A radiomics-based model for the outcome prediction in COVID-19 positive patients through deep learning with both longitudinal chest x-ray and chest computed tomography images

The coronavirus disease 2019 (COVID-19) pandemic had a major impact on global health and was associated with millions of deaths worldwide. During the pandemic, imaging characteristics of chest X-ray (CXR) and chest computed tomography (CT) played an important role in the screening, diagnosis and monitoring the disease progression. Various studies suggested that quantitative image analysis methods including artificial intelligence and radiomics can greatly boost the value of imaging in the management of COVID-19. However, few studies have explored the use of longitudinal multi-modal medical images with varying visit intervals for outcome prediction in COVID-19 patients. This study aims to explore the potential of longitudinal multimodal radiomics in predicting the outcome of COVID-19 patients by integrating both CXR and CT images with variable visit intervals through deep learning. 2274 patients who underwent CXR and/or CT scans during disease progression were selected for this study. Of these, 946 patients were treated at the University of Pennsylvania Health System (UPHS) and the remaining 1328 patients were acquired at Stony Brook University (SBU) and curated by the Medical Imaging and Data Resource Center (MIDRC). 532 radiomic features were extracted with the Cancer Imaging Phenomics Toolkit (CaPTk) from the lung regions in CXR and CT images at all visits. We employed two commonly used deep learning algorithms to analyze the longitudinal multimodal features, and evaluated the prediction results based on the area under the receiver operating characteristic curve (AUC). Our models achieved testing AUC scores of 0.816 and 0.836, respectively, for the prediction of mortality. © 2023 SPIE.

Assessment of Jordanian physicians' knowledge about venous thromboembolism risk and management among COVID-19 patients

Objectives: The aim of this study was to assess Jordanian physicians' awareness about venous thromboembolism (VTE) risk among COVID-19 patients and its treatment protocol. Method(s): This was a cross-sectional-based survey that was conducted in Jordan in 2020. During the study period, a convenience sample of physicians working in various Jordanian hospitals were invited to participate in this study. Physicians' knowledge was evaluated and physicians gained one point for each correct answer. Then, a knowledge score out of 23 was calculated for each. Key Findings: In this study, 102 physicians were recruited. Results from this study showed that most of the physicians realize that all COVID-19 patients need VTE risk assessment (n = 69, 67.6%). Regarding VTE prophylaxis, the majority of physicians (n = 91, 89.2%) agreed that low molecular weight heparin (LMWH) is the best prophylactic option for mild-moderate COVID-19 patients with high VTE risk. Regarding severe/critically ill COVID-19 patients, 75.5% of physicians (n = 77) recognized that LMWH is the correct prophylactic option in this case, while 80.4% of them (n = 82) knew that mechanical prevention is the preferred prophylactic option for severe/critically ill COVID-19 patients with high bleeding risk. Moreover, 77.5% of physicians (n = 79) knew that LMWH is the treatment of choice for COVID-19 patients diagnosed with VTE. Finally, linear regression analysis showed that consultants had an overall higher knowledge score about VTE prevention and treatment in COVID-19 patients compared with residents (P = 0.009). Conclusion(s): All physicians knew about VTE risk factors for COVID-19 patients. However, consultants showed better awareness of VTE prophylaxis and treatment compared with residents. We recommend educational workshops be conducted to enhance physicians' knowledge and awareness about VTE thromboprophylaxis and management in COVID-19 patients.Copyright © 2022 The Author(s). Published by Oxford University Press on behalf of the Royal Pharmaceutical Society. All rights reserved.

SAUNet++: an automatic segmentation model of COVID-19 lesion from CT slices

The coronavirus disease 2019 (COVID-19) epidemic has spread worldwide and the healthcare system is in crisis. Accurate, automated and rapid segmentation of COVID-19 lesion in computed tomography (CT) images can help doctors diagnose and provide prognostic information. However, the variety of lesions and small regions of early lesion complicate their segmentation. To solve these problems, we propose a new SAUNet++ model with squeeze excitation residual (SER) module and atrous spatial pyramid pooling (ASPP) module. The SER module can assign more weights to more important channels and mitigate the problem of gradient disappearance;the ASPP module can obtain context information by atrous convolution using various sampling rates. In addition, the generalized dice loss (GDL) can reduce the correlation between lesion size and dice loss, and is introduced to solve the problem of small regions segmentation of COVID-19 lesion. We collected multinational CT scan data from China, Italy and Russia and conducted extensive comparative and ablation studies. The experimental results demonstrated that our method outperforms state-of-the-art models and can effectively improve the accuracy of COVID-19 lesion segmentation on the dice similarity coefficient (our: 87.38% vs. U-Net++: 84.25%), sensitivity (our: 93.28% vs. U-Net++: 89.85%) and Hausdorff distance (our: 19.99 mm vs. U-Net++: 26.79 mm), respectively.

