls) between the ages of 1 month and 14 years. All cases were subjected to detailed history, full clinical examination and imaging modalities as CXR, ECG and ECHO. Results: The highest accuracy of CXR was detection of right atrium (96.9%) followed by left ventricle (95.1%). The highest accuracy of ECG was detection of right atrium (98.5%) and left ventricle (95.5%). The agreement between CXR and ECHO in detection of pulmonary hypertension was 0.4 while that for ECG was 0.6. The agreement between CXR and ECHO in diagnosis of congenital heart diseases was 0.6 and that for ECG was 0.9. The accuracy of ECG in diagnosis of congenital heart disease was 95.4% while that for CXR was 70.4%. Conclusion: ECG is more accurate than CXR in detection of cardiac chamber enlargement and heart diseases in general but echocardiography still the most accurate method in this regard. The chest radiograph and electrocardiogram had limited values in the diagnosis of pulmonary pressure changes and echocardiography is the method to be used in this regard.

Use of Lung Ultrasound in Detection of Complications of Respiratory Distress Syndrome

Repeated chest radiography is required for the diagnosis and follow-up of neonates with respiratory distress syndrome (RDS) and carries the risk of radiation hazards. Lung ultrasound (LUS) is a non-invasive bedside diagnostic tool that has proven to be effective in the diagnosis of RDS. Our aim was to assess the role of LUS with respect to the standard chest X-ray (CXR) in the detection of complications of RDS in neonates. Ninety premature newborns of both genders with RDS (mean gestational age = 29.91 ± 1.33 wk) and 40 premature babies as a control group were involved in this study. All patients underwent initial clinical assessment as well as CXR and LUS. Those who presented with respiratory distress and/or exhibited deterioration of oxygenation parameters were followed by CXR and, within 4 h, by LUS. Alveolo-interstitial syndrome and pleural line abnormalities were detected in all cases (100%) in the initial assessment, patchy consolidation was detected in 34 cases and white lung was detected in 80 cases. Alveolo-interstitial syndrome was detected in 19 controls. In follow-up of the patients, LUS was superior to CXR in detection of consolidation and sub-pleural atelectasis, but not in detection of pneumothorax. We concluded that bedside LUS is a good non-hazardous alternative tool in the early detection and follow-up of RDS in the neonatal intensive care unit; it could be of value in reducing exposure to unnecessary radiation.

Diagnostic Accuracy of Chest x-Ray and Ultrasonography in Detection of Community Acquired Pneumonia; a Brief Report.

Chest x-ray (CXR) is the simplest diagnostic tool of Community Acquired Pneumonia (CAP), but it has some limitation. Therefore, the aim of this study is comparing the diagnostic accuracy of CXR and chest ultrasonography (CUS) in detection of CAP. In the present study, a consecutive sample of suspected patients with CAP was underwent CUS, CXR, and chest computed tomography (CT) scan. Diagnostic accuracy of CUS and CXR was assessed by calculating the sensitivity, specificity, predictive values, and likelihood ratios using SPSS 20 statistical software. 30 patients with CAP were enrolled (93.3% male with mean age of 63.8 ± 18.3 years). Sensitivity of CUS and CXR in detection of CAP were 100.0% (95% Cl: 85.4-100.0) and 93.1% (95% Cl: 75.8-98.8), respectively. Specificity of CXR was 0.0 (95% Cl: 0.0-94.5), while the CUS specificity was not calculable. Findings of the present study demonstrated on the higher diagnostic accuracy of CUS versus CXR in detection of pneumonia.

Chest X-ray in the follow-up of renal cell carcinoma

To evaluate the value of chest X-ray in the follow-up of surgically treated T1-3N0M0 renal cell carcinoma. We performed retrospective analysis of patients that underwent surgical treatment of a localized renal cell carcinoma (T1-3N0M0) between January 1993 and July 2010. Data on frequency and results of performed chest X-rays were collected from patients' records. In 17.5 years, 249 patients with a T1-3N0M0 renal cell carcinoma underwent a radical or partial nephrectomy. In 221 patients, 823 chest X-rays were performed during a median follow-up of 3.3 years (range 0.5-17 years). In 19 patients, a pulmonary recurrence occurred, of which 10 were not detected by the regular follow-up. Of the 9 patients that were diagnosed with a pulmonary recurrence with a chest X-ray during follow-up, 7 were asymptomatic at the time of diagnosis, and the chest X-ray has led to the detection; 0.85 % of the performed chest X-rays (7/823) have led to the detection of asymptomatic lung metastases. Due to the low yield of chest X-ray for detection of asymptomatic pulmonary recurrences, it has very low clinical value in the follow-up after nephrectomy for T1-3N0M0 renal cell carcinoma.

