Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 20 de 127
Filter
1.
IEEE J Biomed Health Inform ; 26(3): 1080-1090, 2022 03.
Article in English | MEDLINE | ID: covidwho-1759116

ABSTRACT

Pneumonia is one of the most common treatable causes of death, and early diagnosis allows for early intervention. Automated diagnosis of pneumonia can therefore improve outcomes. However, it is challenging to develop high-performance deep learning models due to the lack of well-annotated data for training. This paper proposes a novel method, called Deep Supervised Domain Adaptation (DSDA), to automatically diagnose pneumonia from chest X-ray images. Specifically, we propose to transfer the knowledge from a publicly available large-scale source dataset (ChestX-ray14) to a well-annotated but small-scale target dataset (the TTSH dataset). DSDA aligns the distributions of the source domain and the target domain according to the underlying semantics of the training samples. It includes two task-specific sub-networks for the source domain and the target domain, respectively. These two sub-networks share the feature extraction layers and are trained in an end-to-end manner. Unlike most existing domain adaptation approaches that perform the same tasks in the source domain and the target domain, we attempt to transfer the knowledge from a multi-label classification task in the source domain to a binary classification task in the target domain. To evaluate the effectiveness of our method, we compare it with several existing peer methods. The experimental results show that our method can achieve promising performance for automated pneumonia diagnosis.


Subject(s)
Deep Learning , Pneumonia , Early Diagnosis , Humans , Pneumonia/diagnostic imaging , Tomography, X-Ray Computed/methods , X-Rays
2.
Sci Rep ; 12(1): 4270, 2022 03 11.
Article in English | MEDLINE | ID: covidwho-1740475

ABSTRACT

Inflammatory burden is associated with COVID-19 severity and outcomes. Residual computed tomography (CT) lung abnormalities have been reported after COVID-19. The aim was to evaluate the association between inflammatory burden during COVID-19 and residual lung CT abnormalities collected on follow-up CT scans performed 2-3 and 6-7 months after COVID-19, in severe COVID-19 pneumonia survivors. C-reactive protein (CRP) curves describing inflammatory burden during the clinical course were built, and CRP peaks, velocities of increase, and integrals were calculated. Other putative determinants were age, sex, mechanical ventilation, lowest PaO2/FiO2 ratio, D-dimer peak, and length of hospital stay (LOS). Of the 259 included patients (median age 65 years; 30.5% females), 202 (78%) and 100 (38.6%) had residual, predominantly non-fibrotic, abnormalities at 2-3 and 6-7 months, respectively. In age- and sex-adjusted models, best CRP predictors for residual abnormalities were CRP peak (odds ratio [OR] for one standard deviation [SD] increase = 1.79; 95% confidence interval [CI] = 1.23-2.62) at 2-3 months and CRP integral (OR for one SD increase = 2.24; 95%CI = 1.53-3.28) at 6-7 months. Hence, inflammation is associated with short- and medium-term lung damage in COVID-19. Other severity measures, including mechanical ventilation and LOS, but not D-dimer, were mediators of the relationship between CRP and residual abnormalities.


Subject(s)
COVID-19/pathology , Pneumonia/diagnostic imaging , Aged , C-Reactive Protein/analysis , COVID-19/complications , COVID-19/diagnostic imaging , Female , Humans , Male , Middle Aged , Patient Acuity , Pneumonia/etiology , Pneumonia/pathology , Retrospective Studies , Risk Factors , Time Factors , Tomography, X-Ray Computed
3.
Medicine (Baltimore) ; 101(9): e28950, 2022 Mar 04.
Article in English | MEDLINE | ID: covidwho-1730758