Acute or chronic? Failing left ventricle in a 34 year old

Submission content Introduction: An unusual case of a very young patient without previously known cardiac disease presenting with severe left ventricular failure, detected by a point of care echocardiogram. Main Body: A 34 year old previously well man was brought to hospital after seeing his general practitioner with one month of progressive shortness of breath on exertion. This began around the time the patient received his second covid-19 vaccination. He was sleeping in a chair as he was unable to lie flat. Abnormal observations led the GP to call an ambulance. In the emergency department, the patient required oxygen 5L/min to maintain SpO2 >94%, but he was not in respiratory distress at rest. Blood pressure was 92/53mmHg, mean 67mmHg. Point of care testing for COVID-19 was negative. He was alert, with warm peripheries. Lactate was 1.0mmol/L and he was producing more than 0.5ml/kg/hr of urine. There was no ankle swelling. ECG showed sinus tachycardia. He underwent CT pulmonary angiography which demonstrated no pulmonary embolus, but there was bilateral pulmonary edema. Troponin was 17ng/l, BNP was 2700pg/ml. Furosemide 40mg was given intravenously by the general medical team. Critical care outreach asked for an urgent intensivist review given the highly unusual diagnosis of pulmonary edema in a man of this age. An immediate FUSIC Heart scan identified a dilated left ventricle with end diastolic diameter 7cm and severe global systolic impairment. The right ventricle was not severely impaired, with TAPSE 18mm. There was no significant pericardial effusion. Multiple B lines and trace pulmonary effusions were identified at the lung bases. The patient was urgently discussed with the regional cardiac unit in case of further deterioration, basic images were shared via a cloud system. A potential diagnosis of vaccination-associated myocarditis was considered,1 but in view of the low troponin, the presentation was felt most likely to represent decompensated chronic dilated cardiomyopathy. The patient disclosed a family history of early cardiac death in males. Aggressive diuresis was commenced. The patient was admitted to a monitored bed given the potential risk of arrhythmia or further haemodynamic deterioration. Advice was given that in the event of worsening hypotension, fluids should not be administered but the cardiac centre should be contacted immediately. Formal echocardiography confirmed the POCUS findings, with ejection fraction <35%. He was initiated on ACE inhibitors and beta adrenergic blockade. His symptoms improved and he was able to return home and to work, and is currently undergoing further investigations to establish the etiology of his condition. Conclusion(s): Early echocardiography provided early evidence of a cardiac cause for the patient's presentation and highlighted the severity of the underlying pathology. This directed early aggressive diuresis and safety-netting by virtue of discussion with a tertiary cardiac centre whilst it was established whether this was an acute or decompensated chronic pathology. Ultrasound findings: PLAX, PSAX and A4Ch views demonstrating a severely dilated (7cm end diastolic diameter) left ventricle with global severe systolic impairment.

Detecting Coronavirus (COVID -19) Disease Cues from Chest Radiography Images

This paper proposes a deep learning-based approach to detect COVID-19 infections in lung tissues from chest Computed Tomography (CT) images. A two-stage classification model is designed to identify the infection from CT scans of COVID-19 and Community Acquired Pneumonia (CAP) patients. The proposed neural model named, Residual C-NiN uses a modified convolutional neural network (CNN) with residual connections and a Network-in-Network (NiN) architecture for COVID-19 and CAP detection. The model is trained with the Signal Processing Grand Challenge (SPGC) 2021 COVID dataset. The proposed neural model achieves a slice-level classification accuracy of 93.54% on chest CT images and patient-level classification accuracy of 86.59% with class-wise sensitivity of 92.72%, 55.55%, and 95.83% for COVID-19, CAP, and Normal classes, respectively. Experimental results show the benefit of adding NiN and residual connections in the proposed neural architecture. Experiments conducted on the dataset show significant improvement over the existing state-of-the-art methods reported in the literature. © 2022 ACM.