Computer-aided diagnosis in breast tomosynthesis and chest CT.

Computer-aided diagnosis (CAD), in the general sense, includes computer-aided detection and characterization of abnormalities on medical images. The usefulness of CAD for assisting radiologists in detection of breast cancer in screening mammography has been demonstrated by a number of prospective clinical trials in recent years. The development of CAD in other areas is also being actively pursued by researchers. In this talk, the recent work in two areas of CAD, digital breast tomosynthesis (DBT) and chest computed tomography (CT), in the CAD Research Laboratory at the University of Michigan will be reviewed. DBT is a new modality under development for breast imaging. The quasi-3D information in DBT alleviates the problem of overlapping tissue in mammography and holds the promise to improve the sensitivity for cancer detection. DBT image analysis can be performed in the 3D reconstructed volume of the 2D projection view (PV) images. DBT image quality depend on the image acquisition parameters, reconstruction method and parameters. The flexibility in image processing approaches makes CAD development for DBT interesting and challenging. out early experiences in the development of image segmentation and features extraction technique for mass detection and characterization in DBT will be discussed. The performances of the CAD systems using the 2D, 3D, and combined 2D and 3D approaches will be compared. CT has been shown to be superior to chest x-ray in detection of small lung nodules and thus lung cancer screening with CT is still being debated, many research groups are developing CAD methods for detection and characterization of lung nodules in chest CT scans. The specific prescreening, segmentation, and feature extraction techniques designed for our lung nodule detection and characterization systems will be discussed. The effects of CAD on radiologists' accuracy in nodules detection and characterization in CT scans will be demonstrated by results of observer ROC studies. There are similarities in the approaches to developing CAD methods in 3D image volumes such as DBT and CT, these experiences will facilitate the development of CAD systems for other diseases in 3D modalities.

High-resolution CT pulmonary findings in idiopathic ankylosing spondylitis: correlations with clinical assessment, plain chest X-ray and pulmonary function tests

Aim: The aim of this study is to evaluate pulmonary manifestations of ankylosing spondylitis on high-resolution computed tomography (CT) scan and to correlate these findings with clinical assessment, plain chest X-ray and pulmonary function tests. Methods: The study comprised 32 patients (26 males [81.3%], 6 females [18.8%]) who met the modified New York criteria for diagnosis of idiopathic ankylosing spondylitis; in addition 10 normal subjects not complaining of any respiratory symptoms and matched for age and sex served as a control group. All patients were subjected to full history-taking, full clinical examination, chest X-ray, high-resolution computed tomography (HRCT) chest and pulmonary function tests. Results: The abnormalities on HRCT included evidence of apical lung fibrosis in two patients (6.3%), interstitial lung disease in six (18.8%), minor interstitial abnormalities in eight (25%), bronchiectasis in four (12.5%), lung nodules in three (6.3%) and pleural thickening in five patients (6.3%). Nine patients (28.1%) showed negative findings by chest X-ray which revealed positive findings on HRCT. Five patients (15.6%) showed positive findings on both HRCT and plain chest X-ray, and 18 patients (56.3%) showed no findings on both HRCT and chest X-ray. Four patients (12.5%) showed normal pulmonary function tests, 25 (78.13%) showed restrictive pattern, 17 (53.13%) had obstructive pattern and 26 patients (81.3%) had diffusion defects. Conclsuion: The identification of non-apical minor basal interstitial lung disease in our study which was previously reported in other studies, raises a possible association to ankylosing spondylitis. High-resolution CT scan is more sensitive than chest X-rays in detection of such minor interstitial lung disease (ILD), and other parenchymal lung changes.

Chest X-rays for detection of lung cancer

Chest X-rays for detection of lung cancer