Enhancing thoracic disease detection using chest X-rays from PubMed Central Open Access

Large chest X-rays (CXR) datasets have been collected to train deep learning models to detect thorax pathology on CXR. However, most CXR datasets are from single-center studies and the collected pathologies are often imbalanced. The aim of this study was to automatically construct a public, weakly-labeled CXR database from articles in PubMed Central Open Access (PMC-OA) and to assess model performance on CXR pathology classification by using this database as additional training data. Our framework includes text extraction, CXR pathology verification, subfigure separation, and image modality classification. We have extensively validated the utility of the automatically generated image database on thoracic disease detection tasks, including Hernia, Lung Lesion, Pneumonia, and pneumothorax. We pick these diseases due to their historically poor performance in existing datasets: the NIH-CXR dataset (112,120 CXR) and the MIMIC-CXR dataset (243,324 CXR). We find that classifiers fine-tuned with additional PMC-CXR extracted by the proposed framework consistently and significantly achieved better performance than those without (e.g., Hernia: 0.9335 vs 0.9154; Lung Lesion: 0.7394 vs. 0.7207; Pneumonia: 0.7074 vs. 0.6709; Pneumothorax 0.8185 vs. 0.7517, all in AUC with p< 0.0001) for CXR pathology detection. In contrast to previous approaches that manually submit the medical images to the repository, our framework can automatically collect figures and their accompanied figure legends. Compared to previous studies, the proposed framework improved subfigure segmentation and incorporates our advanced self-developed NLP technique for CXR pathology verification. We hope it complements existing resources and improves our ability to make biomedical image data findable, accessible, interoperable, and reusable.

PneuNet: deep learning for COVID-19 pneumonia diagnosis on chest X-ray image analysis using Vision Transformer

A long-standing challenge in pneumonia diagnosis is recognizing the pathological lung texture, especially the ground-glass appearance pathological texture. One main difficulty lies in precisely extracting and recognizing the pathological features. The patients, especially those with mild symptoms, show very little difference in lung texture, neither conventional computer vision methods nor convolutional neural networks perform well on pneumonia diagnosis based on chest X-ray (CXR) images. In the meanwhile, the Coronavirus Disease 2019 (COVID-19) pandemic continues wreaking havoc around the world, where quick and accurate diagnosis backed by CXR images is in high demand. Rather than simply recognizing the patterns, extracting feature maps from the original CXR image is what we need in the classification process. Thus, we propose a Vision Transformer (VIT)-based model called PneuNet to make an accurate diagnosis backed by channel-based attention through X-ray images of the lung, where multi-head attention is applied on channel patches rather than feature patches. The techniques presented in this paper are oriented toward the medical application of deep neural networks and VIT. Extensive experiment results show that our method can reach 94.96% accuracy in the three-categories classification problem on the test set, which outperforms previous deep learning models.

Development and validation of a CT-based radiomics model for differentiating pneumonia-like primary pulmonary lymphoma from infectious pneumonia: A multicenter study

BACKGROUND

Pneumonia-like primary pulmonary lymphoma (PPL) was commonly misdiagnosed as infectious pneumonia, leading to delayed treatment. The purpose of this study was to establish a computed tomography (CT)-based radiomics model to differentiate pneumonia-like PPL from infectious pneumonia.

METHODS

In this retrospective study, 79 patients with pneumonia-like PPL and 176 patients with infectious pneumonia from 12 medical centers were enrolled. Patients from center 1 to center 7 were assigned to the training or validation cohort, and the remaining patients from other centers were used as the external test cohort. Radiomics features were extracted from CT images. A three-step procedure was applied for radiomics feature selection and radiomics signature building, including the inter- and intra-class correlation coefficients (ICCs), a one-way analysis of variance (ANOVA), and least absolute shrinkage and selection operator (LASSO). Univariate and multivariate analyses were used to identify the significant clinicoradiological variables and construct a clinical factor model. Two radiologists reviewed the CT images for the external test set. Performance of the radiomics model, clinical factor model, and each radiologist were assessed by receiver operating characteristic, and area under the curve (AUC) was compared.

RESULTS

A total of 144 patients (44 with pneumonia-like PPL and 100 infectious pneumonia) were in the training cohort, 38 patients (12 with pneumonia-like PPL and 26 infectious pneumonia) were in the validation cohort, and 73 patients (23 with pneumonia-like PPL and 50 infectious pneumonia) were in the external test cohort. Twenty-three radiomics features were selected to build the radiomics model, which yielded AUCs of 0.95 (95% confidence interval [CI]: 0.94-0.99), 0.93 (95% CI: 0.85-0.98), and 0.94 (95% CI: 0.87-0.99) in the training, validation, and external test cohort, respectively. The AUCs for the two readers and clinical factor model were 0.74 (95% CI: 0.63-0.83), 0.72 (95% CI: 0.62-0.82), and 0.73 (95% CI: 0.62-0.84) in the external test cohort, respectively. The radiomics model outperformed both the readers' interpretation and clinical factor model ( P <0.05).