Semi-Supervised Modified-UNet for Lung Infection Image Segmentation

Automatic lung infection segmentation in computed tomography (CT) scans can offer great assistance in radiological diagnosis by improving accuracy and reducing time required for diagnosis. The biggest challenges for deep learning (DL) models in segmenting infection region are the high variances in infection characteristics, fuzzy boundaries between infected and normal tissues, and the troubles in getting large number of annotated data for training. To resolve such issues, we propose a Modified U-Net (Mod-UNet) model with minor architectural changes and significant modifications in the training process of vanilla 2D UNet. As part of these modifications, we updated the loss function, optimization function, and regularization methods, added a learning rate scheduler and applied advanced data augmentation techniques. Segmentation results on two Covid-19 Lung CT segmentation datasets show that the performance of Mod-UNet is considerably better than the baseline U-Net. Furthermore, to mitigate the issue of lack of annotated data, the Mod-UNet is used in a semi-supervised framework (Semi-Mod-UNet) which works on a random sampling approach to progressively enlarge the training dataset from a large pool of unannotated CT slices. Exhaustive experiments on the two Covid-19 CT segmentation datasets and on a real lung CT volume show that the Mod-UNet and Semi-Mod-UNet significantly outperform other state-of-theart approaches in automated lung infection segmentation. IEEE

Improved method for COVID-19 Classification of Complex-Architecture CNN from Chest CT volumes using Orthogonal Ensemble Networks

This paper introduces the improved method for the COVID-19 classification based on computed tomography (CT) volumes using a combination of a complex-architecture convolutional neural network (CNN) and orthogonal ensemble networks (OEN). The novel coronavirus disease reported in 2019 (COVID-19) is still spreading worldwide. Early and accurate diagnosis of COVID-19 is required in such a situation, and the CT scan is an essential examination. Various computer-aided diagnosis (CAD) methods have been developed to assist and accelerate doctors' diagnoses. Although one of the effective methods is ensemble learning, existing methods combine some major models which do not specialize in COVID-19. In this study, we attempted to improve the performance of a CNN for the COVID-19 classification based on chest CT volumes. The CNN model specializes in feature extraction from anisotropic chest CT volumes. We adopt the OEN, an ensemble learning method considering inter-model diversity, to boost its feature extraction ability. For the experiment, We used chest CT volumes of 1283 cases acquired in multiple medical institutions in Japan. The classification result on 257 test cases indicated that the combination could improve the classification performance. © 2023 SPIE.

Integrating deep learning algorithms for the lung segmentation and body-part-specific anatomical classification with Medical Imaging and Data Resource Center (MIDRC)

Medical Imaging and Data Resource Center (MIDRC) has been built to support AI-based research in response to the COVID-19 pandemic. One of the main goals of MIDRC is to make data collected in the repository ready for AI analysis. Due to data heterogeneity, there is a need to standardize data and make data-mining easier. Our study aims to stratify imaging data according to underlying anatomy using open-source image processing tools. The experiments were performed using Google Colaboratory on computed tomography (CT) imaging data available from the MIDRC. We adopted the existing open-source tools to process CT series (N=389) to define the image sub-volumes according to body part classification, and additionally identified series slices containing specific anatomic landmarks. Cases with automatically identified chest regions (N=369) were then processed to automatically segment the lungs. In order to assess the accuracy of segmentation, we performed outlier analysis using 3D shape radiomics features extracted from the left and right lungs. Standardized DICOM objects were created to store the resulting segmentations, regions, landmarks and radiomics features. We demonstrated that the MIDRC chest CT collections can be enriched using open-source analysis tools and that data available in MIDRC can be further used to evaluate the robustness of publicly available tools. © 2023 SPIE.

Invasive aspergillosis in patients with COVID-19 in intensive care units: results of a multicenter study.

