Effective deep learning approaches for predicting COVID-19 outcomes from chest computed tomography volumes.

The rapid evolution of the novel coronavirus disease (COVID-19) pandemic has resulted in an urgent need for effective clinical tools to reduce transmission and manage severe illness. Numerous teams are quickly developing artificial intelligence approaches to these problems, including using deep learning to predict COVID-19 diagnosis and prognosis from chest computed tomography (CT) imaging data. In this work, we assess the value of aggregated chest CT data for COVID-19 prognosis compared to clinical metadata alone. We develop a novel patient-level algorithm to aggregate the chest CT volume into a 2D representation that can be easily integrated with clinical metadata to distinguish COVID-19 pneumonia from chest CT volumes from healthy participants and participants with other viral pneumonia. Furthermore, we present a multitask model for joint segmentation of different classes of pulmonary lesions present in COVID-19 infected lungs that can outperform individual segmentation models for each task. We directly compare this multitask segmentation approach to combining feature-agnostic volumetric CT classification feature maps with clinical metadata for predicting mortality. We show that the combination of features derived from the chest CT volumes improve the AUC performance to 0.80 from the 0.52 obtained by using patients' clinical data alone. These approaches enable the automated extraction of clinically relevant features from chest CT volumes for risk stratification of COVID-19 patients.

Pneumonia detection in chest X-ray images using an ensemble of deep learning models.

Pneumonia is a respiratory infection caused by bacteria or viruses; it affects many individuals, especially in developing and underdeveloped nations, where high levels of pollution, unhygienic living conditions, and overcrowding are relatively common, together with inadequate medical infrastructure. Pneumonia causes pleural effusion, a condition in which fluids fill the lung, causing respiratory difficulty. Early diagnosis of pneumonia is crucial to ensure curative treatment and increase survival rates. Chest X-ray imaging is the most frequently used method for diagnosing pneumonia. However, the examination of chest X-rays is a challenging task and is prone to subjective variability. In this study, we developed a computer-aided diagnosis system for automatic pneumonia detection using chest X-ray images. We employed deep transfer learning to handle the scarcity of available data and designed an ensemble of three convolutional neural network models: GoogLeNet, ResNet-18, and DenseNet-121. A weighted average ensemble technique was adopted, wherein the weights assigned to the base learners were determined using a novel approach. The scores of four standard evaluation metrics, precision, recall, f1-score, and the area under the curve, are fused to form the weight vector, which in studies in the literature was frequently set experimentally, a method that is prone to error. The proposed approach was evaluated on two publicly available pneumonia X-ray datasets, provided by Kermany et al. and the Radiological Society of North America (RSNA), respectively, using a five-fold cross-validation scheme. The proposed method achieved accuracy rates of 98.81% and 86.85% and sensitivity rates of 98.80% and 87.02% on the Kermany and RSNA datasets, respectively. The results were superior to those of state-of-the-art methods and our method performed better than the widely used ensemble techniques. Statistical analyses on the datasets using McNemar's and ANOVA tests showed the robustness of the approach. The codes for the proposed work are available at https://github.com/Rohit-Kundu/Ensemble-Pneumonia-Detection.

Hemothorax after oocyte retrieval in a patient with a history of COVID-19: a case report.

Spontaneous hemothorax is a rare disorder characterized by pleural fluid hematocrit greater than 50% of the peripheral blood hematocrit without natural or iatrogenic trauma to the lungs or pleural space. Since the first case of COVID-19, more than 85 million cases have been confirmed and most patients have sustained symptoms after more than six months of acute infection. This paper reports the case of a 38-year-old woman without signs of endometriosis and a history of COVID-19 infection who developed spontaneous hemothorax after oocyte retrieval. Three months before undergoing assisted reproductive technology (ART) treatment, the patient had a symptomatic COVID-19 infection with a negative PCR test and a positive IgG test four weeks after the onset of symptoms. Controlled ovarian stimulation and oocyte retrieval were conducted uneventfully. Two hours after oocyte retrieval, the patient developed nausea and mild hypogastric pain. Ten hours after the procedure, the patient went to the emergency department with abdominal pain. Chest computed tomography scans revealed moderate right pleural effusion and laminar left pleural effusion. Since the patient had respiratory symptoms, the choice was made to drain the pleural fluid. Fluid analysis confirmed the patient had right hemothorax (400 mL). After drainage, the patient's clinical and imaging signs improved gradually without complications. The patient was asymptomatic one week after the procedure.