CONCLUSIONS

The CT-based radiomics model may provide an effective and non-invasive tool to differentiate pneumonia-like PPL from infectious pneumonia, which might provide assistance for clinicians in tailoring precise therapy.

Computed tomography-based radiomics for the differential diagnosis of pneumonitis in stage IV non-small cell lung cancer patients treated with immune checkpoint inhibitors

INTRODUCTION

Immunotherapy-induced pneumonitis (IIP) is a serious side-effect which requires accurate diagnosis and management with high-dose corticosteroids. The differential diagnosis between IIP and other types of pneumonitis (OTP) remains challenging due to similar radiological patterns. This study was aimed to develop a prediction model to differentiate IIP from OTP in patients with stage IV non-small cell lung cancer (NSCLC) who developed pneumonitis during immunotherapy.

METHODS

Consecutive patients with metastatic NSCLC treated with immunotherapy in six centres in the Netherlands and Belgium from 2017 to 2020 were reviewed and cause-specific pneumonitis events were identified. Seven regions of interest (segmented lungs and spheroidal/cubical regions surrounding the inflammation) were examined to extract the most predictive radiomic features from the chest computed tomography images obtained at pneumonitis manifestation. Models were internally tested regarding discrimination, calibration and decisional benefit. To evaluate the clinical application of the models, predicted labels were compared with the separate clinical and radiological judgements.

RESULTS

A total of 556 patients were reviewed; 31 patients (5.6%) developed IIP and 41 patients developed OTP (7.4%). The line of immunotherapy was the only predictive factor in the clinical model (2nd versus 1st odds ratio = 0.08, 95% confidence interval:0.01-0.77). The best radiomic model was achieved using a 75-mm spheroidal region of interest which showed an optimism-corrected area under the receiver operating characteristic curve of 0.83 (95% confidence interval:0.77-0.95) with negative and positive predictive values of 80% and 79%, respectively. Good calibration and net benefits were achieved for the radiomic model across the entire range of probabilities. A correct diagnosis was provided by the radiomic model in 10 out of 12 cases with non-conclusive radiological judgements.

CONCLUSION

Radiomic biomarkers applied to computed tomography imaging may support clinicians making the differential diagnosis of pneumonitis in patients with NSCLC receiving immunotherapy, especially when the radiologic assessment is non-conclusive.

Adenosquamous carcinoma lung radiologically mimicking pneumonia: A potentially disastrous diagnostic challenge in an unusual malignancy

Lung cancer is among the most frequently diagnosed cancers and the world's leading cause of cancer-related death. Radiology remains the mainstay for timely diagnosis; however, atypical radiologic patterns are known, and these may be misdiagnosed as infectious or inflammatory pathology, particularly in the absence of smoking history. We report herein an account of an older male nonsmoker who presented radiologically with bilateral diffuse pulmonary infiltrates, simulating pneumonia, but was eventually diagnosed with adenosquamous lung carcinoma. The delay in diagnosis and subsequent unfortunate rapid deterioration of our patient serves as a reminder for clinicians to consider lung cancer in patients with clinical/radiologic findings suggestive of pneumonia, especially in nonsmokers or cases refractory to antibiotic therapy.

Organizing Pneumonia with NGS False-Positive Imaging Resembling Tuberculosis: a Case Report

BACKGROUND

Organizing pneumonia (OP) is a pathologic concept characterized by the formation of granulation tissue from fibroblasts, myofibroblasts, collagen, and fibrotic exudate in the respiratory fine bronchi, alveolar ducts, and alveoli. The clinical imaging of mechanized pneumonia is variable, and histopathological examination is required to clarify the nature of the lesion when imaging is atypical. We report a case of OP with imaging resem-blance to pulmonary tuberculosis and false-positive next-generation sequencing (NGS), which was first misdiag-nosed as pulmonary tuberculosis.

METHODS

Appropriate laboratory tests, alveolar lavage fluid NGS, chest CT, bronchoscopy, percutaneous lung puncture, pathology.

RESULTS

Chest CT showed a nodular high-density shadow in the lower lobe of the right lung. According to the chest CT, bronchoalveolar lavage was performed in the dorsal segment of the right lower lobe of the lung. NGS of lavage fluid: the sequence number of Moraxella osseae was 1,423; the sequence number of Prevotella melanogaster was 1,129. Based on lung histopathology, fibrous emboli and necrotic material were seen in the alveolar lumen, and the final diagnosis of the OP was confirmed.

CONCLUSIONS

It should be noted that physicians should not blindly believe the NGS result report. When the diagnosis is not clear and anti-infection treatment is ineffective, lung tissue should be obtained promptly for pathological examination to obtain pathological evidence to differentiate from misdiagnosed diseases.

Efficiency of Lung Ultrasonography in the Diagnosis and Follow-up of Viral Pneumonia in Newborn

OBJECTIVE

Lung ultrasonography (LUS) is a useful method for diagnosis of lung diseases such as respiratory distress syndrome, transient tachypnea of the newborn, pneumonia, and pneumothorax in the neonatal period. LUS has become an important tool in the diagnosis and follow-up of lung diseases. LUS is easy to apply at the bedside and is a practical and low-cost method for diagnosing pneumonia.