Objective. To study risk factors, clinical and radiological features and effectiveness of the treatment of invasive aspergillosis (IA) in adult patients with COVID-19 (COVID-IA) in intensive care units (ICU). Materials and methods. A total of 60 patients with COVID-IA treated in ICU (median age 62 years, male - 58%) were included in this multicenter prospective study. The comparison group included 34 patients with COVID-IA outside the ICU (median age 62 years, male - 68%). ECMM/ISHAM 2020 criteria were used for diagnosis of CAPA, and EORTC/MSGERC 2020 criteria were used for evaluation of the treatment efficacy. A case-control study (one patient of the main group per two patients of the control group) was conducted to study risk factors for the development and features of CAPA. The control group included 120 adult COVID-19 patients without IA in the ICU, similar in demographic characteristics and background conditions. The median age of patients in the control group was 63 years, male - 67%. Results. 64% of patients with COVID-IA stayed in the ICU. Risk factors for the COVID-IA development in the ICU: chronic obstructive pulmonary disease (OR = 3.538 [1.104-11.337], p = 0.02), and prolonged (> 10 days) lymphopenia (OR = 8.770 [4.177-18.415], p = 0.00001). The main location of COVID-IA in the ICU was lungs (98%). Typical clinical signs were fever (97%), cough (92%), severe respiratory failure (72%), ARDS (64%) and haemoptysis (23%). Typical CT features were areas of consolidation (97%), hydrothorax (63%), and foci of destruction (53%). The effective methods of laboratory diagnosis of COVID-IA were test for galactomannan in BAL (62%), culture (33%) and microscopy (22%) of BAL. The main causative agents of COVID-IA are A. fumigatus (61%), A. niger (26%) and A. flavus (4%). The overall 12-week survival rate of patients with COVID-IA in the ICU was 42%, negative predictive factors were severe respiratory failure (27.5% vs 81%, p = 0.003), ARDS (14% vs 69%, p = 0.001), mechanical ventilation (25% vs 60%, p = 0.01), and foci of destruction in the lung tissue on CT scan (23% vs 59%, p = 0.01). Conclusions. IA affects predominantly ICU patients with COVID-19 who have concomitant medical conditions, such as diabetes mellitus, hematological malignancies, cancer, and COPD. Risk factors for COVID-IA in ICU patients are prolonged lymphopenia and COPD. The majority of patients with COVID-IA have their lungs affected, but clinical signs of IA are non-specific (fever, cough, progressive respiratory failure). The overall 12-week survival in ICU patients with COVID-IA is low. Prognostic factors of poor outcome in adult ICU patients are severe respiratory failure, ARDS, mechanical ventilation as well as CT signs of lung tissue destruction.Copyright © 2022, Interregional Association for Clinical Microbiology and Antimicrobial Chemotherapy. All rights reserved.

Deep learning-based COVID-19 diagnosis using CT scans with laboratory and physiological parameters

The global economy has been dramatically impacted by COVID-19, which has spread to be a pandemic. COVID-19 virus affects the respiratory system, causing difficulty breathing in the patient. It is crucial to identify and treat infections as soon as possible. Traditional diagnostic reverse transcription-polymerase chain reaction (RT-PCR) methods require more time to find the infection. A high infection rate, slow laboratory analysis, and delayed test results caused the widespread and uncontrolled spread of the disease. This study aims to diagnose the COVID-19 epidemic by leveraging a modified convolutional neural network (CNN) to quickly and safely predict the disease's appearance from computed tomography (CT) scan images and a laboratory and physiological parameters dataset. A dataset representing 500 patients was used to train, test, and validate the CNN model with results in detecting COVID-19 having an accuracy, sensitivity, specificity, and F1-score of 99.33%, 99.09%, 99.52%, and 99.24%, respectively. These experimental results suggest that our strategy performs better than previously published approaches.

A serum interleukin-6-based analysis of patients with nasopharyngeal swab reverse-transcriptase polymerase chain reaction negative for SARS-CoV-2 infection with CO-RADS 4 and 5 on computed tomography of the chest

Background: Serum interleukin 6 (IL-6) levels have been studied in the diagnostic evaluation of patients with severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) disease (COVID-19). Methods: We studied the utility of treatment with tocilizumab in COVID-19 patients (n=19) with a negative nasopharyngeal swab real time reverse transcriptase polymerase chain reaction (RT-PCR) test for SARS-CoV-2 who had suggestive computed tomography (CT) findings, namely, COVID-19 Reporting and Data System (CO-RADS) 4,5. Results: Receiver operator characteristic (ROC) curve analysis showed that serum IL 6 at a cut-off of >56.9 pg/L was a predictor of mortality in nasopharyngeal swab RT-PCR negative patients with suggestive CT findings. Tocilizumab had no significant effect on the mortality. Conclusions: In nasopharyngeal swab RT-PCR negative patients with suggestive chest CT findings, elevated serum IL-6 levels > 56.9 pg/L predicted mortality. However, treatment with tocilizumab had no effect on mortality.

Lung cancer screening use before & during the COVID-19 pandemic in the US: Population-based evidence from 2019-2021

