COVID­19 detection from chest X-ray images using transfer learning.

COVID-19 is a kind of coronavirus that appeared in China in the Province of Wuhan in December 2019. The most significant influence of this virus is its very highly contagious characteristic which may lead to death. The standard diagnosis of COVID-19 is based on swabs from the throat and nose, their sensitivity is not high enough and so they are prone to errors. Early diagnosis of COVID-19 disease is important to provide the chance of quick isolation of the suspected cases and to decrease the opportunity of infection in healthy people. In this research, a framework for chest X-ray image classification tasks based on deep learning is proposed to help in early diagnosis of COVID-19. The proposed framework contains two phases which are the pre-processing phase and classification phase which uses pre-trained convolution neural network models based on transfer learning. In the pre-processing phase, different image enhancements have been applied to full and segmented X-ray images to improve the classification performance of the CNN models. Two CNN pre-trained models have been used for classification which are VGG19 and EfficientNetB0. From experimental results, the best model achieved a sensitivity of 0.96, specificity of 0.94, precision of 0.9412, F1 score of 0.9505 and accuracy of 0.95 using enhanced full X-ray images for binary classification of chest X-ray images into COVID-19 or normal with VGG19. The proposed framework is promising and achieved a classification accuracy of 0.935 for 4-class classification.

Investigating the impact of data heterogeneity on the performance of federated learning algorithm using medical imaging.

In recent years, Federated Learning (FL) has gained traction as a privacy-centric approach in medical imaging. This study explores the challenges posed by data heterogeneity on FL algorithms, using the COVIDx CXR-3 dataset as a case study. We contrast the performance of the Federated Averaging (FedAvg) algorithm on non-identically and independently distributed (non-IID) data against identically and independently distributed (IID) data. Our findings reveal a notable performance decline with increased data heterogeneity, emphasizing the need for innovative strategies to enhance FL in diverse environments. This research contributes to the practical implementation of FL, extending beyond theoretical concepts and addressing the nuances in medical imaging applications. This research uncovers the inherent challenges in FL due to data diversity. It sets the stage for future advancements in FL strategies to effectively manage data heterogeneity, especially in sensitive fields like healthcare.

The prognostic value of lung ultrasound score (LUSS) in patients with COVID-19 admitted in Emergency Department: a prospective observational study.

BACKGROUND: The Lung Ultrasound (LUS) is routinely used as a point-of-care imaging tool in Emergency Department (ED) and its role in COVID-19 is being studied. The Lung UltraSound Score (LUSS) is a semi quantitative score of lung damage severity. Alongside instrumental diagnostic, the PaO2/FiO2 (P/F) ratio, obtained from arterial blood gas analysis, is the index used to assess the severity of the acute respiratory distress syndrome (ARDS), according to the Berlin definition. OBJECTIVES: The primary objective of the study was to evaluate a possible correlation between the LUSS score and the P/F Ratio, obtained from the arterial sampling in COVID-19 positive patients. MATERIALS AND METHODS: This was a cross-perspective monocentric observational study and it was carried out in the Emergency Department of the "AOU delle Marche" (Ancona, Italy), from 1 January 2023 to 28 February 2023. The study foresaw, once the patient was admitted to the ED, the execution of the LUS exam and the subsequent calculation of the LUSS score. RESULTS: The sample selected for the study was of 158 patients. The proportion of LUSS ≤4 was statistically higher in those with a P/F >300 (76.2%), compared to those with a P/F ≤300 (13.2%). On the other end, the proportion of LUSS >4 was lower in those who have P/F >300 (23.8%), while it was higher in those who have P/F ≤300 (86.8%). Those patients with a LUSS >4 were 1.76 (95% CI: 1.57 - 1.99) times more likely to have a P/F ≤300, compared to those with LUSS ≤4. The Odds Ratio of having a P/F ≤300 value in those achieving a LUSS >4, compared to those achieving a LUSS ≤4, was 21.0 (95% CI: 8.4 - 52.4). The study identified pO2, Hb and dichotomous LUSS as predictors of the level of P/F ≤300 or P/F >300. DISCUSSION: We found that the LUSS score defined by our study was closely related to the P/F ratio COVID-19 positive patients. Our study presented provides evidence on the potential rule of the LUSS for detecting the stage of lung impairment and the need for oxygen therapy in COVID-19 positive patients.