STUDY DESIGN

This study was conducted in neonatal intensive care unit of Dr. Sami Ulus Obstetrics, Children's Health and Diseases Training and Research Hospital. From September 2019 to April 2020, 50 patients who were diagnosed with viral pneumonia were included in the study. Also, 24 patients with sepsis-related respiratory failure were included in the study as a control group. LUS was performed at the bedside three times, by a single expert, once each before treatment for diagnosis, on discharge, and after discharge in outpatient clinic control.

RESULTS

Before treatment, LUS findings were lung consolidation with air bronchograms (50/50), pleural line abnormalities (35/50), B-pattern (25/50), disappearance of lung sliding (21/50), lung pulse (5/50), and pleural effusion (9/50). During discharge, we found significant changes: lung consolidation with air bronchograms (6/50), pleural line abnormalities (7/50), B-pattern (12/50), and pleural effusion (1/50) (p < 0.05). Outpatient clinic control LUS findings were lung consolidation with air bronchograms (0/50), pleural line abnormalities (0/50), B-pattern (0/50), disappearance of lung sliding (0/50), and pleural effusion (0/50) (p < 0.05). Also, B-pattern image, disappearance of lung sliding, and pleural line abnormalities were higher in control group (p < 0.05).

CONCLUSION

Ultrasound gives no hazard, and the application of bedside ultrasonography is comfortable for the patients. Pneumonia is a serious infection in the neonatal period. Repeated chest radiography may be required depending on the clinical condition of the patient with pneumonia. This study focuses on adequacy of LUS in neonatal pneumonia.

KEY POINTS

· Lung ultrasound is a practical and low-cost method in diagnosing pneumonia.. · Neonatal pneumonia is a very important cause of morbidity and mortality in NICU.. · We can evaluate neonatal pneumonia with combination of clinical presentations and LUS findings..

Cardiovascular abnormalities in chest radiographs of children with pneumonia, Uganda

Objective

To describe chest radiograph findings among children hospitalized with clinically diagnosed severe pneumonia and hypoxaemia at three tertiary facilities in Uganda.

Methods

The study involved clinical and radiograph data on a random sample of 375 children aged 28 days to 12 years enrolled in the Children's Oxygen Administration Strategies Trial in 2017. Children were hospitalized with a history of respiratory illness and respiratory distress complicated by hypoxaemia, defined as a peripheral oxygen saturation (SpO2) < 92%. Radiologists blinded to clinical findings interpreted chest radiographs using standardized World Health Organization method for paediatric chest radiograph reporting. We report clinical and chest radiograph findings using descriptive statistics.

Findings

Overall, 45.9% (172/375) of children had radiological pneumonia, 36.3% (136/375) had a normal chest radiograph and 32.8% (123/375) had other radiograph abnormalities, with or without pneumonia. In addition, 28.3% (106/375) had a cardiovascular abnormality, including 14.9% (56/375) with both pneumonia and another abnormality. There was no significant difference in the prevalence of radiological pneumonia or of cardiovascular abnormalities or in 28-day mortality between children with severe hypoxaemia (SpO2: < 80%) and those with mild hypoxaemia (SpO2: 80 to < 92%).

Conclusion

Cardiovascular abnormalities were relatively common among children hospitalized with severe pneumonia in Uganda. The standard clinical criteria used to identify pneumonia among children in resource-poor settings were sensitive but lacked specificity. Chest radiographs should be performed routinely for all children with clinical signs of severe pneumonia because it provides useful information on both cardiovascular and respiratory systems.

Classification of COVID-19 from community-acquired pneumonia: Boosting the performance with capsule network and maximum intensity projection image of CT scans

BACKGROUND

The coronavirus disease 2019 (COVID-19) and community-acquired pneumonia (CAP) present a high degree of similarity in chest computed tomography (CT) images. Therefore, a procedure for accurately and automatically distinguishing between them is crucial.

METHODS

A deep learning method for distinguishing COVID-19 from CAP is developed using maximum intensity projection (MIP) images from CT scans. LinkNet is employed for lung segmentation of chest CT images. MIP images are produced by superposing the maximum gray of intrapulmonary CT values. The MIP images are input into a capsule network for patient-level pred iction and diagnosis of COVID-19. The network is trained using 333 CT scans (168 COVID-19/165 CAP) and validated on three external datasets containing 3581 CT scans (2110 COVID-19/1471 CAP).

RESULTS

LinkNet achieves the highest Dice coefficient of 0.983 for lung segmentation. For the classification of COVID-19 and CAP, the capsule network with the DenseNet-121 feature extractor outperforms ResNet-50 and Inception-V3, achieving an accuracy of 0.970 on the training dataset. Without MIP or the capsule network, the accuracy decreases to 0.857 and 0.818, respectively. Accuracy scores of 0.961, 0.997, and 0.949 are achieved on the external validation datasets. The proposed method has higher or comparable sensitivity compared with ten state-of-the-art methods.

CONCLUSIONS

The proposed method illustrates the feasibility of applying MIP images from CT scans to distinguish COVID-19 from CAP using capsule networks. MIP images provide conspicuous benefits when exploiting deep learning to detect COVID-19 lesions from CT scans and the capsule network improves COVID-19 diagnosis.

