DSSAE: Deep Stacked Sparse Autoencoder Analytical Model for COVID-19 Diagnosis by Fractional Fourier Entropy

( Aim ) COVID-19 has caused more than 2.28 million deaths till 4/Feb/2021 while it is still spreading across the world. This study proposed a novel artificial intelligence model to diagnose COVID-19 based on chest CT images. ( Methods ) First, the two-dimensional fractional Fourier entropy was used to extract features. Second, a custom deep stacked sparse autoencoder (DSSAE) model was created to serve as the classifier. Third, an improved multiple-way data augmentation was proposed to resist overfitting. ( Results ) Our DSSAE model obtains a micro-averaged F1 score of 92.32% in handling a four-class problem (COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis, and healthy control). ( Conclusion ) Our method outperforms 10 state-of-the-art approaches.

Deep Ensemble Learning-Based Models for Diagnosis of COVID-19 from Chest CT Images

Novel coronavirus (COVID-19) has been endangering human health and life since 2019. The timely quarantine, diagnosis, and treatment of infected people are the most necessary and important work. The most widely used method of detecting COVID-19 is real-time polymerase chain reaction (RT-PCR). Along with RT-PCR, computed tomography (CT) has become a vital technique in diagnosing and managing COVID-19 patients. COVID-19 reveals a number of radiological signatures that can be easily recognized through chest CT. These signatures must be analyzed by radiologists. It is, however, an error-prone and time-consuming process. Deep Learning-based methods can be used to perform automatic chest CT analysis, which may shorten the analysis time. The aim of this study is to design a robust and rapid medical recognition system to identify positive cases in chest CT images using three Ensemble Learning-based models. There are several techniques in Deep Learning for developing a detection system. In this paper, we employed Transfer Learning. With this technique, we can apply the knowledge obtained from a pre-trained Convolutional Neural Network (CNN) to a different but related task. In order to ensure the robustness of the proposed system for identifying positive cases in chest CT images, we used two Ensemble Learning methods namely Stacking and Weighted Average Ensemble (WAE) to combine the performances of three fine-tuned Base-Learners (VGG19, ResNet50, and DenseNet201). For Stacking, we explored 2-Levels and 3-Levels Stacking. The three generated Ensemble Learning-based models were trained on two chest CT datasets. A variety of common evaluation measures (accuracy, recall, precision, and F1-score) are used to perform a comparative analysis of each method. The experimental results show that the WAE method provides the most reliable performance, achieving a high recall value which is a desirable outcome in medical applications as it poses a greater risk if a true infected patient is not identified.

Therapeutic effect of cycloserine combined with anti-tuberculosis drugs in the treatment of multidrug-resistant tuberculosis

Purpose: The present study analyzes the comprehensive therapeutic effect of cycloserine, in combination with anti-tuberculosis drugs using chest X-ray and chest CT (computed tomography) scan techniques. Methods: A total of 90 patients, diagnosed with multidrug resistant tuberculosis (MDR TB) were subjected to chest x-ray and CT scan before and after treatment in the two groups. Different views such as sagittal, coronal, lung window and multiplanar imaging of mediastinal window were taken. Some parameters such as case detection rate (CDR) in chest X-ray and CT scan and comprehensive curative effect were observed in two groups. Further, the changes in chest CT signs in addition to absorption of focus, cavity closure and changes in CT extra pulmonary signs were also observed. Results: The clinical profile of the patients and the course of disease were statistically insignificant (p > 0.05). Total effectiveness rate and case detection rate (CDR) values exhibited a significant difference between the groups (p < 0.05). Lung consolidation, nodules and cavities significantly improved in both groups before and after the treatment (p < 0.05). Both groups showed significant improvements in extrapulmonary signs in CT scan (p < 0.05) after the treatment. Conclusion: Based on the study outcomes, the CT scan method has good potentials for diagnosing and treating MDR TB at the early stages. Further, it can clarify the signs and outcomes of the disease at early stages, thus providing the medical fraternity a great opportunity to cure the disease.