Neuroimaging Features of Cytokine-related Diseases.

Cytokines are small secreted proteins that have specific effects on cellular interactions and are crucial for functioning of the immune system. Cytokines are involved in almost all diseases, but as microscopic chemical compounds they cannot be visualized at imaging for obvious reasons. Several imaging manifestations have been well recognized owing to the development of cytokine therapies such as those with bevacizumab (antibody against vascular endothelial growth factor) and chimeric antigen receptor (CAR) T cells and the establishment of new disease concepts such as interferonopathy and cytokine release syndrome. For example, immune effector cell-associated neurotoxicity is the second most common form of toxicity after CAR T-cell therapy toxicity, and imaging is recommended to evaluate the severity. The emergence of COVID-19, which causes a cytokine storm, has profoundly impacted neuroimaging. The central nervous system is one of the systems that is most susceptible to cytokine storms, which are induced by the positive feedback of inflammatory cytokines. Cytokine storms cause several neurologic complications, including acute infarction, acute leukoencephalopathy, and catastrophic hemorrhage, leading to devastating neurologic outcomes. Imaging can be used to detect these abnormalities and describe their severity, and it may help distinguish mimics such as metabolic encephalopathy and cerebrovascular disease. Familiarity with the neuroimaging abnormalities caused by cytokine storms is beneficial for diagnosing such diseases and subsequently planning and initiating early treatment strategies. The authors outline the neuroimaging features of cytokine-related diseases, focusing on cytokine storms, neuroinflammatory and neurodegenerative diseases, cytokine-related tumors, and cytokine-related therapies, and describe an approach to diagnosing cytokine-related disease processes and their differentials. ©RSNA, 2024 Supplemental material is available for this article.

Interest of Chest CT to Assess the Prognosis of SARS-CoV-2 Pneumonia: An In-Hospital-Based Experience in Sub-Saharan Africa.

Methods: We included all patients with respiratory symptoms (dyspnea, fever, and cough) and/or respiratory failure admitted to the SOS Médecins de nuit SARL hospital, DR Congo, during the 2nd and 3rd waves of the COVID-19 pandemic. The diagnosis of COVID-19 was established based on RT-PCR anti-SARS-CoV-2 tests (G1 (RT-PCR positive) vs. G2 (RT-PCR negative)), and all patients had a chest CT on the day of admission. We retrieved the digital files of patients, precisely the clinical, biological, and chest CT parameters of the day of admission as well as the vital outcome (survival or death). Chest CT were read by a very high-definition console using Advantage Windows software and exported to the hospital network using the RadiAnt DICOM viewer. To determine the threshold for the percentage of lung lesions associated with all-cause mortality, we used ROC curves. Factors associated with death, including chest CT parameters, were investigated using logistic regression analysis. Results: The study included 200 patients (average age 56.2 ± 15.2 years; 19% diabetics and 4.5% obese), and COVID-19 was confirmed among 56% of them (G1). Chest CT showed that ground glass (72.3 vs. 39.8%), crazy paving (69.6 vs. 17.0%), and consolidation (83.9 vs. 22.7%), with bilateral and peripheral locations (68.8 vs. 30.7%), were more frequent in G1 vs. G2 (p < 0.001). No case of pulmonary embolism and fibrosis had been documented. The lung lesions affecting 30% of the parenchyma were informative in predicting death (area under the ROC curve at 0.705, p = 0.017). In multivariate analysis, a percentage of lesions affecting 50% of the lung parenchyma increased the risk of dying by 7.194 (1.656-31.250). Conclusion: The chest CT demonstrated certain characteristic lesions more frequently in patients in whom the diagnosis of COVID-19 was confirmed. The extent of lesions affecting at least half of the lung parenchyma from the first day of admission to hospital increases the risk of death by a factor of 7.

