Severity prediction of pulmonary diseases using chest CT scans via cost-sensitive label multi-kernel distribution learning.

BACKGROUND AND OBJECTIVES: The progression of pulmonary diseases is a complex progress. Timely predicting whether the patients will progress to the severe stage or not in its early stage is critical to take appropriate hospital treatment. However, this task suffers from the "insufficient and incomplete" data issue since it is clinically impossible to have adequate training samples for one patient at each day. Besides, the training samples are extremely imbalanced since the patients who will progress to the severe stage is far less than those who will not progress to the non-severe stage. METHOD: We consider the severity prediction of pulmonary diseases as a time estimation problem based on CT scans. To handle the issue of "insufficient and incomplete" training samples, we introduced label distribution learning (LDL). Specifically, we generate a label distribution for each patient, making a CT image contribute to not only the learning of its chronological day, but also the learning of its neighboring days. In addition, a cost-sensitive mechanism is introduced to explore the imbalance data issue. To identify the importance of pulmonary segments in pulmonary disease severity prediction, multi-kernel learning in composite kernel space is further incorporated and particle swarm optimization (PSO) is used to find the optimal kernel weights. RESULTS: We compare the performance of the proposed CS-LD-MKSVR algorithm with several classical machine learning algorithms and deep learning (DL) algorithms. The proposed method has obtained the best classification results on the in-house data, fully indicating its effectiveness in pulmonary disease severity prediction. CONTRIBUTIONS: The severity prediction of pulmonary diseases is considered as a time estimation problem, and label distribution is introduced to describe the conversion time from non-severe stage to severe stage. The cost-sensitive mechanism is also introduced to handle the data imbalance issue to further improve the classification performance.

[Follow-up of patients after COVID-19 pneumonia. Pulmonary sequelae]. / Seguimiento de los pacientes después de neumonía por COVID-19. Secuelas pulmonares.

Coronavirus disease 2019 (COVID-19) is an infection caused by SARS-CoV-2 that has caused an unprecedented pandemic with a high rate of morbidity and mortality worldwide. Although most cases are mild, there are a considerable number of patients who develop pneumonia or even acute respiratory distress syndrome (ARDS). After having recovered from the initial disease, many patients continue with various symptoms (fatigue, dry cough, fever, dyspnea, anosmia, and chest pain, among others.), which has led to consider the possible existence of "post-COVID-19 syndrome". Although the definition and validity of this syndrome are not clear yet, several studies report that individuals who have recovered from COVID-19 may have persistent symptoms, radiological abnormalities, and compromised respiratory function. Current evidence suggests that there is a large number of pulmonary sequelae after COVID-19 pneumonia (interstitial thickening, ground glass opacities, crazy paving pattern, and bronchiectasis, among others.). Likewise, it seems that pulmonary function tests (spirometry, DLCO, 6MWT, and measurement of maximum respiratory pressures), in addition to high-resolution computed axial tomographies (CAT scan), are useful for the assessment of these post-COVID-19 pulmonary sequelae. This review aims to describe the possible pulmonary sequelae after COVID-19 pneumonia, as well as to suggest diagnostic procedures for their correct assessment and follow-up; thus, allowing proper management by a multidisciplinary medical team.