Pulmonary Sequelae in Patients After Recovery From Coronavirus Disease 2019: A Follow-Up Study With Chest CT

Objective: The pulmonary sequelae of coronavirus disease 2019 (COVID-19) have not been comprehensively evaluated. We performed a follow-up study analyzing chest computed tomography (CT) findings of COVID-19 patients at 3 and 6 months after hospital discharge.Methods: Between February 2020 and May 2020, a total of 273 patients with COVID-19 at the Shenzhen Third People's Hospital were recruited and followed for 6 months after discharge. Chest CT scanning was performed with the patient in the supine position at end-inspiration. A total of 957 chest CT scans was obtained at different timepoints. A semi-quantitative score was used to assess the degree of lung involvement.Results: Most chest CT scans showed bilateral lung involvement with peripheral location at 3 and 6 months follow-up. The most common CT findings were ground-glass opacity and parenchymal band, which were found in 136 (55.3%) and 94 (38.2%) of the 246 patients at 3 months follow-up, and 82 (48.2%) and 76 (44.7%) of 170 patients at 6 months follow-up, respectively. The number of lobes involved and the total CT severity score declined over time. The total CT score gradually increased with the increasement of disease severity at both 3 months follow-up (trend test P &lt; 0.001) and 6 months follow-up (trend test P &lt; 0.001). Patients with different disease severity represented diverse CT patterns over time.Conclusions: The most common CT findings were ground-glass opacity and parenchymal bands at the 3 and 6 months follow-up. Patients with different disease severity represent diverse CT manifestations, indicating the necessary for long-term follow-up monitoring of patients with severe and critical conditions.

Liver Transplant Imaging prior to and during the COVID-19 Pandemic

Background. The suspension of the surgical activity, the burden of the infection in immunosuppressed patients, and the comorbidities underlying end-stage organ disease have impacted transplant programs significantly, even life-saving procedures, such as liver transplantation. Methods. A review of the literature was conducted to explore the challenges faced by transplant programs and the adopted strategies to overcome them, with a focus on indications for imaging in liver transplant candidates. Results. Liver transplantation relies on an appropriate imaging method for its success. During the Coronavirus Disease 2019 (COVID-19) pandemic, chest CT showed an additional value to detect early signs of SARS-CoV-2 infection and other screening modalities are less accurate than radiology. Conclusion. There is an emerging recognition of the chest CT value to recommend its use and help COVID-19 detection in patients. This examination appears highly sensitive for liver transplant candidates and recipients, who otherwise would have not undergone it, particularly when asymptomatic.

Venous Air Microembolism in Chest CT Angiography: Evidence of Normalization of Deviance Phenomenon

Purpose This article provides evidence that detection of venous air microbubbles (VAMB) in chest computed tomography angiography (CTA) can be an indicator for “normalization of deviance” phenomenon in CT. Method and Materials Institutional review board-approved retrospective study, with waiver for informed consent. Contrast-enhanced chest CT performed during 6 months were reviewed for presence of VAMB in venous segments visible in chest CT (subclavian, brachiocephalic vein, superior vena cava) and cardiac chambers. VAMB volumes were quantified through a semiautomatic method (MIAlite plugin for OsiriX), using a region of interest (ROI) covering the bubble. With basal results, protocols for correct injection technique were reinforced, and VAMB were estimated again at 1 and 3 months. Six months later, questionnaires were sent to the CT technologists to inquire about their perception of VAMB. Descriptive measures with central distribution and dispersion were performed; statistical significance was considered at p < 0.05. Results A total of 602 chest CTA were analyzed, 332 were women (55.14%), with a median age of 58 (interquartile range [IQR] 44–72) years. Among those, 16.11% (100 cases) presented VAMB. Most were emergency department patients (51.6%), male (50.3%), with a median age of 54 (IQR 26) years. There was no difference on detection of VAMB regarding sex (p = 0.19), age (p = 0.46), or referral diagnosis (p = 0.35). Mean air bubbles volume was 0.2 mL (range 0.01–3.4 mL). After intervention, the number of exams with VAMB dropped to 3.29 % (3/91) (p < 0.001). On the 6-month query, 50% of the technicians still considered that VMAB is inevitable, and 60% thought that the occurrence is not associated to risk, and therefore, not actionable. Conclusion VAMB are a frequent finding in chest CTA, and being independent from patient-related variables, it is likely due to technical issues such as intravenous access manipulation during the exam. Reduction after reinforcement of proper performance, and certification of a low concern from CT technicians for any risk associated, provides evidence that there is normalization of deviance in this everyday procedure.