ABSTRACT

ABSTRACT: To characterize computed tomography (CT) findings of coronavirus disease 2019 (COVID-19) pneumonia and their value in outcome prediction.Chest CTs of 182 patients with a confirmed diagnosis of COVID-19 infection by real-time reverse transcription polymerase chain reaction were evaluated for the presence of CT-abnormalities and their frequency. Regarding the patient outcome each patient was categorized in 5 progressive stages and the duration of hospitalization was determined. Regression analysis was performed to find which CT findings are predictive for patient outcome and to assess prognostic factors for the hospitalization duration.Multivariate statistical analysis confirmed a higher age (OR = 1.023, P  =  .025), a higher total visual severity score (OR = 1.038, P  =  .002) and the presence of crazy paving (OR = 2.160, P  =  .034) as predictive parameters for patient outcome. A higher total visual severity score (+0.134 days; P  =  .012) and the presence of pleural effusion (+13.985 days, P  =  0.005) were predictive parameters for a longer hospitalization duration. Moreover, a higher sensitivity of chest CT (false negatives 10.4%; true positives 78.6%) in comparison to real-time reverse transcription polymerase chain reaction was obtained.An increasing percentage of lung opacity as well as the presence of crazy paving and a higher age are associated with a worse patient outcome. The presence of a higher total visual severity score and pleural effusion are significant predictors for a longer hospitalization duration. These results are underscoring the value of chest CT as a diagnostic and prognostic tool in the pandemic outbreak of COVID-19, to facilitate fast detection and to preserve the limited (intensive) care capacity only for the most vulnerable patients.


Subject(s)
COVID-19/diagnostic imaging , Lung/diagnostic imaging , Pneumonia/diagnostic imaging , Tomography, X-Ray Computed/methods , Adult , Aged , Aged, 80 and over , Female , Humans , Male , Middle Aged , Pleural Effusion , Retrospective Studies , SARS-CoV-2
4.
Biomed Res Int ; 2022: 8925930, 2022.
Article in English | MEDLINE | ID: covidwho-1723968

ABSTRACT

COVID-19 is a fatal disease caused by the SARS-CoV-2 virus that has caused around 5.3 Million deaths globally as of December 2021. The detection of this disease is a time taking process that have worsen the situation around the globe, and the disease has been identified as a world pandemic by the WHO. Deep learning-based approaches are being widely used to diagnose the COVID-19 cases, but the limitation of immensity in the publicly available dataset causes the problem of model over-fitting. Modern artificial intelligence-based techniques can be used to increase the dataset to avoid from the over-fitting problem. This research work presents the use of various deep learning models along with the state-of-the-art augmentation methods, namely, classical and generative adversarial network- (GAN-) based data augmentation. Furthermore, four existing deep convolutional networks, namely, DenseNet-121, InceptionV3, Xception, and ResNet101 have been used for the detection of the virus in X-ray images after training on augmented dataset. Additionally, we have also proposed a novel convolutional neural network (QuNet) to improve the COVID-19 detection. The comparative analysis of achieved results reflects that both QuNet and Xception achieved high accuracy with classical augmented dataset, whereas QuNet has also outperformed and delivered 90% detection accuracy with GAN-based augmented dataset.


Subject(s)
COVID-19/diagnostic imaging , Deep Learning , Image Processing, Computer-Assisted/methods , Computer Graphics , Databases, Factual , Humans , Neural Networks, Computer , Pneumonia/diagnostic imaging , Radiography
5.
Gac Med Mex ; 157(4): 391-396, 2021.
Article in English | MEDLINE | ID: covidwho-1705708

ABSTRACT

INTRODUCTION: By the end 2019 there was an outbreak of pneumonia caused by a new coronavirus, a disease that was called coronavirus disease 2019 (COVID-19). Computed tomography (CT) has played an important role in the diagnosis of COVID-19 patients. OBJECTIVE: To demonstrate inter-observer variability with five scales proposed for measuring the extent of COVID-19 pneumonia on tomography. METHODS: Thirty five initial chest CT scans of patients who attended respiratory triage for suspected COVID-19 pneumonia were analyzed. Three radiologists classified the tomographic images according to the severity scales proposed by Yang (1), Yuan (2), Chun (3), Wang (4) and Instituto Nacional de Enfermedades Respiratorias-Chung-Pan (5). The percentage of agreement between the evaluators for each scale was calculated using the intra-class correlation index. RESULTS: In most patients were five pulmonary lobes compromised (77.1% of the patients). Scales 1, 2, 4 and 5 showed an intra-class correlation > 0.91 (p < 0.0001), with agreement thus being almost perfect. CONCLUSIONS: Scale 4 (proposed by Wang) showed the best inter-observer agreement, with a coefficient of 0.964 (p = 0.001).