COVID-19 detection using federated machine learning.

The current COVID-19 pandemic threatens human life, health, and productivity. AI plays an essential role in COVID-19 case classification as we can apply machine learning models on COVID-19 case data to predict infectious cases and recovery rates using chest x-ray. Accessing patient's private data violates patient privacy and traditional machine learning model requires accessing or transferring whole data to train the model. In recent years, there has been increasing interest in federated machine learning, as it provides an effective solution for data privacy, centralized computation, and high computation power. In this paper, we studied the efficacy of federated learning versus traditional learning by developing two machine learning models (a federated learning model and a traditional machine learning model)using Keras and TensorFlow federated, we used a descriptive dataset and chest x-ray (CXR) images from COVID-19 patients. During the model training stage, we tried to identify which factors affect model prediction accuracy and loss like activation function, model optimizer, learning rate, number of rounds, and data Size, we kept recording and plotting the model loss and prediction accuracy per each training round, to identify which factors affect the model performance, and we found that softmax activation function and SGD optimizer give better prediction accuracy and loss, changing the number of rounds and learning rate has slightly effect on model prediction accuracy and prediction loss but increasing the data size did not have any effect on model prediction accuracy and prediction loss. finally, we build a comparison between the proposed models' loss, accuracy, and performance speed, the results demonstrate that the federated machine learning model has a better prediction accuracy and loss but higher performance time than the traditional machine learning model.

Chest Computed Tomography (CT) as a Predictor of Clinical Course in Coronavirus Disease.

BACKGROUND Chest imaging may be taken into consideration in detecting viral lung infections, especially if there are no tests available or there is a need for a prompt diagnosis. Imaging modalities enable evaluation of the character and extent of pulmonary lesions and monitoring of the disease course. The aim of this study was to verify the prognostic value of chest CT in COVID-19 patients. MATERIAL AND METHODS We conducted a retrospective review of clinical data and CT scans of 156 patients with SARS-CoV-2 infection confirmed by real-time reverse-transcription polymerase-chain-reaction (rRT-PCR) assay hospitalized in the Central Clinical Hospital of the Ministry of the Interior in Warsaw and in the Medical Centre in Lancut, Poland. The total severity score (TSS) was used to quantify the extent of lung opacification in CT scans. RESULTS The dominant pattern in discharged patients was ground-glass opacities, whereas in the non-survivors, the dominant pulmonary changes were consolidations. The non-survivors were more likely to have pleural effusion, pleural thickening, lymphadenopathy, air bronchogram, and bronchiolectasis. There were no statistically significant differences among the 3 analyzed groups (non-survivors, discharged patients, and patients who underwent prolonged hospitalization) in the presence of fibrotic lesions, segmental or subsegmental pulmonary vessel enlargement, subpleural lines, air bubble sign, and halo sign. CONCLUSIONS Lung CT is a diagnostic tool with prognostic utility in COVID-19 patients. The correlation of the available clinical data with semi-quantitative radiological features enables evaluation of disease severity. The occurrence of specific radiomics shows a positive correlation with prognosis.

Semi-supervised learning for an improved diagnosis of COVID-19 in CT images.