CT abnormalities 3 and 12 months after hospitalization for COVID-19 and association with disease severity: A prospective cohort study.

OBJECTIVES: To investigate changes in chest CT between 3 and 12 months and associations with disease severity in patients hospitalized for COVID-19 during the first wave in 2020. MATERIALS AND METHODS: Longitudinal cohort study of patients hospitalized for COVID-19 in 2020. Chest CT was performed 3 and 12 months after admission. CT images were evaluated using a CT severity score (CSS) (0-12 scale) and recoded to an abbreviated version (0-3 scale). We analyzed determinants of the abbreviated CSS with multivariable mixed effects ordinal regression. RESULTS: 242 patients completed CT at 3 months, and 124 (mean age 62.3±13.3, 78 men) also at 12 months. Between 3 and 12 months (n = 124) CSS (0-12 scale) for ground-glass opacities (GGO) decreased from median 3 (25th-75th percentile: 0-12) at 3 months to 0.5 (0-12) at 12 months (p<0.001), but increased for parenchymal bands (p<0.001). In multivariable analysis of GGO, the odds ratio for more severe abbreviated CSS (0-3 scale) at 12 months was 0.11 (95%CI 0.11 0.05 to 0.21, p<0.001) compared to 3 months, for WHO severity category 5-7 (high-flow oxygen/non-invasive ventilation/ventilator) versus 3 (non-oxygen use) 37.16 (1.18 to 43.47, p = 0.032), and for age ≥60 compared to <60 years 4.8 (1.33 to 17.6, p = 0.016). Mosaicism was reduced at 12 compared to 3 months, OR 0.33 (95%CI 0.16 to 0.66, p = 0.002). CONCLUSIONS: GGO and mosaicism decreased, while parenchymal bands increased from 3 to 12 months. Persistent GGO were associated with initial COVID-19 severity and age ≥60 years.

A Narrative Review on the Clinical Relevance of Imaging the Circumventricular Brain Organs and Performing Their Anatomical and Histopathological Examination in Acute and Postacute COVID-19.

ABSTRACT: Autopsy followed by histopathological examination is foundational in clinical and forensic medicine for discovering and understanding pathological changes in disease, their underlying processes, and cause of death. Imaging technology has become increasingly important for advancing clinical research and practice, given its noninvasive, in vivo and ex vivo applicability. Medical and forensic autopsy can benefit greatly from advances in imaging technology that lead toward minimally invasive, whole-brain virtual autopsy. Brain autopsy followed by histopathological examination is still the hallmark for understanding disease and a fundamental modus operandi in forensic pathology and forensic medicine, despite the fact that its practice has become progressively less frequent in medical settings. This situation is especially relevant with respect to new diseases such as COVID-19 caused by the SARS-CoV-2 virus, for which our neuroanatomical knowledge is sparse. In this narrative review, we show that ad hoc clinical autopsies and histopathological analyses combined with neuroimaging of the principal circumventricular organs are critical to gaining insight into the reconstruction of the pathophysiological mechanisms and the explanation of cause of death (ie, atrium mortis) related to the cardiovascular effects of SARS-CoV-2 infection in forensic and clinical medicine.

Characteristics of brain lesions found using MRI imaging in patients with post-COVID with signs of cognitive decline.