COVID-19 es la enfermedad causada por el virus SARS-CoV-2, la cual ha ocasionado una pandemia sin precedentes, con gran cantidad de infectados y muertos en el mundo. Aunque la mayoría de los casos son leves, existe una cantidad considerable de pacientes que desarrollan neumonía o, incluso, síndrome de distrés respiratorio agudo (SDRA). Luego de recuperarse del cuadro inicial, muchos pacientes continúan con diversos síntomas (fatiga, tos seca, fiebre, disnea, anosmia, dolor torácico, entre otras), lo que ha llevado a considerar la posible existencia del "síndrome pos-COVID-19". Aunque la definición y validez de este síndrome aún no son claras, varios estudios reportan que los individuos recuperados de la COVID-19 pueden tener persistencia de síntomas, anormalidades radiológicas y compromiso en la función respiratoria. La evidencia actual sugiere que existe gran cantidad de secuelas pulmonares despues de una neumonía por COVID-19 (engrosamiento intersticial, infiltrado en vidrio esmerilado, patrón en empedrado, bronquiectasias, entre otras.). De igual forma, parece ser que las pruebas de función pulmonar (espirometría, prueba de difusión pulmonar de monóxido de carbono, prueba de caminata de seis minutos y la medición de las presiones respiratorias máximas), además de la tomografía axial computarizada de alta resolución, son útiles para evaluar las secuelas pulmonares pos-COVID-19. En esta revisión se pretende describir las posibles secuelas a nivel pulmonar posteriores a neumonía por COVID-19, así como sugerir procedimientos diagnósticos para su correcta evaluación y seguimiento, que permitan el manejo adecuado por parte de un equipo médico multidisciplinario.

Development of Hands-On Skills in Diagnostics of Lung Diseases Using Ultrasonography in Undergraduate Medical Education.

This study evaluated advanced pulmonary ultrasonography training for COVID-19 lung examination. Students completed identical pretests and post-tests and a survey. Changes were found for individual questions and overall scores (all P≤.02), specifically image identification, previous material, and COVID-19 questions. Students were receptive to the training for education and future practice (P<.001), and they felt capable using ultrasound for diagnosis and management of COVID-19 patients. Pulmonary ultrasonography training should be considered for the medical school curriculum.

High-Resolution Computed Tomography of Cystic Lung Disease.

The cystic lung diseases (CLD) are characterized by the presence of multiple, thin-walled, air-filled spaces in the pulmonary parenchyma. Cyst formation may occur with congenital, autoimmune, inflammatory, infectious, or neoplastic processes. Recognition of cyst mimics such as emphysema and bronchiectasis is important to prevent diagnostic confusion and unnecessary evaluation. Chest CT can be diagnostic or may guide the workup based on cyst number, distribution, morphology, and associated lung, and extrapulmonary findings. Diffuse CLD (DCLDs) are often considered those presenting with 10 or more cysts. The more commonly encountered DCLDs include lymphangioleiomyomatosis, pulmonary Langerhans' cell histiocytosis, lymphoid interstitial pneumonia, Birt-Hogg-Dubé syndrome, and amyloidosis/light chain deposition disease.

MR Imaging for the Evaluation of Diffuse Lung Disease: Where Are We?

Patients with diffuse lung diseases require thorough medical and social history and physical examinations, coupled with a multitude of laboratory tests, pulmonary function tests, and radiologic imaging to discern and manage the specific disease. This review summarizes the current state of imaging of various diffuse lung diseases by hyperpolarized MR imaging. The potential of hyperpolarized MR imaging as a clinical tool is outlined as a novel imaging approach that enables further understanding of the cause of diffuse lung diseases, permits earlier detection of disease progression before that found with pulmonary function tests, and can delineate physiologic response to lung therapies.

Three-stage segmentation of lung region from CT images using deep neural networks.

BACKGROUND: Lung region segmentation is an important stage of automated image-based approaches for the diagnosis of respiratory diseases. Manual methods executed by experts are considered the gold standard, but it is time consuming and the accuracy is dependent on radiologists' experience. Automated methods are relatively fast and reproducible with potential to facilitate physician interpretation of images. However, these benefits are possible only after overcoming several challenges. The traditional methods that are formulated as a three-stage segmentation demonstrate promising results on normal CT data but perform poorly in the presence of pathological features and variations in image quality attributes. The implementation of deep learning methods that can demonstrate superior performance over traditional methods is dependent on the quantity, quality, cost and the time it takes to generate training data. Thus, efficient and clinically relevant automated segmentation method is desired for the diagnosis of respiratory diseases. METHODS: We implement each of the three stages of traditional methods using deep learning methods trained on five different configurations of training data with ground truths obtained from the 3D Image Reconstruction for Comparison of Algorithm Database (3DIRCAD) and the Interstitial Lung Diseases (ILD) database. The data was augmented with the Lung Image Database Consortium (LIDC-IDRI) image collection and a realistic phantom. A convolutional neural network (CNN) at the preprocessing stage classifies the input into lung and none lung regions. The processing stage was implemented using a CNN-based U-net while the postprocessing stage utilize another U-net and CNN for contour refinement and filtering out false positives, respectively. RESULTS: The performance of the proposed method was evaluated on 1230 and 1100 CT slices from the 3DIRCAD and ILD databases. We investigate the performance of the proposed method on five configurations of training data and three configurations of the segmentation system; three-stage segmentation and three-stage segmentation without a CNN classifier and contrast enhancement, respectively. The Dice-score recorded by the proposed method range from 0.76 to 0.95. CONCLUSION: The clinical relevance and segmentation accuracy of deep learning models can improve though deep learning-based three-stage segmentation, image quality evaluation and enhancement as well as augmenting the training data with large volume of cheap and quality training data. We propose a new and novel deep learning-based method of contour refinement.