Coronavirus disease 2019 (COVID-19) has been spread out all over the world. Although a real-time reverse-transcription polymerase chain reaction (RT-PCR) test has been used as a primary diagnostic tool for COVID-19, the utility of CT based diagnostic tools have been suggested to improve the diagnostic accuracy and reliability. Herein we propose a semi-supervised deep neural network for an improved detection of COVID-19. The proposed method utilizes CT images in a supervised and unsupervised manner to improve the accuracy and robustness of COVID-19 diagnosis. Both labeled and unlabeled CT images are employed. Labeled CT images are used for supervised leaning. Unlabeled CT images are utilized for unsupervised learning in a way that the feature representations are invariant to perturbations in CT images. To systematically evaluate the proposed method, two COVID-19 CT datasets and three public CT datasets with no COVID-19 CT images are employed. In distinguishing COVID-19 from non-COVID-19 CT images, the proposed method achieves an overall accuracy of 99.83%, sensitivity of 0.9286, specificity of 0.9832, and positive predictive value (PPV) of 0.9192. The results are consistent between the COVID-19 challenge dataset and the public CT datasets. For discriminating between COVID-19 and common pneumonia CT images, the proposed method obtains 97.32% accuracy, 0.9971 sensitivity, 0.9598 specificity, and 0.9326 PPV. Moreover, the comparative experiments with respect to supervised learning and training strategies demonstrate that the proposed method is able to improve the diagnostic accuracy and robustness without exhaustive labeling. The proposed semi-supervised method, exploiting both supervised and unsupervised learning, facilitates an accurate and reliable diagnosis for COVID-19, leading to an improved patient care and management.

Generalized chest CT and lab curves throughout the course of COVID-19.

A better understanding of temporal relationships between chest CT and labs may provide a reference for disease severity over the disease course. Generalized curves of lung opacity volume and density over time can be used as standardized references from well before symptoms develop to over a month after recovery, when residual lung opacities remain. 739 patients with COVID-19 underwent CT and RT-PCR in an outbreak setting between January 21st and April 12th, 2020. 29 of 739 patients had serial exams (121 CTs and 279 laboratory measurements) over 50 ± 16 days, with an average of 4.2 sequential CTs each. Sequential volumes of total lung, overall opacity and opacity subtypes (ground glass opacity [GGO] and consolidation) were extracted using deep learning and manual segmentation. Generalized temporal curves of CT and laboratory measurements were correlated. Lung opacities appeared 3.4 ± 2.2 days prior to symptom onset. Opacity peaked 1 day after symptom onset. GGO onset was earlier and resolved later than consolidation. Lactate dehydrogenase, and C-reactive protein peaked earlier than procalcitonin and leukopenia. The temporal relationships of quantitative CT features and clinical labs have distinctive patterns and peaks in relation to symptom onset, which may inform early clinical course in patients with mild COVID-19 pneumonia, or may shed light upon chronic lung effects or mechanisms of medical countermeasures in clinical trials.

Explainable COVID-19 Detection Using Chest CT Scans and Deep Learning.

This paper explores how well deep learning models trained on chest CT images can diagnose COVID-19 infected people in a fast and automated process. To this end, we adopted advanced deep network architectures and proposed a transfer learning strategy using custom-sized input tailored for each deep architecture to achieve the best performance. We conducted extensive sets of experiments on two CT image datasets, namely, the SARS-CoV-2 CT-scan and the COVID19-CT. The results show superior performances for our models compared with previous studies. Our best models achieved average accuracy, precision, sensitivity, specificity, and F1-score values of 99.4%, 99.6%, 99.8%, 99.6%, and 99.4% on the SARS-CoV-2 dataset, and 92.9%, 91.3%, 93.7%, 92.2%, and 92.5% on the COVID19-CT dataset, respectively. For better interpretability of the results, we applied visualization techniques to provide visual explanations for the models' predictions. Feature visualizations of the learned features show well-separated clusters representing CT images of COVID-19 and non-COVID-19 cases. Moreover, the visualizations indicate that our models are not only capable of identifying COVID-19 cases but also provide accurate localization of the COVID-19-associated regions, as indicated by well-trained radiologists.