OBJECTIVE: Aim: To describe and evaluate abnormalities of the brain in post-COVID patients with neurologic symptoms and cognitive deficits using MRI imaging of the brain. PATIENTS AND METHODS: Materials and Methods: We included 21 patients with a previous positive PCR testing for SARS-CoV-2 and one or more of the following symptoms: memory and cognitive decline, dizziness, anxiety, depression, chronic headaches. All patients had MRI imaging done at onset of symptoms, but after at least 1 year after positive testing for COVID-19 based on the patient's previous medical history. RESULTS: Results: All of the patients complained of lack of concentration, forgetfulness, hard to process information. 15 patients suffered from confusion, 10 from anxiety. Of the 21 patients 14 had isolated chronic headaches, 3 had isolated dizziness, 4 patients had both symptoms upon inclusion. All patients underwent MRI imaging as a part of the diagnostic workup and had varying degrees of neurodegeneration. CONCLUSION: Conclusions: Our data correlates with existing research and shows tendency for cognitive decline in post-COVID patients. This provides groundwork for further research to determine correlation between acceleration of neurodegeneration and post-COVID.

Association of chest computed tomography severity score at ICU admission and respiratory outcomes in critically ill COVID-19 patients.

OBJECTIVE: To evaluate the association of a validated chest computed tomography (Chest-CT) severity score in COVID-19 patients with their respiratory outcome in the Intensive Care Unit. METHODS: A single-center, prospective study evaluated patients with positive RT-PCR for COVID-19, who underwent Chest-CT and had a final COVID-19 clinical diagnosis needing invasive mechanical ventilation in the ICU. The admission chest-CT was evaluated according to a validated Chest-CT Severity Score in COVID-19 (Chest-CTSS) divided into low ≤50% (<14 points) and >50% high (≥14 points) lung parenchyma involvement. The association between the initial score and their pulmonary clinical outcomes was evaluated. RESULTS: 121 patients were clustered into the > 50% lung involvement group and 105 patients into the ≤ 50% lung involvement group. Patients ≤ 50% lung involvement (<14 points) group presented lower PEEP levels and FiO2 values, respectively GEE P = 0.09 and P = 0.04. The adjusted COX model found higher hazard to stay longer on invasive mechanical ventilation HR: 1.69, 95% CI, 1.02-2.80, P = 0.042 and the adjusted logistic regression model showed increased risk ventilator-associated pneumonia OR = 1.85 95% CI 1.01-3.39 for COVID-19 patients with > 50% lung involvement (≥14 points) on Chest-CT at ICU admission. CONCLUSION: COVID-19 patients with >50% lung involvement on Chest-CT admission presented higher chances to stay longer on invasive mechanical ventilation and more chances to developed ventilator-associated pneumonia.

Cerebral microstructural alterations in Post-COVID-condition are related to cognitive impairment, olfactory dysfunction and fatigue.

After contracting COVID-19, a substantial number of individuals develop a Post-COVID-Condition, marked by neurologic symptoms such as cognitive deficits, olfactory dysfunction, and fatigue. Despite this, biomarkers and pathophysiological understandings of this condition remain limited. Employing magnetic resonance imaging, we conduct a comparative analysis of cerebral microstructure among patients with Post-COVID-Condition, healthy controls, and individuals that contracted COVID-19 without long-term symptoms. We reveal widespread alterations in cerebral microstructure, attributed to a shift in volume from neuronal compartments to free fluid, associated with the severity of the initial infection. Correlating these alterations with cognition, olfaction, and fatigue unveils distinct affected networks, which are in close anatomical-functional relationship with the respective symptoms.

Longitudinal Echocardiographic Follow-Up of a Pediatric Multisystem Inflammatory Syndrome Cohort.