Nomogram Using CT Radiomics Features for Differentiation of Pneumonia-Type Invasive Mucinous Adenocarcinoma and Pneumonia: Multicenter Development and External Validation Study

BACKGROUND. Pneumonia-type invasive mucinous adenocarcinoma (IMA) and pneumonia show overlapping chest CT features as well as overlapping clinical characteristics. OBJECTIVE. The purpose of our study was to develop and validate a nomogram combining clinical and CT-based radiomics features to differentiate pneumonia-type IMA and pneumonia. METHODS. This retrospective study included 314 patients (172 men, 142 women; mean age, 60.3 ± 14.5 [SD] years) from six hospitals who underwent noncontrast chest CT showing consolidation and were diagnosed with pneumonia-type IMA (n = 106) or pneumonia (n = 208). Patients from three hospitals formed a training set (n = 195) and a validation set (n = 50), and patients from the other three hospitals formed the external test set (n = 69). A model for predicting pneumonia-type IMA was built using clinical characteristics that were significant independent predictors of this diagnosis. Radiomics features were extracted from CT images by placing ROIs on areas of consolidation, and a radiomics signature of pneumonia-type IMA was constructed. A nomogram for predicting pneumonia-type IMA was constructed that combined features in the clinical model and the radiomics signature. Two cardiothoracic radiologists independently reviewed CT images in the external test set to diagnose pneumonia-type IMA. Diagnostic performance was compared among models and radiologists. Decision curve analysis (DCA) was performed. RESULTS. The clinical model included fever and family history of lung cancer. The radiomics signature included 15 radiomics features. DCA showed higher overall net benefit from the nomogram than from the clinical model. In the external test set, AUC was higher for the nomogram (0.85) than for the clinical model (0.71, p = .01), radiologist 1 (0.70, p = .04), and radiologist 2 (0.67, p = .01). In the external test set, the nomogram had sensitivity of 46.9%, specificity of 94.6%, and accuracy of 72.5%. CONCLUSION. The nomogram combining clinical variables and CT-based radiomics features outperformed the clinical model and two cardiothoracic radiologists in differentiating pneumonia-type IMA from pneumonia. CLINICAL IMPACT. The findings support potential clinical use of the nomogram for diagnosing pneumonia-type IMA in patients with consolidation on chest CT.

LDDNet: A Deep Learning Framework for the Diagnosis of Infectious Lung Diseases

This paper proposes a new deep learning (DL) framework for the analysis of lung diseases, including COVID-19 and pneumonia, from chest CT scans and X-ray (CXR) images. This framework is termed optimized DenseNet201 for lung diseases (LDDNet). The proposed LDDNet was developed using additional layers of 2D global average pooling, dense and dropout layers, and batch normalization to the base DenseNet201 model. There are 1024 Relu-activated dense layers and 256 dense layers using the sigmoid activation method. The hyper-parameters of the model, including the learning rate, batch size, epochs, and dropout rate, were tuned for the model. Next, three datasets of lung diseases were formed from separate open-access sources. One was a CT scan dataset containing 1043 images. Two X-ray datasets comprising images of COVID-19-affected lungs, pneumonia-affected lungs, and healthy lungs exist, with one being an imbalanced dataset with 5935 images and the other being a balanced dataset with 5002 images. The performance of each model was analyzed using the Adam, Nadam, and SGD optimizers. The best results have been obtained for both the CT scan and CXR datasets using the Nadam optimizer. For the CT scan images, LDDNet showed a COVID-19-positive classification accuracy of 99.36%, a 100% precision recall of 98%, and an F1 score of 99%. For the X-ray dataset of 5935 images, LDDNet provides a 99.55% accuracy, 73% recall, 100% precision, and 85% F1 score using the Nadam optimizer in detecting COVID-19-affected patients. For the balanced X-ray dataset, LDDNet provides a 97.07% classification accuracy. For a given set of parameters, the performance results of LDDNet are better than the existing algorithms of ResNet152V2 and XceptionNet.

Relationship between lung ultrasound and electrical impedance tomography as regional assessment tools during PEEP titration in acute respiratory distress syndrome caused by multi-lobar pneumonia: a pilot study

Acute respiratory distress syndrome (ARDS) caused by multilobar pneumonia (MLP) is markedly different from typical ARDS in pathology, imaging characteristics, and lung mechanics. Regional lung assessment is required. We aimed to analyze the relationship between two regional assessment tools, lung ultrasound (LUS) and electrical impedance tomography (EIT) during positive end-expiratory pressure (PEEP) titration, and determine an appropriate PEEP level. We conducted a prospective study of patients with ARDS caused by MLP with PaO₂/FiO₂ < 150 mmHg. All subjects were equipped with two EIT belts connected with a single EIT machine to measure upper and lower hemithorax impedance change alternatingly at each PEEP level. LUS score was simultaneously determined in chest wall regions corresponding to the EIT regions during PEEP titration. We acquired EIT and LUS data in eight regions of interest at seven PEEP levels in 12 subjects. Therefore, 672 pairs of data were obtained for analysis. There were significant relationships between LUS score and tidal impedance variation and pixel compliance (C_pix). The Spearman's rho between LUS score vs. tidal impedance variation and LUS score vs. the C_pix were - 0.142, P < 0.001, and - 0.195, P < 0.001, respectively. The relationship between the LUS score and C_pix remained the same at every PEEP level but did not reach statistical significance. The individual's mean expected PEEP by LUS was similar to the EIT [10.33(± 1.67) vs. 10.33(± 1.44) cm H₂O, P = 0.15]. Regarding the MLP, the LUS scores were associated with EIT parameters, and LUS scores might proof helpful for finding individual PEEP settings in MLP.