INTRODUCCIÓN: A finales de 2019 se presentó un brote de neumonía causada por un nuevo coronavirus, enfermedad a la que se denominó COVID-19. La tomografía computarizada ha desempeñado un papel importante en el diagnóstico de los pacientes con COVID-19. OBJETIVO: Demostrar la variabilidad interobservador con cinco escalas propuestas para la medición de la extensión de la neumonía ocasionada por COVID-19 mediante tomografía. MÉTODOS: Se analizaron 35 tomografías de tórax iniciales de pacientes que asistieron al triaje respiratorio por sospecha de neumonía por COVID-19. Tres radiólogos realizaron la clasificación de las imágenes tomográficas de acuerdo con las escalas de severidad propuestas por Yang (1), Yuan (2), Chun (3), Wang (4) e INER-Chung-Pan (5). Se calculó el porcentaje de concordancia entre los evaluadores para cada escala con el índice de correlación intraclase. RESULTADOS: La mayoría de los pacientes presentó afección de cinco lóbulos pulmonares (77.1 % de los pacientes). Las escalas 1, 2, 4 y 5 mostraron una correlación intraclase > 0.91, con p < 0.0001, por lo que la concordancia fue casi perfecta. CONCLUSIONES: La escala 4 (de Wang) mostró la mejor concordancia interobservador, con un coeficiente de 0.964 (p = 0.001).


Subject(s)
COVID-19 , Pneumonia , Humans , Observer Variation , Pneumonia/diagnostic imaging , Pneumonia/epidemiology , Retrospective Studies , SARS-CoV-2 , Tomography, X-Ray Computed
6.
Medicina (Kaunas) ; 58(2)2022 Feb 01.
Article in English | MEDLINE | ID: covidwho-1667239

ABSTRACT

Background and Objectives: Scant data regarding early post-COVID-19 effects are available, especially in younger people. Therefore, the objective of this study was to explore the early clinical impacts of post-COVID-19 pneumonia, comparing severe and non-severe patients. Materials and Methods: A cross-sectional study was conducted in adult patients admitted with COVID-19 pneumonia from April to May 2021. Demographic data, symptoms and signs, quality of life, Hospital Anxiety and Depression Scale (HADS), chest radiograph (CXR), pulmonary function tests (spirometry, impulse oscillometry), fractional exhaled nitric oxide (FeNO), and exercise capacity were assessed one month after hospital discharge. Twenty-five healthy control subjects that were age- and gender-matched were recruited for comparisons. Results: One hundred and five patients, with a mean age of 35.6 ± 15.8 years and 54 (51.4%) males, participated and were categorized into the non-severe pneumonia (N = 68) and severe pneumonia groups (N = 37). At a one-month follow-up visit (the time from the onset of the disease symptoms = 45.4 ± 5.9 days), the severe group had more cough, fatigue, and skin rash with higher dyspnea scale, more residual CXR lesions, and lower quality of life scores. Forced vital capacity (FVC) was lower in the severe group (88.3% of predicted value) and non-severe group (94.6% of predicted value) than in the healthy controls (p = 0.001). The six-minute walk distance was significantly lower in the non-severe group, at 79.2 m, and in the severe group, at 103.8 m, than in the healthy control subjects (p < 0.001). Conclusions: Adult patients with COVID-19, especially those with clinically severe pneumonia, still had residual symptoms and chest radiographic abnormalities, together with poorer quality of life and lower exercise capacity, one month after hospital discharge.