OBJECTIVE: Significant involvement of the cardiovascular system is known in multisystem inflammatory syndrome in children (MIS-C). This study aimed to examine the recovery of affected cardiovascular parameters over a medium-term follow-up. METHODS: A cohort of 69 children was studied prospectively. Assessments of left ventricular (LV) function and coronary artery abnormalities (CAA) were conducted at admission, 1.5 months, and 3 months. Coronavirus Disease 2019 (COVID-19) antibody titers were assessed at these three time points. Echocardiographic and antibody parameters (rising/decreasing) were analyzed for correlation. Outcomes were assessed using logistic regression. RESULTS: At admission, among the 78.2% of patients who were tested, 88.9% tested positive for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). A quarter of the patients had pericardial effusion, and half had valvulitis. Decreased ejection fraction, global circumferential strain (GCS), and global longitudinal strain (GLS) were seen in 54.4%, 68.6%, and 35.8% of patients, respectively. CAAs were observed in 27.78% of patients. Systolic dysfunction was significantly associated with older age. During follow-up, severe LV dysfunction normalized within 6-7 weeks, while mild to moderate dysfunction reached normalcy by two weeks. Both GCS and GLS reached normalcy within a median of two weeks. Diastolic parameters recovered by six weeks. Most small and moderate coronary aneurysms resolved, but a giant aneurysm in an infant remained large even after 15 months. Trends in antibodies and ejection fraction (EF) at three months were significantly correlated. Admission EF, GLS (at 6 weeks) and deceleration time (at 3 months) were significantly associated with intensive care unit (ICU) admission. The median segmental strain of the cohort remained low in certain segments at three months. CONCLUSION: Smaller CAAs resolve, whereas giant CAAs persist. EF and GLS are important predictors of Pediatric Intensive Care Unit (PICU) stay. The residual impairment of median segmental strain and persistent diastolic dysfunction at three months indicate the need for long-term follow-up.

COVID-19 diagnostic prediction on chest CT scan images using hybrid quantum-classical convolutional neural network.

Notwithstanding the extensive research efforts directed towards devising a dependable approach for the diagnosis of coronavirus disease 2019 (COVID-19), the inherent complexity and capriciousness of the virus continue to pose a formidable challenge to the precise identification of affected individuals. In light of this predicament, it is essential to devise a model for COVID-19 prediction utilizing chest computed tomography (CT) scans. To this end, we present a hybrid quantum-classical convolutional neural network (HQCNN) model, which is founded on stochastic quantum circuits that can discern COVID-19 patients from chest CT images. Two publicly available chest CT image datasets were employed to evaluate the performance of our model. The experimental outcomes evinced diagnostic accuracies of 99.39% and 97.91%, along with precisions of 99.19% and 98.52%, respectively. These findings are indicative of the fact that the proposed model surpasses recently published works in terms of performance, thus providing a superior ability to precisely predict COVID-19 positive instances.Communicated by Ramaswamy H. Sarma.

Clinical and imaging features of patients with COVID-19 in a double-center cohort study.

INTRODUCTION: There is extensive published data on coronavirus disease 2019 (COVID-19). However, information on the effective factors that improve the pulmonary involvement of COVID-19 patients, and long-term clinical and imaging follow-up of these patients is limited. METHODOLOGY: This is a prospective cohort study on patients with COVID-19 who were hospitalized in two major academic hospitals in Yazd, Iran. The correlation between the baseline demographic and clinical/para-clinical data with the imaging resolution status at day 60 was assessed. RESULTS: 122 patients, including 65 males, with an average age of 53.43 years participated in this study. Age, gender, baseline oxygen saturation (O2Sat), and the percentage of lung involvement were the main prognostic factors. Our results suggest that with every year increase in age, the probability of complete imaging resolution decreases by 6.4%. In addition, women are 2.07 times more likely to recover completely. Moreover, each percent increase of baseline O2Sat makes the patients 15.4% more likely to fully recover. As the patients' shortness of breath increases, the probability of recovery decreases by 9.8%.;56.7% of patients who did not recover after 60 days had persistent shortness of breath, while only 21% of those who recovered had symptoms of dyspnea after day 60. CONCLUSIONS: Age, gender, baseline O2Sat, percentage of lung involvement, and shortness of breath were identified as the main risk factors in the recovery of patients with COVID-19. Long-term follow-up of patients with COVID-19, especially patients with high-risk factors, is necessary.