Introduction: Lung cancer is the leading cause of cancer-related death in the US with an estimated 236,740 new cases and 130,180 deaths expected in 2022. While early detection with low-dose computed tomography reduces lung cancer mortality by at least 20%, there has been a low uptake of lung cancer screening (LCS) use in the US. The COVID-19 pandemic caused significant disruption in cancer screening. Yet, little is known about how COVID-19 impacted already low use of LCS. This study aims to estimate LCS use before (2019) and during (2020 and 2021) the COVID-19 pandemic among LCS-eligible population in the US. Method(s): We used population-based, nationally representative, cross-section data from the 2019 (n=4,484), 2020 (n=1,239) and 2021 (n=1,673) Behavioral Risk Factor Surveillance System, Lung Cancer Screening module. The outcome was self-reported LCS use among eligible adults in the past 12 months. For 2019 and 2020, the eligibility was defined based on US Preventive Services Task Force (USPSTF) initial criteria-adults aged 55 to 80 years old, who were current and former smokers (had quit within the past 15 years) with at least 30 pack years of smoking history. For 2021, we used the USPSTF updated criteria- adults aged 50 to 80 years, current and former smokers (who had quit within the past 15 years) with at least 20 pack years of smoking history. We applied sampling weights to account for the complex survey design to generate population estimates and conducted weighted descriptive statistics and logistic regression models. Result(s): Overall, there were an estimated 1,559,137 LCS-eligible respondents from 16 US states in 2019 (AZ, ID, KY, ME, MN, MS, MT, NC, ND, PA, RI, SC, UT, VT, WV, WI), 200,301 LCS-eligible respondents from five states in 2020 (DE, ME, NJ, ND, SD), and 668,359 LCS-eligible respondents from four states in 2021 (ME, MI, NJ, RI). Among 2,427,797 LCS-eligible adults, 254,890;38,875;and 122,240 individuals reported receiving LCS in 2019, 2020 and 2021, respectively. Overall, 16.4% (95% CI 14.4-18.5), 19.4% (95% CI 15.3-24.3), and 18.3% (95% CI 15.6-21.3) received LCS during 2019, 2020, and 2021, respectively. In all years, the proportion of LCS use was higher among adults aged 65-74, insured, those with fair and poor health, lung disease and history of cancer (other than lung cancer). In 2020, a higher proportion of adults living in urban areas reported receiving LCS compared to those living in rural areas (20.36% vs. 12.7%, p=0.01). Compared to non-Hispanic White adults, the odds of receiving LCS was lower among Hispanic adults and higher among Non-Hispanic American Indian/Alaskan Native adults in 2020 and 2021, respectively. Conclusion(s): LCS uptake remains low in the US. An estimated 2,011,792 adults at high-risk for developing lung cancer did not receive LCS during 2019, 2020 and 2021. Efforts should be focused to increase LCS awareness and uptake across the US to reduce lung cancer burden.

Case Report of Succesful Multidisiplinary Treatment for Patient with Metastatic Lung Cancer and Relapsed/Refractory Chronic Lymphatic Leukemia/Small Lymphocytic Lymphoma