Early chest CT-scan in emergency patients affected by community-acquired pneumonia is associated with improved diagnosis consistency

Chest CT-scan (CT) exceeds chest X-ray (CXR) to diagnose community-acquired pneumonia (CAP) but actual use and results remain unclear. We examine whether CT performed at ED visit improved ED diagnosis of CAP as compared to a final diagnosis of CAP at hospital discharge (gold standard diagnosis for the study), and how it impacts relevant clinical outcomes. This retrospective monocenter observational study was based on the analysis of the hospital database. Patients with a diagnosis of CAP in the ED (ICD-10 codes: J110, J111, from J12- to J18-, J440, J690, U0710, and U0711) were included. We compared ED patients who were diagnosed with CAP using CXR and CT. We measured diagnostic consistency, duration of ED visit, percentage of CXR and CT during hospital stay, hospital length-of-stay, ICU admission, and in-hospital mortality. Multivariate analysis was adjusted for CRB65 score by multiple logistic regression analysis for binary outcomes and by multivariate analysis of variance for continuous outcomes. We included 994 ED patients with an initial diagnosis of CAP (751 receiving CXR, 243 receiving CT). CT prescription in the ED increased over time ( P < 0.001). In patients admitted after ED, CT improved diagnosis consistency for CAP [88.2% vs. 80.9%; difference 7.3% (95% confidence interval 1.2-13.3%)] with a trend for lower hospital length-of-stay [10.2 vs. 12.2 days; difference -2.0 (95% confidence interval -3.9 to -0.1)], but not ICU admission ( P = 0.09) and in-hospital mortality ( P = 0.056). Diagnosis of patients admitted with CAP improved when CT was obtained at ED visit. These results should be reproduced at a larger scale to test whether early CT conserves healthcare resources.

Artificial Intelligence-Augmented Pediatric Lung POCUS: A Pilot Study of Novice Learners

OBJECTIVE

Respiratory symptoms are among the most common chief complaints of pediatric patients in the emergency department (ED). Point-of-care ultrasound (POCUS) outperforms conventional chest X-ray and is user-dependent, which can be challenging to novice ultrasound (US) users. We introduce a novel concept using artificial intelligence (AI)-enhanced pleural sweep to generate complete panoramic views of the lungs, and then assess its accuracy among novice learners (NLs) to identify pneumonia.

METHODS

Previously healthy 0- to 17-year-old patients presenting to a pediatric ED with cardiopulmonary chief complaint were recruited. NLs received a 1-hour training on traditional lung POCUS and the AI-assisted software. Two POCUS-trained experts interpreted the images, which served as the criterion standard. Both expert and learner groups were blinded to each other's interpretation, patient data, and outcomes. Kappa was used to determine agreement between POCUS expert interpretations.

RESULTS

Seven NLs, with limited to no prior POCUS experience, completed examinations on 32 patients. The average patient age was 5.53 years (±1.07). The median scan time of 7 minutes (minimum-maximum 3-43; interquartile 8). Three (8.8%) patients were diagnosed with pneumonia by criterion standard. Sensitivity, specificity, and accuracy for NLs AI-augmented interpretation were 66.7% (confidence interval [CI] 9.4-99.1%), 96.5% (CI 82.2-99.9%), and 93.7% (CI 79.1-99.2%). The average image quality rating was 2.94 (±0.16) out of 5 across all lung fields. Interrater reliability between expert sonographers was high with a kappa coefficient of 0.8.

CONCLUSION

This study shows that AI-augmented lung US for diagnosing pneumonia has the potential to increase accuracy and efficiency.

Follow-up Study of Pulmonary Function, Exercise Capacity and Radiological Changes after Recovery from Moderate to Severe COVID Pneumonia without Mechanical Ventilation

BACKGROUND

The long-term effects of COVID on the lungs remain unclear, but, given the extent of the pandemic, it has the potential to become a significant chronic global health problem .Aim of our study was to ascertain the proportion of patients with moderate to severe pneumonia but without mechanical ventilation who have compromised exercise capacity, pulmonary function test and presence of radiological abnormalities and to study any correlation between clinical features with radiological abnormalities.

METHODS

In a hospital-based study, COVID-19 patients with moderate and severe pneumonia were followed 3 months after discharge and assessed with chest computed tomography (CT) imaging, 6 minute walk test and pulmonary function tests.