Target sign in COVID-19, radiological interpretation and diagnostic contribution of digital thoracic tomosynthesis.

INTRODUCTION: Our objectives are: To describe the radiological semiology, clinical-analytical features and prognosis related to the target sign (TS) in COVID-19. To determine whether digital thoracic tomosynthesis (DTT) improves the diagnostic ability of radiography. MATERIAL AND METHODS: Retrospective, descriptive, single-centre, case series study, accepted by our ethical committee. Radiological, clinical, analytical and follow-up characteristics of patients with COVID-19 and TS on radiography and DTT between November 2020 and January 2021 were analysed. RESULTS: Eleven TS were collected in 7 patients, median age 35 years, 57% male. All TS presented with a central nodule and a peripheral ring, and in at least 82%, the lung in between was of normal density. All TS were located in peripheral, basal regions and 91% in posterior regions. TS were multiple in 43%. Contiguous TS shared the peripheral ring. Other findings related to pneumonia were associated in 86% of patients. DTT detected 82% more TS than radiography. Only one patient underwent a CT angiography of the pulmonary arteries, positive for acute pulmonary thromboembolism. Seventy-one per cent presented with pleuritic pain. No distinctive laboratory findings or prognostic worsening were detected. CONCLUSIONS: TS in COVID-19 predominates in peripheral and declining regions and can be multiple. Pulmonary thromboembolism was detected in one case. It occurs in young people, frequently with pleuritic pain and does not worsen the prognosis. DTT detects more than 80 % of TS than radiography.

Artificial intelligence in the healthcare sector: comparison of deep learning networks using chest X-ray images.

Purpose: Artificial intelligence has led to significant developments in the healthcare sector, as in other sectors and fields. In light of its significance, the present study delves into exploring deep learning, a branch of artificial intelligence. Methods: In the study, deep learning networks ResNet101, AlexNet, GoogLeNet, and Xception were considered, and it was aimed to determine the success of these networks in disease diagnosis. For this purpose, a dataset of 1,680 chest X-ray images was utilized, consisting of cases of COVID-19, viral pneumonia, and individuals without these diseases. These images were obtained by employing a rotation method to generate replicated data, wherein a split of 70 and 30% was adopted for training and validation, respectively. Results: The analysis findings revealed that the deep learning networks were successful in classifying COVID-19, Viral Pneumonia, and Normal (disease-free) images. Moreover, an examination of the success levels revealed that the ResNet101 deep learning network was more successful than the others with a 96.32% success rate. Conclusion: In the study, it was seen that deep learning can be used in disease diagnosis and can help experts in the relevant field, ultimately contributing to healthcare organizations and the practices of country managers.

High-fidelity intravoxel incoherent motion parameter mapping using locally low-rank and subspace modeling.

PURPOSE: Intravoxel incoherent motion (IVIM) is a quantitative magnetic resonance imaging (MRI) method used to quantify perfusion properties of tissue non-invasively without contrast. However, clinical applications are limited by unreliable parameter estimates, particularly for the perfusion fraction (f) and pseudodiffusion coefficient (D*). This study aims to develop a high-fidelity reconstruction for reliable estimation of IVIM parameters. The proposed method is versatile and amenable to various acquisition schemes and fitting methods. METHODS: To address current challenges with IVIM, we adapted several advanced reconstruction techniques. We used a low-rank approximation of IVIM images and temporal subspace modeling to constrain the magnetization dynamics of the bi-exponential diffusion signal decay. In addition, motion-induced phase variations were corrected between diffusion directions and b-values, facilitating the use of high SNR real-valued diffusion data. The proposed method was evaluated in simulations and in vivo brain acquisitions in six healthy subjects and six individuals with a history of SARS-CoV-2 infection and compared with the conventionally reconstructed magnitude data. Following reconstruction, IVIM parameters were estimated voxel-wise. RESULTS: Our proposed method reduced noise contamination in simulations, resulting in a 60%, 58.9%, and 83.9% reduction in the NRMSE for D, f, and D*, respectively, compared to the conventional reconstruction. In vivo, anisotropic properties of D, f, and D* were preserved with the proposed method, highlighting microvascular differences in gray matter between individuals with a history of COVID-19 and those without (p = 0.0210), which wasn't observed with the conventional reconstruction. CONCLUSION: The proposed method yielded a more reliable estimation of IVIM parameters with less noise than the conventional reconstruction. Further, the proposed method preserved anisotropic properties of IVIM parameter estimates and demonstrated differences in microvascular perfusion in COVID-affected subjects, which weren't observed with conventional reconstruction methods.