Subject(s)
COVID-19 , Pneumonia , Adult , Cross-Sectional Studies , Humans , Lung/diagnostic imaging , Male , Middle Aged , Pneumonia/diagnostic imaging , Pneumonia/epidemiology , Quality of Life , SARS-CoV-2 , Young Adult
7.
Comput Biol Med ; 141: 105143, 2022 02.
Article in English | MEDLINE | ID: covidwho-1654260

ABSTRACT

BACKGROUND: Even though antibiotics agents are widely used, pneumonia is still one of the most common causes of death around the world. Some severe, fast-spreading pneumonia can even cause huge influence on global economy and life security. In order to give optimal medication regimens and prevent infectious pneumonia's spreading, recognition of pathogens is important. METHOD: In this single-institution retrospective study, 2,353 patients with their CT volumes are included, each of whom was infected by one of 12 known kinds of pathogens. We propose Deep Diagnostic Agent Forest (DDAF) to recognize the pathogen of a patient based on ones' CT volume, which is a challenging multiclass classification problem, with large intraclass variations and small interclass variations and very imbalanced data. RESULTS: The model achieves 0.899 ± 0.004 multi-way area under curves of receiver (AUC) for level-I pathogen recognition, which are five rough groups of pathogens, and 0.851 ± 0.003 AUC for level-II recognition, which are 12 fine-level pathogens. The model also outperforms the average result of seven human readers in level-I recognition and outperforms all readers in level-II recognition, who can only reach an average result of 7.71 ± 4.10% accuracy. CONCLUSION: Deep learning model can help in recognition pathogens using CTs only, which might help accelerate the process of etiological diagnosis.


Subject(s)
Deep Learning , Pneumonia , Forests , Humans , Pneumonia/diagnostic imaging , Retrospective Studies , Tomography, X-Ray Computed
8.
Cardiovasc Ultrasound ; 20(1): 2, 2022 Jan 19.
Article in English | MEDLINE | ID: covidwho-1633049

ABSTRACT

BACKGROUND: This study aimed to investigate the relationship between echocardiography results and lung ultrasound score (LUS) in coronavirus disease 2019 (COVID-19) pneumonia patients and evaluate the impact of the combined application of these techniques in the evaluation of COVID-19 pneumonia. METHODS: Hospitalized COVID-19 pneumonia patients who underwent daily lung ultrasound and echocardiography were included in this study. Patients with tricuspid regurgitation within three days of admission were enrolled. Moreover, the correlation and differences between their pulmonary artery pressure (PAP) and LUS on days 3, 8, and 13 were analyzed. The inner diameter of the pulmonary artery root as well as the size of the atria and ventricles were also considered. RESULTS: The PAP on days 3, 8, and 13 of hospitalization was positively correlated with the LUS (r = 0.448, p = 0.003; r = 0.738, p < 0.001; r = 0.325, p = 0.036, respectively). On day 8, the values of both PAP and LUS were higher than on days 3 and 13 (p < 0.01). Similarly, PAP and LUS were significantly increased in 92.9% (39/42) and 90.5% (38/42) of patients, respectively, and at least one of these two values was positive in 97.6% (41/42) of cases. The inner diameters of the right atrium, right ventricle, and pulmonary artery also differed significantly from their corresponding values on days 3 and 13 (p < 0.05). CONCLUSIONS: PAP is positively correlated with LUS in COVID-19 pneumonia. The two values could be combined for a more precise assessment of disease progression and recovery status.


Subject(s)
COVID-19 , Pneumonia , Echocardiography , Humans , Lung/diagnostic imaging , Pilot Projects , Pneumonia/diagnostic imaging , SARS-CoV-2 , Ultrasonography
9.
Comput Biol Med ; 142: 105220, 2022 03.
Article in English | MEDLINE | ID: covidwho-1611676

ABSTRACT

The coronavirus disease 2019 (COVID-19) has severely stressed the sanitary systems of all countries in the world. One of the main issues that physicians are called to tackle is represented by the monitoring of pauci-symptomatic COVID-19 patients at home and, generally speaking, everyone the access to the hospital might or should be severely reduced. Indeed, the early detection of interstitial pneumonia is particularly relevant for the survival of these patients. Recent studies on rheumatoid arthritis and interstitial lung diseases have shown that pathological pulmonary sounds can be automatically detected by suitably developed algorithms. The scope of this preliminary work consists of proving that the pathological lung sounds evidenced in patients affected by COVID-19 pneumonia can be automatically detected as well by the same class of algorithms. In particular the software VECTOR, suitably devised for interstitial lung diseases, has been employed to process the lung sounds of 28 patient recorded in the emergency room at the university hospital of Modena (Italy) during December 2020. The performance of VECTOR has been compared with diagnostic techniques based on imaging, namely lung ultrasound, chest X-ray and high resolution computed tomography, which have been assumed as ground truth. The results have evidenced a surprising overall diagnostic accuracy of 75% even if the staff of the emergency room has not been suitably trained for lung auscultation and the parameters of the software have not been optimized to detect interstitial pneumonia. These results pave the way to a new approach for monitoring the pulmonary implication in pauci-symptomatic COVID-19 patients.