RESULTS

A total of 102 participants were enrolled, including64 patients who had recovered from moderate disease and 38 patients from severe COVID-19. The patients with critical disease and who required mechanical ventilation or who had previously known chronic lung disease were excluded. High proportion of patients of both groups showed radiological abnormalities and deranged pulmonary function tests 3 months after recovery from acute illness which had significant correlation with severity of disease.

CONCLUSIONS

Pulmonary function and radiological abnormalities remained in significant propotion of patients 3 months after recovery from COVID-19 that needs more attention on pulmonary rehabilitation and long term follow up of these patients.

IEViT: An enhanced vision transformer architecture for chest X-ray image classification

BACKGROUND AND OBJECTIVE

Chest X-ray imaging is a relatively cheap and accessible diagnostic tool that can assist in the diagnosis of various conditions, including pneumonia, tuberculosis, COVID-19, and others. However, the requirement for expert radiologists to view and interpret chest X-ray images can be a bottleneck, especially in remote and deprived areas. Recent advances in machine learning have made possible the automated diagnosis of chest X-ray scans. In this work, we examine the use of a novel Transformer-based deep learning model for the task of chest X-ray image classification.

METHODS

We first examine the performance of the Vision Transformer (ViT) state-of-the-art image classification machine learning model for the task of chest X-ray image classification, and then propose and evaluate the Input Enhanced Vision Transformer (IEViT), a novel enhanced Vision Transformer model that can achieve improved performance on chest X-ray images associated with various pathologies.

RESULTS

Experiments on four chest X-ray image data sets containing various pathologies (tuberculosis, pneumonia, COVID-19) demonstrated that the proposed IEViT model outperformed ViT for all the data sets and variants examined, achieving an F1-score between 96.39% and 100%, and an improvement over ViT of up to +5.82% in terms of F1-score across the four examined data sets. IEViT's maximum sensitivity (recall) ranged between 93.50% and 100% across the four data sets, with an improvement over ViT of up to +3%, whereas IEViT's maximum precision ranged between 97.96% and 100% across the four data sets, with an improvement over ViT of up to +6.41%.

CONCLUSIONS

Results showed that the proposed IEViT model outperformed all ViT's variants for all the examined chest X-ray image data sets, demonstrating its superiority and generalisation ability. Given the relatively low cost and the widespread accessibility of chest X-ray imaging, the use of the proposed IEViT model can potentially offer a powerful, but relatively cheap and accessible method for assisting diagnosis using chest X-ray images.

Two-step machine learning to diagnose and predict involvement of lungs in COVID-19 and pneumonia using CT radiomics

OBJECTIVE

To develop a two-step machine learning (ML) based model to diagnose and predict involvement of lungs in COVID-19 and non COVID-19 pneumonia patients using CT chest radiomic features.

METHODS

Three hundred CT scans (3-classes: 100 COVID-19, 100 pneumonia, and 100 healthy subjects) were enrolled in this study. Diagnostic task included 3-class classification. Severity prediction score for COVID-19 and pneumonia was considered as mild (0-25%), moderate (26-50%), and severe (>50%). Whole lungs were segmented utilizing deep learning-based segmentation. Altogether, 107 features including shape, first-order histogram, second and high order texture features were extracted. Pearson correlation coefficient (PCC≥90%) followed by different features selection algorithms were employed. ML-based supervised algorithms (Naïve Bays, Support Vector Machine, Bagging, Random Forest, K-nearest neighbors, Decision Tree and Ensemble Meta voting) were utilized. The optimal model was selected based on precision, recall and area-under-curve (AUC) by randomizing the training/validation, followed by testing using the test set.

RESULTS

Nine pertinent features (2 shape, 1 first-order, and 6 second-order) were obtained after features selection for both phases. In diagnostic task, the performance of 3-class classification using Random Forest was 0.909±0.026, 0.907±0.056, 0.902±0.044, 0.939±0.031, and 0.982±0.010 for precision, recall, F1-score, accuracy, and AUC, respectively. The severity prediction task using Random Forest achieved 0.868±0.123 precision, 0.865±0.121 recall, 0.853±0.139 F1-score, 0.934±0.024 accuracy, and 0.969±0.022 AUC.

CONCLUSION

The two-phase ML-based model accurately classified COVID-19 and pneumonia patients using CT radiomics, and adequately predicted severity of lungs involvement. This 2-steps model showed great potential in assessing COVID-19 CT images towards improved management of patients.

CX-DaGAN: Domain Adaptation for Pneumonia Diagnosis on a Small Chest X-Ray Dataset

Recent advances in deep learning led to several algorithms for the accurate diagnosis of pneumonia from chest X-rays. However, these models require large training medical datasets, which are sparse, isolated, and generally private. Furthermore, these models in medical imaging are known to over-fit to a particular data domain source, i.e., these algorithms do not conserve the same accuracy when tested on a dataset from another medical center, mainly due to image distribution discrepancies. In this work, a domain adaptation and classification technique is proposed to overcome the over-fit challenges on a small dataset. This method uses a private-small dataset (target domain), a public-large labeled dataset from another medical center (source domain), and consists of three steps. First, it performs a data selection of the source domain's most representative images based on similarity constraints through principal component analysis subspaces. Second, the selected samples from the source domain are fit to the target distribution through an image to image translation based on a cycle-generative adversarial network. Finally, the target train dataset and the adapted images from the source dataset are used within a convolutional neural network to explore different settings to adjust the layers and perform the classification of the target test dataset. It is shown that fine-tuning a few specific layers together with the selected-adapted images increases the sorting accuracy while reducing the trainable parameters. The proposed approach achieved a notable increase in the target dataset's overall classification accuracy, reaching up to 97.78 % compared to 90.03 % by standard transfer learning.