Reduced Cortical Thickness Correlates of Cognitive Dysfunction in Post-COVID-19 Condition: Insights from a Long-Term Follow-up.

BACKGROUND AND PURPOSE: There is a paucity of data on long-term neuroimaging findings from individuals who have developed the post-coronavirus 2019 (COVID-19) condition. Only 2 studies have investigated the correlations between cognitive assessment results and structural MR imaging in this population. This study aimed to elucidate the long-term cognitive outcomes of participants with the post-COVID-19 condition and to correlate these cognitive findings with structural MR imaging data in the post-COVID-19 condition. MATERIALS AND METHODS: A cohort of 53 participants with the post-COVID-19 condition underwent 3T brain MR imaging with T1 and FLAIR sequences obtained a median of 1.8 years after Severe Acute Respiratory Syndrome coronavirus 2 (SARS-CoV-2) infection. A comprehensive neuropsychological battery was used to assess several cognitive domains in the same individuals. Correlations between cognitive domains and whole-brain voxel-based morphometry were performed. Different ROIs from FreeSurfer were used to perform the same correlations with other neuroimaging features. RESULTS: According to the Frascati criteria, more than one-half of the participants had deficits in the attentional (55%, n = 29) and executive (59%, n = 31) domains, while 40% (n = 21) had impairment in the memory domain. Only 1 participant (1.89%) showed problems in the visuospatial and visuoconstructive domains. We observed that reduced cortical thickness in the left parahippocampal region (t(48) = 2.28, P = .03) and the right caudal-middle-frontal region (t(48) = 2.20, P = .03) was positively correlated with the memory domain. CONCLUSIONS: Our findings suggest that cognitive impairment in individuals with the post-COVID-19 condition is associated with long-term alterations in the structure of the brain. These macrostructural changes may provide insight into the nature of cognitive symptoms.

COVID-19 Hierarchical Classification Using a Deep Learning Multi-Modal.

Coronavirus disease 2019 (COVID-19), originating in China, has rapidly spread worldwide. Physicians must examine infected patients and make timely decisions to isolate them. However, completing these processes is difficult due to limited time and availability of expert radiologists, as well as limitations of the reverse-transcription polymerase chain reaction (RT-PCR) method. Deep learning, a sophisticated machine learning technique, leverages radiological imaging modalities for disease diagnosis and image classification tasks. Previous research on COVID-19 classification has encountered several limitations, including binary classification methods, single-feature modalities, small public datasets, and reliance on CT diagnostic processes. Additionally, studies have often utilized a flat structure, disregarding the hierarchical structure of pneumonia classification. This study aims to overcome these limitations by identifying pneumonia caused by COVID-19, distinguishing it from other types of pneumonia and healthy lungs using chest X-ray (CXR) images and related tabular medical data, and demonstrate the value of incorporating tabular medical data in achieving more accurate diagnoses. Resnet-based and VGG-based pre-trained convolutional neural network (CNN) models were employed to extract features, which were then combined using early fusion for the classification of eight distinct classes. We leveraged the hierarchal structure of pneumonia classification within our approach to achieve improved classification outcomes. Since an imbalanced dataset is common in this field, a variety of versions of generative adversarial networks (GANs) were used to generate synthetic data. The proposed approach tested in our private datasets of 4523 patients achieved a macro-avg F1-score of 95.9% and an F1-score of 87.5% for COVID-19 identification using a Resnet-based structure. In conclusion, in this study, we were able to create an accurate deep learning multi-modal to diagnose COVID-19 and differentiate it from other kinds of pneumonia and normal lungs, which will enhance the radiological diagnostic process.