DeBoNet: A deep bone suppression model ensemble to improve disease detection in chest radiographs.

Automatic detection of some pulmonary abnormalities using chest X-rays may be impacted adversely due to obscuring by bony structures like the ribs and the clavicles. Automated bone suppression methods would increase soft tissue visibility and enhance automated disease detection. We evaluate this hypothesis using a custom ensemble of convolutional neural network models, which we call DeBoNet, that suppresses bones in frontal CXRs. First, we train and evaluate variants of U-Nets, Feature Pyramid Networks, and other proposed custom models using a private collection of CXR images and their bone-suppressed counterparts. The DeBoNet, constructed using the top-3 performing models, outperformed the individual models in terms of peak signal-to-noise ratio (PSNR) (36.7977±1.6207), multi-scale structural similarity index measure (MS-SSIM) (0.9848±0.0073), and other metrics. Next, the best-performing bone-suppression model is applied to CXR images that are pooled from several sources, showing no abnormality and other findings consistent with COVID-19. The impact of bone suppression is demonstrated by evaluating the gain in performance in detecting pulmonary abnormality consistent with COVID-19 disease. We observe that the model trained on bone-suppressed CXRs (MCC: 0.9645, 95% confidence interval (0.9510, 0.9780)) significantly outperformed (p < 0.05) the model trained on non-bone-suppressed images (MCC: 0.7961, 95% confidence interval (0.7667, 0.8255)) in detecting findings consistent with COVID-19 indicating benefits derived from automatic bone suppression on disease classification. The code is available at https://github.com/sivaramakrishnan-rajaraman/Bone-Suppresion-Ensemble.

Lung Disease Classification in CXR Images Using Hybrid Inception-ResNet-v2 Model and Edge Computing.

Chest X-ray (CXR) imaging is one of the most widely used and economical tests to diagnose a wide range of diseases. However, even for expert radiologists, it is a challenge to accurately diagnose diseases from CXR samples. Furthermore, there remains an acute shortage of trained radiologists worldwide. In the present study, a range of machine learning (ML), deep learning (DL), and transfer learning (TL) approaches have been evaluated to classify diseases in an openly available CXR image dataset. A combination of the synthetic minority over-sampling technique (SMOTE) and weighted class balancing is used to alleviate the effects of class imbalance. A hybrid Inception-ResNet-v2 transfer learning model coupled with data augmentation and image enhancement gives the best accuracy. The model is deployed in an edge environment using Amazon IoT Core to automate the task of disease detection in CXR images with three categories, namely pneumonia, COVID-19, and normal. Comparative analysis has been given in various metrics such as precision, recall, accuracy, AUC-ROC score, etc. The proposed technique gives an average accuracy of 98.66%. The accuracies of other TL models, namely SqueezeNet, VGG19, ResNet50, and MobileNetV2 are 97.33%, 91.66%, 90.33%, and 76.00%, respectively. Further, a DL model, trained from scratch, gives an accuracy of 92.43%. Two feature-based ML classification techniques, namely support vector machine with local binary pattern (SVM + LBP) and decision tree with histogram of oriented gradients (DT + HOG) yield an accuracy of 87.98% and 86.87%, respectively.