Subject(s)
COVID-19 , Pneumonia , Algorithms , Humans , Lung , Pneumonia/diagnostic imaging , Respiratory Sounds , SARS-CoV-2
10.
Tuberk Toraks ; 69(4): 492-498, 2021 Dec.
Article in English | MEDLINE | ID: covidwho-1580009

ABSTRACT

Introduction: To date, there is limited data on the long-term changes in the lungs of patients recovering from coronavirus (COVID-19) pneumonia. In order to evaluate pulmonary sequelae, it was planned to investigate fibrotic changes observed as sequelae in lung tissue in 3-6-month control thorax computerized tomography (CT) scans of moderate-to-severe COVID-19 pneumonia survivors. Materials and Methods: A total of 84 patients (mean age: 67.3 years ±15) with moderate-to-severe pneumonia on chest tomography at the time of diagnosis were included in the study, of which 51 (61%) were males and 33 (39%) were females. Initial and follow-up CT scans averaged 8.3 days ± 2.2 and 112.1 days ± 14.6 after symptom onset, respectively. Participants were recorded in two groups as those with and without fibrotic-like changes such as traction bronchiectasis, fibrotic - parenchymal bands, honeycomb appearance according to 3-6 months follow-up CT scans. Differences between the groups were evaluated with a two-sampled t-test. Logistic regression analyzes were performed to determine independent predictive factors of fibrotic-like sequelae changes. Result: On follow-up CTs, fibrotic-like changes were observed in 29 (35%) of the 84 participants (Group 1), while the remaining 55 (65%) showed complete radiological recovery (Group 2). With logistic regression analysis, hospital stay of 22 days or longer (OR: 4.9; 95% CI: 20, 32; p< 0.05) and a CT score of 15 or more at diagnosis (OR: 2.2; 95% CI: 13.5, 18; p< 0.05) were found to be an independent predictor for sequelae fibrotic changes in lung tissue. Conclusions: More than one-third of patients who survived COVID-19 pneumonia had fibrotic-like sequelae changes in the lung parenchyma. These changes were found to be associated with the presence of severe pneumonia at the time of diagnosis and longer hospital stay.


Subject(s)
COVID-19 , Pneumonia , Aged , Female , Follow-Up Studies , Humans , Lung/diagnostic imaging , Male , Pneumonia/diagnostic imaging , Retrospective Studies , SARS-CoV-2 , Thorax , Tomography, X-Ray Computed
11.
J Med Virol ; 94(4): 1289-1291, 2022 Apr.
Article in English | MEDLINE | ID: covidwho-1589037

ABSTRACT

In December 2019, a new type of virus, coronavirus disease 2019 broke out globally and caused great harm. The virus mutates rapidly, and more research reports are urgently needed to increase our understanding of the disease. We found the reversed halo sign (RHS) occurred in the imaging manifestations of severe acute respiratory syndrome coronavirus 2 delta variant of concern pneumonia. In the absence of pathology, the mechanism is unknown. Therefore, we reported two cases of RHS and tried to speculate the pathological mechanism through multiple computed tomography follow-up comparisons to judge the prognosis of the disease.