Calcification of the thoracic aorta on low-dose chest CT predicts severe COVID-19.

BACKGROUND: Cardiovascular comorbidity anticipates poor prognosis of SARS-CoV-2 disease (COVID-19) and correlates with the systemic atherosclerotic transformation of the arterial vessels. The amount of aortic wall calcification (AWC) can be estimated on low-dose chest CT. We suggest quantification of AWC on the low-dose chest CT, which is initially performed for the diagnosis of COVID-19, to screen for patients at risk of severe COVID-19. METHODS: Seventy consecutive patients (46 in center 1, 24 in center 2) with parallel low-dose chest CT and positive RT-PCR for SARS-CoV-2 were included in our multi-center, multi-vendor study. The outcome was rated moderate (no hospitalization, hospitalization) and severe (ICU, tracheal intubation, death), the latter implying a requirement for intensive care treatment. The amount of AWC was quantified with the CT vendor's software. RESULTS: Of 70 included patients, 38 developed a moderate, and 32 a severe COVID-19. The average volume of AWC was significantly higher throughout the subgroup with severe COVID-19, when compared to moderate cases (771.7 mm3 (Q1 = 49.8 mm3, Q3 = 3065.5 mm3) vs. 0 mm3 (Q1 = 0 mm3, Q3 = 57.3 mm3)). Within multivariate regression analysis, including AWC, patient age and sex, as well as a cardiovascular comorbidity score, the volume of AWC was the only significant regressor for severe COVID-19 (p = 0.004). For AWC > 3000 mm3, the logistic regression predicts risk for a severe progression of 0.78. If there are no visually detectable AWC risk for severe progression is 0.13, only. CONCLUSION: AWC seems to be an independent biomarker for the prediction of severe progression and intensive care treatment of COVID-19 already at the time of patient admission to the hospital; verification in a larger multi-center, multi-vendor study is desired.

Computed tomography characterization and outcome evaluation of COVID-19 pneumonia complicated by venous thromboembolism.

BACKGROUND: COVID-19 is frequently complicated by venous thromboembolism (VTE). Computed tomography (CT) of the chest-primarily usually conducted as low-dose, non-contrast enhanced CT-plays an important role in the diagnosis and follow-up of COVID-19 pneumonia. Performed as contrast-enhanced CT pulmonary angiography, it can reliably detect or rule-out pulmonary embolism (PE). Several imaging characteristics of COVID-19 pneumonia have been described for chest CT, but no study evaluated CT findings in the context of VTE/PE. PURPOSE: In our retrospective study, we analyzed clinical, laboratory and CT imaging characteristics of 50 consecutive patients with RT-PCR proven COVID-19 pneumonia who underwent contrast-enhanced chest CT at two tertiary care medical centers. MATERIAL AND METHODS: All patients with RT-PCR proven COVID-19 pneumonia and contrast-enhanced chest CT performed at two tertiary care hospitals between March 1st and April 20th 2020 were retrospectively identified. Patient characteristics (age, gender, comorbidities), symptoms, date of symptom onset, RT-PCR results, imaging results of CT and leg ultrasound, laboratory findings (C-reactive protein, differential blood count, troponine, N-terminal pro-B-type natriuretic peptide (NT-proBNP), fibrinogen, interleukin-6, D-dimer, lactate dehydrogenase (LDH), creatine kinase (CK), creatine kinase muscle-brain (CKmb) and lactate,) and patient outcome (positive: discharge or treatment on normal ward; negative: treatment on intensive care unit (ICU), need for mechanical ventilation, extracorporeal membrane oxygenation (ECMO), or death) were analyzed. Follow-up was performed until May 10th. Patients were assigned to two groups according to two endpoints: venous thromboembolism (VTE) or no VTE. For statistical analysis, univariate logistic regression models were calculated. RESULTS: This study includes 50 patients. In 14 out of 50 patients (28%), pulmonary embolism was detected at contrast-enhanced chest CT. The majority of PE was detected on CTs performed on day 11-20 after symptom onset. Two patients (14%) with PE simultaneously had evidence of deep vein thrombosis. 15 patients (30%) had a negative outcome (need for intensive care, mechanical ventilation, extracorporeal membrane oxygenation, or death), and 35 patients (70%) had a positive outcome (transfer to regular ward, or discharge). Patients suffering VTE had a statistically significant higher risk of an unfavorable outcome (p = 0.028). In univariate analysis, two imaging characteristics on chest CT were associated with VTE: crazy paving pattern (p = 0.024) and air bronchogram (n = 0.021). Also, elevated levels of NT-pro BNP (p = 0.043), CK (p = 0.023) and D-dimers (p = 0.035) were significantly correlated with VTE. CONCLUSION: COVID-19 pneumonia is frequently complicated by pulmonary embolism (incidence of 28% in our cohort), remarkably with lacking evidence of deep vein thrombosis in nearly all thus affected patients of our cohort. As patients suffering VTE had an adverse outcome, we call for a high level of alertness for PE and advocate a lower threshold for contrast-enhanced CT in COVID-19 pneumonia. According to our observations, this might be particularly justified in the second week of disease and if a crazy paving pattern and / or air bronchogram is present on previous non-enhanced CT.