Quantitative Chest CT Assessment of Small Airways Disease in Post-Acute SARS-CoV-2 Infection.

Background The long-term effects of SARS-CoV-2 infection on pulmonary structure and function remain incompletely characterized. Purpose To test whether SARS-CoV-2 infection leads to small airways disease in patients with persistent symptoms. Materials and Methods In this single-center study at a university teaching hospital, adults with confirmed COVID-19 who remained symptomatic more than 30 days following diagnosis were prospectively enrolled from June to December 2020 and compared with healthy participants (controls) prospectively enrolled from March to August 2018. Participants with post-acute sequelae of COVID-19 (PASC) were classified as ambulatory, hospitalized, or having required the intensive care unit (ICU) based on the highest level of care received during acute infection. Symptoms, pulmonary function tests, and chest CT images were collected. Quantitative CT analysis was performed using supervised machine learning to measure regional ground-glass opacity (GGO) and using inspiratory and expiratory image-matching to measure regional air trapping. Univariable analyses and multivariable linear regression were used to compare groups. Results Overall, 100 participants with PASC (median age, 48 years; 66 women) were evaluated and compared with 106 matched healthy controls; 67% (67 of 100) of the participants with PASC were classified as ambulatory, 17% (17 of 100) were hospitalized, and 16% (16 of 100) required the ICU. In the hospitalized and ICU groups, the mean percentage of total lung classified as GGO was 13.2% and 28.7%, respectively, and was higher than that in the ambulatory group (3.7%, P < .001 for both comparisons). The mean percentage of total lung affected by air trapping was 25.4%, 34.6%, and 27.3% in the ambulatory, hospitalized, and ICU groups, respectively, and 7.2% in healthy controls (P < .001). Air trapping correlated with the residual volume-to-total lung capacity ratio (ρ = 0.6, P < .001). Conclusion In survivors of COVID-19, small airways disease occurred independently of initial infection severity. The long-term consequences are unknown. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Elicker in this issue.

Chest CT Findings in Coronavirus Disease-19 (COVID-19): Relationship to Duration of Infection.

In this retrospective study, chest CTs of 121 symptomatic patients infected with coronavirus disease-19 (COVID-19) from four centers in China from January 18, 2020 to February 2, 2020 were reviewed for common CT findings in relationship to the time between symptom onset and the initial CT scan (i.e. early, 0-2 days (36 patients), intermediate 3-5 days (33 patients), late 6-12 days (25 patients)). The hallmarks of COVID-19 infection on imaging were bilateral and peripheral ground-glass and consolidative pulmonary opacities. Notably, 20/36 (56%) of early patients had a normal CT. With a longer time after the onset of symptoms, CT findings were more frequent, including consolidation, bilateral and peripheral disease, greater total lung involvement, linear opacities, "crazy-paving" pattern and the "reverse halo" sign. Bilateral lung involvement was observed in 10/36 early patients (28%), 25/33 intermediate patients (76%), and 22/25 late patients (88%).

Acute Dyspnea and Hemoptysis in an 84-Year-Old Man With Multiple Comorbidities.

CASE PRESENTATION: An 84-year-old man with an active smoking habit presented to the ED with dyspnea, hemoptysis, and thick phlegm that was difficult to clear. He reported no weight loss, no fever, and no chest pain or dysphonia. He denied both international travel and previous contact with confirmed cases of TB or SARS-CoV-2. He had no known occupational exposures. The patient's personal history included a resolved complete atrioventricular block that required a permanent pacemaker, moderate-to-severe COPD, rheumatoid arthritis (treated with oral prednisone, 2.5 mg/d) and B-chronic lymphocytic leukemia (treated with methotrexate and prophylactic oral supplements of ferrous sulfate). Moreover, he was in medical follow up because of a peptic ulcer, atrophic gastritis, and colonic diverticulosis. The patient also had a history of thoracic surgery after an episode of acute mediastinitis from an odontogenic infection, which required ICU management and temporal tracheostomy.