Subject(s)
COVID-19/diagnostic imaging , SARS-CoV-2/pathogenicity , COVID-19/pathology , COVID-19/virology , Female , Follow-Up Studies , Humans , Male , Middle Aged , Pneumonia/diagnostic imaging , Pneumonia/pathology , Pneumonia/virology , Prognosis , SARS-CoV-2/isolation & purification , Tomography, X-Ray Computed
12.
Comput Biol Med ; 141: 105182, 2022 02.
Article in English | MEDLINE | ID: covidwho-1588025

ABSTRACT

BACKGROUND: Chest computed tomography (CT) is crucial in the diagnosis of coronavirus disease 2019 (COVID-19). However, the persistent pandemic and similar CT manifestations between COVID-19 and community-acquired pneumonia (CAP) raise methodological requirements. METHODS: A fully automatic pipeline of deep learning is proposed for distinguishing COVID-19 from CAP using CT images. Inspired by the diagnostic process of radiologists, the pipeline comprises four connected modules for lung segmentation, selection of slices with lesions, slice-level prediction, and patient-level prediction. The roles of the first and second modules and the effectiveness of the capsule network for slice-level prediction were investigated. A dataset of 326 CT scans was collected to train and test the pipeline. Another public dataset of 110 patients was used to evaluate the generalization capability. RESULTS: LinkNet exhibited the largest intersection over union (0.967) and Dice coefficient (0.983) for lung segmentation. For the selection of slices with lesions, the capsule network with the ResNet50 block achieved an accuracy of 92.5% and an area under the curve (AUC) of 0.933. The capsule network using the DenseNet121 block demonstrated better performance for slice-level prediction, with an accuracy of 97.1% and AUC of 0.992. For both datasets, the prediction accuracy of our pipeline was 100% at the patient level. CONCLUSIONS: The proposed fully automatic deep learning pipeline of deep learning can distinguish COVID-19 from CAP via CT images rapidly and accurately, thereby accelerating diagnosis and augmenting the performance of radiologists. This pipeline is convenient for use by radiologists and provides explainable predictions.


Subject(s)
COVID-19 , Deep Learning , Pneumonia , Humans , Neural Networks, Computer , Pneumonia/diagnostic imaging , SARS-CoV-2 , Tomography, X-Ray Computed
13.
PLoS One ; 16(3): e0247839, 2021.
Article in English | MEDLINE | ID: covidwho-1574949

ABSTRACT

As SARS-CoV-2 has spread quickly throughout the world, the scientific community has spent major efforts on better understanding the characteristics of the virus and possible means to prevent, diagnose, and treat COVID-19. A valid approach presented in the literature is to develop an image-based method to support COVID-19 diagnosis using convolutional neural networks (CNN). Because the availability of radiological data is rather limited due to the novelty of COVID-19, several methodologies consider reduced datasets, which may be inadequate, biasing the model. Here, we performed an analysis combining six different databases using chest X-ray images from open datasets to distinguish images of infected patients while differentiating COVID-19 and pneumonia from 'no-findings' images. In addition, the performance of models created from fewer databases, which may imperceptibly overestimate their results, is discussed. Two CNN-based architectures were created to process images of different sizes (512 × 512, 768 × 768, 1024 × 1024, and 1536 × 1536). Our best model achieved a balanced accuracy (BA) of 87.7% in predicting one of the three classes ('no-findings', 'COVID-19', and 'pneumonia') and a specific balanced precision of 97.0% for 'COVID-19' class. We also provided binary classification with a precision of 91.0% for detection of sick patients (i.e., with COVID-19 or pneumonia) and 98.4% for COVID-19 detection (i.e., differentiating from 'no-findings' or 'pneumonia'). Indeed, despite we achieved an unrealistic 97.2% BA performance for one specific case, the proposed methodology of using multiple databases achieved better and less inflated results than from models with specific image datasets for training. Thus, this framework is promising for a low-cost, fast, and noninvasive means to support the diagnosis of COVID-19.


Subject(s)
COVID-19/diagnostic imaging , Databases, Factual , Neural Networks, Computer , Pneumonia/diagnostic imaging , Algorithms , Bias , Deep Learning , Humans , Image Interpretation, Computer-Assisted , Radiography, Thoracic
14.
Annu Int Conf IEEE Eng Med Biol Soc ; 2021: 3970-3973, 2021 11.
Article in English | MEDLINE | ID: covidwho-1566234

ABSTRACT

Pneumonia is a common complication associated with COVID-19 infections. Unlike common versions of pneumonia that spread quickly through large lung regions, COVID-19 related pneumonia starts in small localized pockets before spreading over the course of several days. This makes the infection more resilient and with a high probability of developing acute respiratory distress syndrome. Because of the peculiar spread pattern, the use of pulmonary computerized tomography (CT) scans was key in identifying COVID-19 infections. Identifying uncommon pulmonary diseases could be a strong line of defense in early detection of new respiratory infection-causing viruses. In this paper we describe a classification algorithm based on hyperdimensional computing for the detection of COVID-19 pneumonia in CT scans. We test our algorithm using three different datasets. The highest reported accuracy is 95.2% with an F1 score of 0.90, and all three models had a precision of 1 (0 false positives).