Introduction: Croatian National Cancer Registry of Croatian Institute for Public Health reported that in year 2020 lung cancer was the second most common cancer site diagnosed in men with 16% and the third most common in women with 10% incidence among all cancer sites. Unfortunatelly lung cancer has the highest mortality in both men and women. Haematological malignancies had 7% share in all malignancies in both male and female cances cases. In 2020 190 newly diagnosed cases of lymphatic leukemia in men and 128 cases in women were reporeted, meaning 1.5 and 1.2% of all malignancies, respectively. Chronic lymphatic leukemia (CLL) is an advanced age disease and incidence increases with age. Impaired immunity, T and B cell dysfunction in CLL, chromosomal aberations, long-term immunosuppressive therapy and genetic factors can all cause secondary malignancies. Co- occurence of solid tumors and CLL is very rare. Although patiens with CLL have an increased risk of developing second primary malignancies including lung carcinoma, the data about their clinical outcomes are lacking. Parekh et al. retrospectively analyzed patients with simultaneous CLL and lung carcinoma over a 20-year period, and they found that ~2% of patients with CLL actually developed lung carcinoma. The authors claimed that up to 38% of patients will also develop a third neoplasm more likely of the skin (melanoma and basal cell carcinoma), larynx (laryngeal carcinoma) or colon. Currently there are no specific guidelines for concurrent CLL and non-small cell lung carcinoma (NSCLC) treatment. Usually, when the tumors are diagnosed simultaneously, treatment is based to target the most aggressive malignancy, as the clinical outcomes depend on the response of the tumor with the poorest prognosis. For this reason, a multidisciplinary approach is mandatory. Case report: A patient with history of coronary heart disease, myocardial infarction and paroxysmal atrial fibrillation was diagnosed in 2019 (at the age of 71) with B chronic lymphocytic leukemia with bulky tumor (inguinal lymph nodes 8x5 cm), stage B according to Binet, intermediate risk. He was treated with 6 cycles of chemoimmunotherapy (rituximab/cyclofosfamid/fludarabine). In 10/2019 remission was confirmed, but MSCT described tumor in the posterior segment of upper right lung lobe measuring 20x17 mm and bilateral metastases up to 11 mm. Bronchoscopy and biopsy were performed, and EGFR neg, ALK neg, ROS 1 neg, PD-L1>50% adenocarcinoma was confirmed. He was referred to Clinical Hospital Center Osijek where monotherapy with pembrolizumab in a standard dose of 200 mg intravenously was started in 01/2020. Partial remission was confirmed in October 2020. Immunotherapy was discontinued due to development of pneumonitis, dysphagia and severe weight loss (20kg), but without radiologically confirmed disease progression. At that time he was referred to our hospital for further treatment. Gastroscopy has shown erosive gastritis with active duodenal ulcus, Forrest III. Supportive therapy and proton pump inhibitor were introduced. After complete regression of pneumonitis, improvement of general condition and resolution of dysphagia, no signs of lung cancer progression were found and pembrolizumab was reintroduced in 12/2021. Hypothyroidism was diagnosed in 01/2021 and levothyroxine replacement ther apy was started. In 03/2021 he underwent surgical removal of basal cell carcinoma of skin on the right temporal region with lobe reconstruction. From 02/2021, when pembrolizumab was reintroduced, regression in tumor size was continously confirmed with complete recovery of general condition. He was hospitalized for COVID 19 infection in 09/2021, and due to complications pembrolizumab was discontinued till 11/2021. Lung cancer immunotherapy proceeded till 11/2022, when Multidisciplinary team decided to finish pembrolizumab because of CLL relapse. CLL was in remission till August 2022 when due to B symptoms, lymphcytosis, anemia and generalized lymphadenopathy, hematological workup including biopsy of cervical lymph node was performed and CLL/SLL relapse was confirmed. Initially chlorambucil was introduced, but disease was refractory. Based on cytogenetic test results (IGHV unmutated, negative TP53) and due to cardiovascular comorbidity (contraindication for BTK inhibitors) venetoclax and rituximab were started in 01/2023. After just 1 cycle of treatment normal blood count as well as regression of B symptoms and peripheral lymphadenopathy occured, indicating the probability of complete disease remission. In our patient with metastatic lung adenocarcinoma excellent disease control is achieved during 41 month of treatment in first line setting. Furthermore, relapsed/refractory CLL/SLL is currently in confirmed remission. Conclusion(s): Successful treatment of patients with multiple primary malignancies is based on multidisciplinarity, early recognition and management of side effects, treatment of comorbidities with the aim of prolonging life, controlling symptoms of disease and preserving quality of life.

Convalescent plasma transfusion in severe or critically ill COVID 19 patients: a rapid systematic review

Currently, there is no availability of any proven specific treatment or prevention strategy to fight against COVID-19. Convalescent plasma (CP) therapy is expected to increase survival rates in COVID-19 as in the case of emerging viral infection (SARS-CoV and MERS-CoV). To collect all the studies relevant to CP therapy in critically ill or severe COVID-19 patients and summarize the findings. The systematic review was conducted according to the PRISMA consensus statement. A systematic search was performed in PubMed, Scopus, Web of Science, and Cochrane databases on April 25, 2020. A total of six studies (28 patients) relevant to CP therapy in severe or critical COVID-19 are considered for inclusion. Two authors extracted the data about study characteristics, demographics, symptoms, co-morbidities, clinical classification of COVID-19, drug therapies, oxygen therapy, laboratory results, chest CT, neutralizing antibody titer, SARS-CoV-2 RNA load, aal outcome. The review findings revealed that CP therapy increases lymphocyte count, reduced s serum inflammatory markers (CRP, IL-6, Procalcitonin) and liver enzyme levels (AST or ALT). There was a rise in serum neutralizing antibody titers in 10 of 14 patients after CP transfusion. In 4 of 14 patients, the titer levels remain unchanged after CP transfusion. All 28 cases (100%) achieved negative to the SARS-CoV-2 RNA after CP transfusion. The convalescent plasma transfusion can improve neutralizing antibody titers and reduces the viral load in severe/critical COVID-19 patients. The review recommends a well-controlled trial design is required to give a definite statement on the safety and efficacy of convalescent plasma therapy in severe/critical COVID-19.