An efficient mixture of deep and machine learning models for COVID-19 diagnosis in chest X-ray images.

A newly emerged coronavirus (COVID-19) seriously threatens human life and health worldwide. In coping and fighting against COVID-19, the most critical step is to effectively screen and diagnose infected patients. Among them, chest X-ray imaging technology is a valuable imaging diagnosis method. The use of computer-aided diagnosis to screen X-ray images of COVID-19 cases can provide experts with auxiliary diagnosis suggestions, which can reduce the burden of experts to a certain extent. In this study, we first used conventional transfer learning methods, using five pre-trained deep learning models, which the Xception model showed a relatively ideal effect, and the diagnostic accuracy reached 96.75%. In order to further improve the diagnostic accuracy, we propose an efficient diagnostic method that uses a combination of deep features and machine learning classification. It implements an end-to-end diagnostic model. The proposed method was tested on two datasets and performed exceptionally well on both of them. We first evaluated the model on 1102 chest X-ray images. The experimental results show that the diagnostic accuracy of Xception + SVM is as high as 99.33%. Compared with the baseline Xception model, the diagnostic accuracy is improved by 2.58%. The sensitivity, specificity and AUC of this model reached 99.27%, 99.38% and 99.32%, respectively. To further illustrate the robustness of our method, we also tested our proposed model on another dataset. Finally also achieved good results. Compared with related research, our proposed method has higher classification accuracy and efficient diagnostic performance. Overall, the proposed method substantially advances the current radiology based methodology, it can be very helpful tool for clinical practitioners and radiologists to aid them in diagnosis and follow-up of COVID-19 cases.

Clinical manifestations and radiological features by chest computed tomographic findings of a novel coronavirus disease-19 pneumonia among 92 patients in Japan.