Explanatory classification of CXR images into COVID-19, Pneumonia and Tuberculosis using deep learning and XAI

Chest X-ray (CXR) images are considered useful to monitor and investigate a variety of pulmonary disorders such as COVID-19, Pneumonia, and Tuberculosis (TB). With recent technological advancements, such diseases may now be recognized more precisely using computer-assisted diagnostics. Without compromising the classification accuracy and better feature extraction, deep learning (DL) model to predict four different categories is proposed in this study. The proposed model is validated with publicly available datasets of 7132 chest x-ray (CXR) images. Furthermore, results are interpreted and explained using Gradient-weighted Class Activation Mapping (Grad-CAM), Local Interpretable Modelagnostic Explanation (LIME), and SHapley Additive exPlanation (SHAP) for better understandably. Initially, convolution features are extracted to collect high-level object-based information. Next, shapely values from SHAP, predictability results from LIME, and heatmap from Grad-CAM are used to explore the black-box approach of the DL model, achieving average test accuracy of 94.31 ± 1.01% and validation accuracy of 94.54 ± 1.33 for 10-fold cross validation. Finally, in order to validate the model and qualify medical risk, medical sensations of classification are taken to consolidate the explanations generated from the eXplainable Artificial Intelligence (XAI) framework. The results suggest that XAI and DL models give clinicians/medical professionals persuasive and coherent conclusions related to the detection and categorization of COVID-19, Pneumonia, and TB.

Machine Learning Model Based on Radiomic Features for Differentiation between COVID-19 and Pneumonia on Chest X-ray

Machine learning approaches are employed to analyze differences in real-time reverse transcription polymerase chain reaction scans to differentiate between COVID-19 and pneumonia. However, these methods suffer from large training data requirements, unreliable images, and uncertain clinical diagnosis. Thus, in this paper, we used a machine learning model to differentiate between COVID-19 and pneumonia via radiomic features using a bias-minimized dataset of chest X-ray scans. We used logistic regression (LR), naive Bayes (NB), support vector machine (SVM), k-nearest neighbor (KNN), bagging, random forest (RF), extreme gradient boosting (XGB), and light gradient boosting machine (LGBM) to differentiate between COVID-19 and pneumonia based on training data. Further, we used a grid search to determine optimal hyperparameters for each machine learning model and 5-fold cross-validation to prevent overfitting. The identification performances of COVID-19 and pneumonia were compared with separately constructed test data for four machine learning models trained using the maximum probability, contrast, and difference variance of the gray level co-occurrence matrix (GLCM), and the skewness as input variables. The LGBM and bagging model showed the highest and lowest performances; the GLCM difference variance showed a high overall effect in all models. Thus, we confirmed that the radiomic features in chest X-rays can be used as indicators to differentiate between COVID-19 and pneumonia using machine learning.

Association Between Lung Ultrasound Patterns and Pneumonia

ABSTRACT

Pneumonia is a common respiratory infection that affects the lungs. Lung ultrasound (LUS) is a portable, cost-effective imaging method, which is free of ionizing radiation and has been shown to be useful for evaluating pneumonia. The aim of this retrospective analytical study was to determine the association between lung ultrasound patterns and pneumonia. For the purpose of performing the required analysis, LUS patterns including consolidations, pleural line irregularities, A lines and B lines from 90 subjects (44 patients with confirmed pneumonia and 46 controls) were retrieved from a published open-access data set, which was reviewed and approved by medical experts. A χ 2 test was used for the comparison of categorical variables to determine the association between each LUS pattern and the presence of pneumonia. There is a significant association between LUS consolidation and the presence of pneumonia ( P < 0.0001). Lung ultrasound A lines are significantly associated with the absence of pneumonia ( P < 0.0001), whereas there are no associations between B lines or pleural line irregularities with pneumonia. Lung ultrasound consolidation is found to be associated with the presence of pneumonia. A lines are associated with healthy lungs, and there is no association of B lines and pleural irregularities with the presence of pneumonia. Further studies investigating LUS patterns with clinical information and symptoms of patients with pneumonia are required.

[Chest x-rays do not aid in antibiotic treatment decisions for children with lower respiratory infections in the Emergency Department]

The chest x-ray (CXR) was the gold standard in the diagnosis of pneumonia in children. However, CXR has limitations and cannot discriminate in etiology. Current guidelines recommend against routine use of CXR in children with uncomplicated lower respiratory tract infections (LRTI). We used routine care data from a multicentre RCT including 597 children with LRTI symptoms, to evaluate the influence of CXR on antibiotic prescription in the emergency department (ED). CXR remains frequently performed in non-complex children suspected of LRTI in the ED (18%). Children who underwent CXR were more likely to receive antibiotics, even when adjusted for symptoms, hospital and CXR results. Our study highlights the inferior role of CXR in treatment decisions for children with LRTI as CXR, regardless of its results, is independently associated with more antibiotic prescriptions.