The Utility of Handheld Cardiac and Lung Ultrasound in Predicting Outcomes of Hospitalised Patients With COVID-19.

BACKGROUND: Strict isolation precautions limit formal echocardiography use in the setting of COVID-19 infection. Information on the importance of handheld focused ultrasound for cardiac evaluation in these patients is scarce. This study investigated the utility of a handheld echocardiography device in hospitalised patients with COVID-19 in diagnosing cardiac pathologies and predicting the composite end point of in-hospital death, mechanical ventilation, shock, and acute decompensated heart failure. METHODS: From April 28 through July 27, 2020, consecutive patients diagnosed with COVID-19 underwent evaluation with the use of handheld ultrasound (Vscan Extend with Dual Probe; GE Healthcare) within 48 hours of admission. The patients were divided into 2 groups: "normal" and "abnormal" echocardiogram, as defined by biventricular systolic dysfunction/enlargement or moderate/severe valvular regurgitation/stenosis. RESULTS: Among 102 patients, 26 (25.5%) had abnormal echocardiograms. They were older with more comorbidities and more severe presenting symptoms compared with the group with normal echocardiograms. The prevalences of the composite outcome among low- and high-risk patients (oxygen saturation < 94%) were 3.1% and 27.1%, respectively. Multivariate logistic regression analysis revealed that an abnormal echocardiogram at presentation was independently associated with the composite end point (odds ratio 6.19, 95% confidence interval 1.50-25.57; P = 0.012). CONCLUSIONS: An abnormal echocardiogram in COVID-19 infection settings is associated with a higher burden of medical comorbidities and independently predicts major adverse end points. Handheld focused echocardiography can be used as an important "rule-out" tool among high-risk patients with COVID-19 and should be integrated into their routine admission evaluation. However, its routine use among low-risk patients is not recommended.

The use of ultrasound in establishing COVID-19 infection as part of a trauma evaluation.

PURPOSE: The use of lung ultrasound for diagnosis of COVID-19 has emerged during the pandemic as a beneficial diagnostic modality due to its rapid availability, bedside use, and lack of radiation. This study aimed to determine if routine ultrasound (US) imaging of the lungs of trauma patients with COVID-19 infections who undergo extended focused assessment with sonography for trauma (EFAST) correlates with computed tomography (CT) imaging and X-ray findings, as previously reported in other populations. METHODS: This was a prospective, observational feasibility study performed at two level 1 trauma centers. US, CT, and X-ray imaging were retrospectively reviewed by a surgical trainee and a board-certified radiologist to determine any correlation of imaging findings in patients with active COVID-19 infection. RESULTS: There were 53 patients with lung US images from EFAST available for evaluation and COVID-19 testing. The overall COVID-19 positivity rate was 7.5%. COVID-19 infection was accurately identified by one patient on US by the trainee, but there was a 15.1% false-positive rate for infection based on the radiologist examination. CONCLUSIONS: Evaluation of the lung during EFAST cannot be used in the trauma setting to identify patients with active COVID-19 infection or to stratify patients as high or low risk of infection. This is likely due to differences in lung imaging technique and the presence of concomitant thoracic injury.

Relationship of the pulmonary disease severity scoring with thromboembolic complications in COVID-19.

PURPOSE: To correlate thromboembolic (TE) complications secondary to COVID-19 with the extent of the pulmonary parenchymal disease using CT severity scores and other comorbidities. METHODS: In total, 185 patients with COVID-19 and suspected thromboembolic complications were classified into two groups based on the presence or absence of thromboembolic complications. Thromboembolic complications were categorized based on location. Chest CT severity scoring system was used to assess the pulmonary parenchymal disease severity in all patients. Based into severity scores, patients were categorized into three groups (mild, moderate, and sever disease). RESULTS: The final study cohort consisted of 171 patients (99 male and 72 female) after excluding 14 patients with non-diagnostic CT pulmonary angiography. The TE group included 53 patients with a mean age of 55.1 ± 7.1, while the non-TE group included 118 patients with a mean age of 52.9 ± 10.8. Patients with BMI > 30 kg/m2 or having a history of smoking and HTN were found more frequently in the TE group (p < 0.05). Patients admitted to ICU were significantly higher in the TE group (p < 0.001). There was statistically significant difference (p = 0.002) in chest CT-SS between the TE group (22.8 ± 11.4) and non-TE group (17.6 ± 10.7). The percentage of severe parenchymal disease in the TE group was significantly higher compared to the non-TE group (p < 0.05). Severe parenchymal disease, BMI > 30 kg/m2, smoking, and HTN had a higher and more significant odds ratio for developing TE complications. CONCLUSION: The present data suggest that severe pulmonary parenchymal disease secondary to COVID-19 is associated with a higher incidence of thromboembolic complications.