INTRODUCTION: The novel coronavirus disease (COVID-19) could cause a severe acute respiratory infectious disease, showing a high mortality rate of 12-45% among cases who required intensive care unit admission. COVID-19 pneumonia PATIENTS AND METHODS: For the purpose of identifying clinical manifestations and radiological findings of COVID-19 pneumonia, we reviewed all cases of COVID-19 pneumonia which were published by the homepage of the Japanese Association for Infectious Diseases from Feb 5 2020 until April 30 2020, including our cases. All patients were diagnosed based on positive results of the novel coronavirus-real-time RT-PCR with chest computed tomography (CT) findings. RESULTS: A total of 92 patients were enrolled in this study. The median age was 66 years (range 16-92 years). For all, 50 (54%) were males. The most common underlying disease was hypertension in 32 (36%). Any comorbidity was seen in 60 (67%). The mortality rate was 4 (6%). In terms of clinical symptoms on an initial visit, fever and cough were confirmed in 66 (72%) and 37 (40%). Forty-three (47%) had no respiratory symptoms. As for radiological findings by chest CT scan, ground-glass opacities (GGO)s, peripheral distribution, bilateral lung involvements were seen in 88 (96%), 76 (83%) and 78 (85%), respectively. CONCLUSION: It is difficult to diagnose as COVID-19 pneumonia due to poor respiratory symptoms. Chest CT findings typically show GGO, peripheral and bilateral shadows. Patients should have chest CT performed if suspected for early diagnosis and therapeutic intervention, resulting in a favorable outcome and prevention of secondary nosocomial transmitted infection.

Determination of disease severity in COVID-19 patients using deep learning in chest X-ray images.

PURPOSE: Chest X-ray plays a key role in diagnosis and management of COVID-19 patients and imaging features associated with clinical elements may assist with the development or validation of automated image analysis tools. We aimed to identify associations between clinical and radiographic features as well as to assess the feasibility of deep learning applied to chest X-rays in the setting of an acute COVID-19 outbreak. METHODS: A retrospective study of X-rays, clinical, and laboratory data was performed from 48 SARS-CoV-2 RT-PCR positive patients (age 60±17 years, 15 women) between February 22 and March 6, 2020 from a tertiary care hospital in Milan, Italy. Sixty-five chest X-rays were reviewed by two radiologists for alveolar and interstitial opacities and classified by severity on a scale from 0 to 3. Clinical factors (age, symptoms, comorbidities) were investigated for association with opacity severity and also with placement of central line or endotracheal tube. Deep learning models were then trained for two tasks: lung segmentation and opacity detection. Imaging characteristics were compared to clinical datapoints using the unpaired student's t-test or Mann-Whitney U test. Cohen's kappa analysis was used to evaluate the concordance of deep learning to conventional radiologist interpretation. RESULTS: Fifty-six percent of patients presented with alveolar opacities, 73% had interstitial opacities, and 23% had normal X-rays. The presence of alveolar or interstitial opacities was statistically correlated with age (P = 0.008) and comorbidities (P = 0.005). The extent of alveolar or interstitial opacities on baseline X-ray was significantly associated with the presence of endotracheal tube (P = 0.0008 and P = 0.049) or central line (P = 0.003 and P = 0.007). In comparison to human interpretation, the deep learning model achieved a kappa concordance of 0.51 for alveolar opacities and 0.71 for interstitial opacities. CONCLUSION: Chest X-ray analysis in an acute COVID-19 outbreak showed that the severity of opacities was associated with advanced age, comorbidities, as well as acuity of care. Artificial intelligence tools based upon deep learning of COVID-19 chest X-rays are feasible in the acute outbreak setting.

Chest-CT findings of COVID-19 in patients with pre-existing malignancies; a pictorial review.

As of April 17th, 2020, more than 2,190,010 COVID-19 cases with 147,010 deaths have been recorded worldwide. It has been suggested that a high mortality rate occurs in patients with severe disease and is associated with advanced age and underlying comorbidities, such as malignancies. To the best of our knowledge, no study has been conducted to evaluate chest CT features in patients with malignancy and concomitant COVID-19 infection. In fact, the imaging findings can be challenging and have not yet been fully understood in this setting. In this manuscript, we go over imaging findings in chest CT of patients with COVID-19 and known cancer. With the ongoing COVID-19 pandemic and exponentially increasing incidence throughout the world, in at-risk and vulnerable populations such as patients with known malignancies, infection with SARS-CoV-2 should be included in the differential considerations even with atypical image pictures. Detection of superimposed infection in patients with cancers who present with pulmonary infiltrations warrant correlation with clinical picture, contact history, and RT-PCR confirmatory testing.