Comparative Analysis of Linear Regression and Logistic Regression for the COVID-19 Detection from CT-Scans

The new emerging Coronavirus disease (COVID-19) is a pandemic disease due to its enormous infectious capability. Generally affecting the lungs, COVID-19 engenders fever, dry cough, and tiredness. However, some patients may not show symptoms. An imaging test, such as a chest X-ray or a chest CT scan, is therefore requested for reliable detection of this pneumonia type. Despite the decreasing trends both in the new and death reported cases, there is an extent need for quick, accurate, and inexpensive new methods for diagnosis. In this framework, we propose two machine learning (ML) algorithms: linear regression and logistic regression for effective COVID-19 detection in the abdominal Computed Tomography (CT) dataset. The ML methods proposed in this paper, effectively classify the data into COVID-19 and normal classes without recourse to image preprocessing or analysis. The effectiveness of these algorithms was shown through the use of the performance measures: accuracy, precision, recall, and F1-score. The best classification accuracy was obtained as 96% with logistic regression using the saga solver with no added penalty against 95.3% with linear regression. As for precision, recall, and F1-score the value of 0.89 was reached by logistic regression for all these metrics, as well as the value of 0.87 by linear regression. © 2022 IEEE.

A mask-guided attention deep learning model for COVID-19 diagnosis based on an integrated CT scan images database

The global extent of COVID-19 mutations and the consequent depletion of hospital resources highlighted the necessity of effective computer-assisted medical diagnosis. COVID-19 detection mediated by deep learning models can help diagnose this highly contagious disease and lower infectivity and mortality rates. Computed tomography (CT) is the preferred imaging modality for building automatic COVID-19 screening and diagnosis models. It is well-known that the training set size significantly impacts the performance and generalization of deep learning models. However, accessing a large dataset of CT scan images from an emerging disease like COVID-19 is challenging. Therefore, data efficiency becomes a significant factor in choosing a learning model. To this end, we present a multi-task learning approach, namely, a mask-guided attention (MGA) classifier, to improve the generalization and data efficiency of COVID-19 classification on lung CT scan images. The novelty of this method is compensating for the scarcity of data by employing more supervision with lesion masks, increasing the sensitivity of the model to COVID-19 manifestations, and helping both generalization and classification performance. Our proposed model achieves better overall performance than the single-task (without MGA module) baseline and state-of-the-art models, as measured by various popular metrics.

Cytokine elimination in suspected COVID-19 perymyocarditis

Background: Only recently studies have been able to demonstrate the safety and efficacy of purification therapies in inflammatory diseases. Here we present the management of a young (21y) male patient in severe cardiogenic shock due to COVID-19 perymyocarditis admitted to the ICU at Bolzano Central Hospital. November 30th 2020 the patient developed high fever (>40 C) and diarrhea. After unsuccessfully being treated orally with a macrolide he was admitted to a peripheral hospital the 4th of December. The day after he deteriorated, required transfer to the ICU, endotracheal intubation and pharmacological cardiovascular support (Norepinephrine, Levosimendan). Antimicrobial treatment was started with piperacillin/tazobactam, linezolid and metronidazole. Despite multiple radiological and microbiological diagnostic attempts the origin of this severe septic shock remained unclear. December 6th the patient was transferred to Bolzano Central Hospital for VA-ECMO evaluation. Method(s): The transesophageal echocardiography revealed 15-20% of EF, lactate (5,2 mmol/l), cardiac enzymes (TropT 1400 mcg/l) and inflammatory parameters (PCT 35 ng/ml, IL-6 685 pg/ml) were elevated. We performed cardiac monitoring via Swan-Ganz catheter. The cardiac index was 1,6 l/min/m2. The peak dosage for Norepinephrine reached 7,5mg/h (1,47 mcg/kg/min). At Bolzano ICU we facilitate the pharmacological therapy with milrinone, vasopressin and low dose epinephrine. Furthermore, we impost continuous hemodiafiltration with CytoSorb filter. Result(s): Only hours after the start of filtration therapy the patient improved and we were able to gradually reduce catecholamine therapy, lactate values decreased. A VA-ECMO implantation was no more necessary. December 10th, we saw a stable patient without ventilatory or cardiovascular support, at echocardiography we revealed a normal EF. Conclusion(s): Clinically we saw a young patient in severe septic/cardiogenic shock due to perimyocarditis. Yet diagnostic attempts (CT-scan, multiple blood/urinary/liquor cultures) remained negative. Despite multiple negative PCR tests for SARS-CoV2 infection we performed specific immunoglobulin analysis and received a positive result for IgM. We therefore conclude on a COVID-19 associated perymyocarditis. Furthermore, this case illustrates the potential benefit of cytokine filtration and elimination in COVID-19 patients with altered IL6 levels.