Subject(s)
COVID-19 , Pneumonia , Algorithms , Humans , Pneumonia/diagnostic imaging , SARS-CoV-2 , Tomography, X-Ray Computed
15.
Annu Int Conf IEEE Eng Med Biol Soc ; 2021: 3213-3216, 2021 11.
Article in English | MEDLINE | ID: covidwho-1566218

ABSTRACT

The wide spread of coronavirus pneumonia (COVID-19) has been a severe threat to global health since 2019. Apart from the nucleic acid detection, medical imaging examination is a vital diagnostic modality to confirm and treat the disease. Thus, implementing the automatic diagnosis of the COVID-19 bears particular significance. However, the limitations of data quality and size strongly hinder the clas-sification and segmentation performance and it also result in high misdiagnosis rate. To this end, we propose a novel full scale attention mechanism (FUSA) to capture more contextual dependencies of features, which enables the model easier to classify positive cases and improve the sensitivity. Specifically, FUSA parallelly extracts the information of channel domain and spatial domain, and fuses them together. The experimental study shows FUSA can significantly improve the COVID-19 automated diagnosis performance and eliminate false negative cases compared with other state-of-the-art ones.


Subject(s)
COVID-19 , Pneumonia , COVID-19 Testing , Humans , Pneumonia/diagnostic imaging , SARS-CoV-2 , Tomography, X-Ray Computed
16.
Sci Rep ; 11(1): 23210, 2021 12 01.
Article in English | MEDLINE | ID: covidwho-1545637

ABSTRACT

SARS-CoV2 pandemic exposed the limitations of artificial intelligence based medical imaging systems. Earlier in the pandemic, the absence of sufficient training data prevented effective deep learning (DL) solutions for the diagnosis of COVID-19 based on X-Ray data. Here, addressing the lacunae in existing literature and algorithms with the paucity of initial training data; we describe CovBaseAI, an explainable tool using an ensemble of three DL models and an expert decision system (EDS) for COVID-Pneumonia diagnosis, trained entirely on pre-COVID-19 datasets. The performance and explainability of CovBaseAI was primarily validated on two independent datasets. Firstly, 1401 randomly selected CxR from an Indian quarantine center to assess effectiveness in excluding radiological COVID-Pneumonia requiring higher care. Second, curated dataset; 434 RT-PCR positive cases and 471 non-COVID/Normal historical scans, to assess performance in advanced medical settings. CovBaseAI had an accuracy of 87% with a negative predictive value of 98% in the quarantine-center data. However, sensitivity was 0.66-0.90 taking RT-PCR/radiologist opinion as ground truth. This work provides new insights on the usage of EDS with DL methods and the ability of algorithms to confidently predict COVID-Pneumonia while reinforcing the established learning; that benchmarking based on RT-PCR may not serve as reliable ground truth in radiological diagnosis. Such tools can pave the path for multi-modal high throughput detection of COVID-Pneumonia in screening and referral.


Subject(s)
COVID-19/complications , Deep Learning , Expert Systems , Image Processing, Computer-Assisted/methods , Pneumonia/diagnosis , Radiography, Thoracic/methods , Tomography, X-Ray Computed/methods , Algorithms , COVID-19/virology , Humans , Incidence , India/epidemiology , Neural Networks, Computer , Pneumonia/diagnostic imaging , Pneumonia/epidemiology , Pneumonia/virology , Retrospective Studies , SARS-CoV-2/isolation & purification
17.
Sci Rep ; 11(1): 22892, 2021 11 24.
Article in English | MEDLINE | ID: covidwho-1532105