Chest CT abnormalities in COVID-19: a systematic review.

Computed tomography (CT) of the chest is one of the main diagnositic tools for coronavirus disease 2019 (COVID-19) infection. To document the chest CT findings in patients with confirmed COVID-19 and their association with the clinical severity, we searched related literatures through PubMed, MEDLINE, Embase, Web of Science (inception to May 4, 2020) and reviewed reference lists of previous systematic reviews. A total of 31 case reports (3768 patients) on CT findings of COVID-19 were included. The most common comorbid conditions were hypertension (18.4%) and diabetes mellitus (8.3%). The most common symptom was fever (78.7%), followed by cough (60.2%). It took an average of 5.6 days from symptom onset to admission. The most common chest CT finding was vascular enlargement (84.8%), followed by ground-glass opacity (GGO) (60.1%), air-bronchogram (47.8%), and consolidation (41.4%). Most lung lesions were located in the lung periphery (72.2%) and involved bilateral lung (76%). Most patients showed normal range of laboratory findings such as white blood cell count (96.4%) and lymphocyte (87.2%). Compared to previous published meta-analyses, our study is the first to summarize the different radiologic characteristics of chest CT in a total of 3768 COVID-19 patients by compiling case series studies. A comprehensive diagnostic approach should be adopted for patients with known COVID-19, suspected cases, and for exposed individuals.

CT of Post-Acute Lung Complications of COVID-19.

The acute course of COVID-19 is variable and ranges from asymptomatic infection to fulminant respiratory failure. Patients recovering from COVID-19 can have persistent symptoms and CT abnormalities of variable severity. At 3 months after acute infection, a subset of patients will have CT abnormalities that include ground-glass opacity (GGO) and subpleural bands with concomitant pulmonary function abnormalities. At 6 months after acute infection, some patients have persistent CT changes to include the resolution of GGOs seen in the early recovery phase and the persistence or development of changes suggestive of fibrosis, such as reticulation with or without parenchymal distortion. The etiology of lung disease after COVID-19 may be a sequela of prolonged mechanical ventilation, COVID-19-induced acute respiratory distress syndrome (ARDS), or direct injury from the virus. Predictors of lung disease after COVID-19 include need for intensive care unit admission, mechanical ventilation, higher inflammatory markers, longer hospital stay, and a diagnosis of ARDS. Treatments of lung disease after COVID-19 are being investigated, including the potential of antifibrotic agents for prevention of lung fibrosis after COVID-19. Future research is needed to determine the long-term persistence of lung disease after COVID-19, its impact on patients, and methods to either prevent or treat it. © RSNA, 2021.

[Lung ultrasonography: what the cardiologist should know]. / Ecografia del polmone: quello che il cardiologo dovrebbe conoscere.

In recent years, lung ultrasonography has acquired an important role as a valuable diagnostic tool in clinical practice. The lung is usually poorly explorable, but it provides more acoustic information in pathological conditions that modify the relationship between air, water and tissues. The different acoustic impedance of all these components makes the chest wall a powerful ultrasound reflector: this is responsible for the creation of several artifacts providing valuable information about lung pathophysiology. Lung ultrasonography helps in the diagnostic process of parenchymal and pleural pathologies, in the differential diagnosis of dyspnea and in the clinical and prognostic evaluation of the SARS-CoV-2 infection.