Radiological patterns and prognosis in elderly patients with acute Klebsiella pneumoniae pneumonia: A retrospective study

Although Klebsiella pneumoniae pneumonia is an insidious threat among the elderly, the role of radiological features has not been elucidated. We aimed to evaluate thin-section chest computed tomography (CT) features and assess its associations with disease prognosis in elderly patients with acute K. pneumoniae pneumonia. We retrospectively included elderly patients, admitted for acute K. pneumoniae pneumonia, and investigated thin-section CT findings to determine whether bronchopneumonia or lobar pneumonia was present. The association between the radiological pattern of pneumonia and in-hospital mortality was analyzed. Eighty-six patients with acute K. pneumoniae pneumonia were included, and among them, the bronchopneumonia pattern was observed in 70 (81%) patients. Twenty-five (29%) patients died in hospital, and they had a greater incidence of lobar pneumonia pattern (40% in nonsurvivors vs 10% in survivors; P = .008), low albumin level (2.7 g/dL, range, 1.6-3.8 in nonsurvivors vs 3.0 g/dL, range, 1.7-4.2 in survivors; P = .026) and higher levels of aspartate aminotransferase (30 U/L, range, 11-186 in nonsurvivors vs 23 U/L, range, 11-102 in survivors, P = .017) and C-reactive protein (8.0 mg/dL, range, 0.9-26.5 in nonsurvivors vs 4.7 mg/dL, range, 0.0-24.0 in survivors; P = .047) on admission. Multivariate analysis showed that lobar pneumonia pattern was independently associated with increased in-hospital mortality (adjusted hazard ratio, 3.906; 95% CI, 1.513-10.079; P = .005). In elderly patients with acute K. pneumoniae pneumonia, the lobar pneumonia pattern may be less commonly observed, and this pattern could relate to poor prognosis.

Association of Artificial Intelligence-Aided Chest Radiograph Interpretation With Reader Performance and Efficiency

IMPORTANCE

The efficient and accurate interpretation of radiologic images is paramount.

OBJECTIVE

To evaluate whether a deep learning-based artificial intelligence (AI) engine used concurrently can improve reader performance and efficiency in interpreting chest radiograph abnormalities.

DESIGN, SETTING, AND PARTICIPANTS

This multicenter cohort study was conducted from April to November 2021 and involved radiologists, including attending radiologists, thoracic radiology fellows, and residents, who independently participated in 2 observer performance test sessions. The sessions included a reading session with AI and a session without AI, in a randomized crossover manner with a 4-week washout period in between. The AI produced a heat map and the image-level probability of the presence of the referrable lesion. The data used were collected at 2 quaternary academic hospitals in Boston, Massachusetts: Beth Israel Deaconess Medical Center (The Medical Information Mart for Intensive Care Chest X-Ray [MIMIC-CXR]) and Massachusetts General Hospital (MGH).

MAIN OUTCOMES AND MEASURES

The ground truths for the labels were created via consensual reading by 2 thoracic radiologists. Each reader documented their findings in a customized report template, in which the 4 target chest radiograph findings and the reader confidence of the presence of each finding was recorded. The time taken for reporting each chest radiograph was also recorded. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUROC) were calculated for each target finding.

RESULTS

A total of 6 radiologists (2 attending radiologists, 2 thoracic radiology fellows, and 2 residents) participated in the study. The study involved a total of 497 frontal chest radiographs-247 from the MIMIC-CXR data set (demographic data for patients were not available) and 250 chest radiographs from MGH (mean [SD] age, 63 [16] years; 133 men [53.2%])-from adult patients with and without 4 target findings (pneumonia, nodule, pneumothorax, and pleural effusion). The target findings were found in 351 of 497 chest radiographs. The AI was associated with higher sensitivity for all findings compared with the readers (nodule, 0.816 [95% CI, 0.732-0.882] vs 0.567 [95% CI, 0.524-0.611]; pneumonia, 0.887 [95% CI, 0.834-0.928] vs 0.673 [95% CI, 0.632-0.714]; pleural effusion, 0.872 [95% CI, 0.808-0.921] vs 0.889 [95% CI, 0.862-0.917]; pneumothorax, 0.988 [95% CI, 0.932-1.000] vs 0.792 [95% CI, 0.756-0.827]). AI-aided interpretation was associated with significantly improved reader sensitivities for all target findings, without negative impacts on the specificity. Overall, the AUROCs of readers improved for all 4 target findings, with significant improvements in detection of pneumothorax and nodule. The reporting time with AI was 10% lower than without AI (40.8 vs 36.9 seconds; difference, 3.9 seconds; 95% CI, 2.9-5.2 seconds; P < .001).

CONCLUSIONS AND RELEVANCE

These findings suggest that AI-aided interpretation was associated with improved reader performance and efficiency for identifying major thoracic findings on a chest radiograph.