ABSTRACT

Clinical and prognostic differences between symptomatic and asymptomatic older patients with COVID-19 are of great interest since frail patients often show atypical presentation of illness. Lung Ultrasound (LUS) has been proven to be a reliable tool for detecting early-phase COVID-19 pneumonic alterations. The current prospective bicentric study aimed to compare LUS score and 3-month overall mortality between asymptomatic and symptomatic older patients with COVID-19, according to frailty status. Patients were stratified according to LUS score tertiles and Clinical Frailty Scale categories. Survival rate was assessed by telephone interviews 3 months after discharge. 64 symptomatic (24 women, aged 80.0 ± 10.8 years) and 46 asymptomatic (31 women, aged 84.3 ± 8.8 years) were consecutively enrolled. LUS score resulted an independent predictor of 3-month mortality [OR 2.27 (CI95% 1.09-4.8), p = 0.03], and the highest mortality rate was observed in symptomatic and asymptomatic pre-frail and frail patients (70.6% and 66.7%, respectively) with greater LUS abnormalities (3rd tertile). In conclusion, LUS identified an acute interstitial lung involvement in most of the older asymptomatic patients. Mortality rate progressively increased according to clinical frailty and LUS score degree, resulting a reliable prognostic tool in both symptomatic and asymptomatic patients.


Subject(s)
COVID-19/diagnostic imaging , COVID-19/mortality , Pneumonia/diagnostic imaging , Aged , Aged, 80 and over , Asymptomatic Diseases/epidemiology , COVID-19/complications , Female , Hospitalization , Humans , Lung/diagnostic imaging , Lung/pathology , Male , Pneumonia/immunology , Prognosis , Prospective Studies , SARS-CoV-2/metabolism , SARS-CoV-2/pathogenicity , Tomography, X-Ray Computed/methods , Ultrasonography/methods
18.
IEEE Trans Neural Netw Learn Syst ; 33(1): 12-24, 2022 01.
Article in English | MEDLINE | ID: covidwho-1528340

ABSTRACT

The coronavirus disease 2019 (COVID-19) pandemic is spreading worldwide. Considering the limited clinicians and resources and the evidence that computed tomography (CT) analysis can achieve comparable sensitivity, specificity, and accuracy with reverse-transcription polymerase chain reaction, the automatic segmentation of lung infection from CT scans supplies a rapid and effective strategy for COVID-19 diagnosis, treatment, and follow-up. It is challenging because the infection appearance has high intraclass variation and interclass indistinction in CT slices. Therefore, a new context-aware neural network is proposed for lung infection segmentation. Specifically, the autofocus and panorama modules are designed for extracting fine details and semantic knowledge and capturing the long-range dependencies of the context from both peer level and cross level. Also, a novel structure consistency rectification is proposed for calibration by depicting the structural relationship between foreground and background. Experimental results on multiclass and single-class COVID-19 CT images demonstrate the effectiveness of our work. In particular, our method obtains the mean intersection over union (mIoU) score of 64.8%, 65.2%, and 73.8% on three benchmark datasets for COVID-19 infection segmentation.


Subject(s)
COVID-19/diagnostic imaging , Tomography, X-Ray Computed/methods , Algorithms , Benchmarking , Calibration , Diagnosis, Differential , Humans , Image Processing, Computer-Assisted , Neural Networks, Computer , Pneumonia/diagnosis , Pneumonia/diagnostic imaging
19.
PLoS One ; 16(9): e0256630, 2021.
Article in English | MEDLINE | ID: covidwho-1518353

ABSTRACT

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.


Subject(s)
COVID-19/diagnosis , Early Diagnosis , Pneumonia/diagnosis , Thorax/diagnostic imaging , COVID-19/diagnostic imaging , COVID-19/virology , Deep Learning , Humans , Lung/diagnostic imaging , Lung/pathology , Neural Networks, Computer , North America , Pneumonia/diagnostic imaging , SARS-CoV-2/isolation & purification , SARS-CoV-2/pathogenicity , Thorax/pathology , X-Rays
20.
AJR Am J Roentgenol ; 217(5): 1103, 2021 11.
Article in English | MEDLINE | ID: covidwho-1502235
SELECTION OF CITATIONS
SEARCH DETAIL