[Identifying Novel Coronavirus Pneumonia With CT Images: A Deep Learning Approach With Detail Upsampling and Attention Guidance]. / 从CTå›¾åƒä¸­æ£€æµ‹æ–°åž‹å† çŠ¶ç— æ¯’æ„ŸæŸ“å¯¼è‡´çš„è‚ºç‚Žï¼šä¸€ç§ç»†èŠ‚ä¸Šé‡‡æ ·å’Œæ³¨æ„åŠ›å¼•å¯¼çš„æ·±åº¦å­¦ä¹ æ–¹æ³•.

Objective: To construct a deep learning-based target detection method to help radiologists perform rapid diagnosis of lesions in the CT images of patients with novel coronavirus pneumonia (NCP) by restoring detailed information and mining local information. Methods: We present a deep learning approach that integrates detail upsampling and attention guidance. A linear upsampling algorithm based on bicubic interpolation algorithm was adopted to improve the restoration of detailed information within feature maps during the upsampling phase. Additionally, a visual attention mechanism based on vertical and horizontal spatial dimensions embedded in the feature extraction module to enhance the capability of the object detection algorithm to represent key information related to NCP lesions. Results: Experimental results on the NCP dataset showed that the detection method based on the detail upsampling algorithm improved the recall rate by 1.07% compared with the baseline model, with the AP50 reaching 85.14%. After embedding the attention mechanism in the feature extraction module, 86.13% AP50, 73.92% recall, and 90.37% accuracy were achieved, which were better than those of the popular object detection models. Conclusion: The feature information mining of CT images based on deep learning can further improve the lesion detection ability. The proposed approach helps radiologists rapidly identify NCP lesions on CT images and provides an important clinical basis for early intervention and high-intensity monitoring of NCP patients.

Lung volume measurement using chest CT in COVID-19 patients: a cohort study in Japan.

OBJECTIVE: This study aimed to investigate the utility of CT quantification of lung volume for predicting critical outcomes in COVID-19 patients. METHODS: This retrospective cohort study included 1200 hospitalised patients with COVID-19 from 4 hospitals. Lung fields were extracted using artificial intelligence-based segmentation, and the percentage of the predicted (%pred) total lung volume (TLC (%pred)) was calculated. The incidence of critical outcomes and posthospitalisation complications was compared between patients with low and high CT lung volumes classified based on the median percentage of predicted TLCct (n=600 for each). Prognostic factors for residual lung volume loss were investigated in 208 patients with COVID-19 via a follow-up CT after 3 months. RESULTS: The incidence of critical outcomes was higher in the low TLCct (%pred) group than in the high TLCct (%pred) group (14.2% vs 3.3%, p<0.0001). Multivariable analysis of previously reported factors (age, sex, body mass index and comorbidities) demonstrated that CT-derived lung volume was significantly associated with critical outcomes. The low TLCct (%pred) group exhibited a higher incidence of bacterial infection, heart failure, thromboembolism, liver dysfunction and renal dysfunction than the high TLCct (%pred) group. TLCct (%pred) at 3 months was similarly divided into two groups at the median (71.8%). Among patients with follow-up CT scans, lung volumes showed a recovery trend from the time of admission to 3 months but remained lower in critical cases at 3 months. CONCLUSION: Lower CT lung volume was associated with critical outcomes, posthospitalisation complications and slower improvement of clinical conditions in COVID-19 patients.