СOURSE FEATURES OF A NEW CORONAVIRUS INFECTION IN A COMORBID PATIENT

The appearance of a new coronavirus infection (COVID-19) in 2020 caused by the SARS-CoV-2 virus set tasks for doctors of various specialties to quickly diagnose, treat, and develop effective rehabilitation measures. The medical community's knowledge about the respiratory tract lesions pathogenesis course in COVID-19 is going to improve, but the key accents placement in understanding this pathology course continues today. Suspected SARS-CoV-2 virus reference points are as follows: vascular endothelial dysfunction, coagulopathy, thrombosis resembling the antiphospholipid syndrome. Treatment is carried out in accordance with general recommendations aimed at the average patient despite the higher secondary infectious complications risk in patients suffering from cancer and a high severe COVID-19 risk. A successful inpatient treatment experience in patients suffering from comorbid pulmonary pathology and a new coronavirus infection is demonstrated on a separate example. The treatment duration and the complexity of selecting a rehabilitation measures course were due to the patient's history of central squamous cell lung cancer, surgical intervention (bilobectomy), radio- and chemotherapy, as well as complications in the form of pulmonary embolism. The concomitant respiratory pathology was differentiated (chronic obstructive pulmonary disease) during examination and treatment and basic therapy was selected, which made the medical rehabilitation stage more effective. The patient's condition required a more careful selection of combined anti-inflammatory, broncholytic, mucolytic, and antibacterial therapy than in patients without concomitant pathology. Treatment and a complex of rehabilitation measures, normalization of respiratory function, compensation for concomitant bronchopulmonary pathology was possible to be achieves as a result of the diagnosis of concomitant bronchopulmonary pathology. Careful diagnostic search and optimal treatment of all somatic pathology are important factors in the selection of adequate therapy for elderly patients suffering from coronavirus infection with comorbid pulmonary pathology. All rights reserved © Eco-Vector, 2022.

Barotrauma in a Previously Healthy and Young COVID-19 Patient: an Autopsy Case Report

Introduction/Objective Since the emergence of a novel SARS-CoV-2 virus caused coronavirus disease 2019 (COVID-19), a great number of autopsy studies have been published. However, histopathologic studies focused on pulmonary barotrauma are very rare. Here we report an autopsy confined to the lungs on a young COVID-19 patient. Methods/Case Report The patient was a 37-year-old male, non-smoker, with no significant past medical history, and a body mass index of 24.1, who presented with shortness of breath and cough. A computerized tomography (CT) showed features of atypical pneumonia. The main abnormal laboratory data included elevated partial thromboplastin time, fibrinogen, and D-Dimer. The patient had been on mechanical ventilation for 35 days, and was complicated by recurrent pneumothoraces, hypotension, and worsening hypoxia. An autopsy limited to the lungs was performed after the patient expired. Grossly, the lungs showed increased weight, adhesions on visceral pleural surface, patchy consolidation and dilated subpleural cysts. Histological examination revealed cystically dilated/remodeled airspaces with extensive coagulative necrosis, focal alveolar hemorrhage and edema, focal confluent fibrosis, and subpleural blebs. Fresh fibrinous thrombi were seen in small- and medium-sized vessels. Viral cytopathic changes or significant inflammation were not observed. The findings in the lungs were consistent with barotrauma in COVID-19. Results (if a Case Study enter NA) NA. Conclusion This case demonstrates various histopathologic changes of the lungs in a previously healthy and young COVID-19 patient with prolonged hospital course of mechanical ventilation. The features of diffuse alveolar damage with inflammation usually seen in the early stage of barotrauma are not identified. Our findings in the lungs may represent the histopathologic characteristics of the later stage of barotrauma in COVID-19.

IDDF2023-ABS-0005 Hepatocellular carcinoma (HCC) treatment experience using smart fusion in dialysis patients

BackgroundApproximately 700 dialysis patients are seen at our hospital. Among them are patients with HCC that develop viral hepatitis. Advances in ultrasound systems have improved the accuracy of HCC treatment and diagnosis. This time, we had the opportunity to use microwaves for dialysis patients using Smart Fusion and needle navigation installed in APLIOi800 so that we will report it.MethodsTen dialysis patients were treated from January 2018 to February 2023. An Emprint (Covidien, USA) antenna was used for treatment. Canon APLIOi800(Canon, Tochigi, Japan) was used. The built-in function is Smart Fusion. This method can display ultrasound imaging and volume data from other modalities, such as CT and MRI, in association with positional information using a magnetic sensor. Needle navigation has a function that can confirm the position of the needle. It is possible to treat even when the tumor is overprinted and the visualization is poor due to bubbles. Informed consent was obtained from all patients and the treatment was performed.ResultsIt was possible to visualize all tumors. In this study, CT images were used in 0 cases, and MRI was used in 1 Case. No serious side effects occurred after treatment.ConclusionsUsing this method, it was thought that dialysis patients could be safely and accurately treated.