Deep Regression via Multi-Channel Multi-Modal Learning for Pneumonia Screening

Wang, Qiuli; Yang, Dan; Li, Zhihuan; Zhang, Xiaohong; Liu, Chen

doi:10.1109/access.2020.2990423

Cited by 24 publications

(15 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The average score of multiple lung regions was utilized for the classification scheme, achieving 88.9% accuracy, 85.3% sensitivity, and 90.1% specificity. Wang et al [ 21 ] introduced a deep regression framework for automatic pneumonia identification by jointly learning from CT scan images and clinical information (i.e., age, gender, and clinical complaints). Recurrent Neural Network (RNN) with ResNet50 as the backbone was used to extract visual features from CT images.…”

Section: Introductionmentioning

confidence: 99%

Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

Detecting COVID-19 at an early stage is essential to reduce the mortality risk of the patients. In this study, a cascaded system is proposed to segment the lung, detect, localize, and quantify COVID-19 infections from computed tomography images. An extensive set of experiments were performed using Encoder–Decoder Convolutional Neural Networks (ED-CNNs), UNet, and Feature Pyramid Network (FPN), with different backbone (encoder) structures using the variants of DenseNet and ResNet. The conducted experiments for lung region segmentation showed a Dice Similarity Coefficient (DSC) of 97.19% and Intersection over Union (IoU) of 95.10% using U-Net model with the DenseNet 161 encoder. Furthermore, the proposed system achieved an elegant performance for COVID-19 infection segmentation with a DSC of 94.13% and IoU of 91.85% using the FPN with DenseNet201 encoder. The proposed system can reliably localize infections of various shapes and sizes, especially small infection regions, which are rarely considered in recent studies. Moreover, the proposed system achieved high COVID-19 detection performance with 99.64% sensitivity and 98.72% specificity. Finally, the system was able to discriminate between different severity levels of COVID-19 infection over a dataset of 1110 subjects with sensitivity values of 98.3%, 71.2%, 77.8%, and 100% for mild, moderate, severe, and critical, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…1 ) 15 . The details of this deep learning model have been described in our previous articles 15 , 16 . Three functional modules were included in this model: lung segmentation, pneumonia lesion segmentation, and quantitative analysis.…”

Section: Methodsmentioning

confidence: 99%

Multidimensional Evaluation of All-Cause Mortality Risk and Survival Analysis for Hospitalized Patients with COVID-19

Li¹,

Luo²,

Deng³

et al. 2021

Int. J. Med. Sci.

Self Cite

View full text Add to dashboard Cite

Background: Coronavirus disease 2019 (COVID-19) has caused over 3.8 million deaths globally. Up to date, the number of death in 2021 is more than that in 2020 globally. Here, we aimed to compare clinical characteristics of deceased patients and recovered patients, and analyze the risk factors of death to help reduce mortality of COVID-19. Methods: In this retrospective study, a total of 2719 COVID-19 patients were enrolled, including 109 deceased patients and 2610 recovered patients. Medical records of all patients were collected between February 4, 2020, and April 7, 2020. Clinical characteristics, laboratory indices, treatments, and deep-learning system-assessed lung lesion volumes were analyzed. The effect of different medications on survival time of fatal cases was also investigated. Results: The deceased patients were older (73 years versus 60 years) and had a male predominance. Nausea (10.1% versus 4.1%) and dyspnea (54.1% versus 39.2%) were more common in deceased patients. The proportion of patients with comorbidities in deceased patients was significantly higher than those in recovered patients. The median times from hospital admission to outcome in deceased patients and recovered patients were 9 days and 13 days, respectively. Patients with severe or critical COVID-19 were more frequent in deceased group. Leukocytosis (11.35×10 9 /L versus 5.60×10 9 /L) and lymphocytopenia (0.52×10 9 /L versus 1.58×10 9 /L) were shown in patients who died. The level of prothrombin time, activated partial prothrombin time, D-dimer, aspartate aminotransferase, alanine aminotransferase, urea, creatinine, creatine kinase, glucose, brain natriuretic peptide, and inflammatory indicators were significantly higher in deceased patients than in recovered patients. The volumes of ground-glass, consolidation, total lesions and total lung in all patients were quantified. Complications were more common in deceased patients than in recovered patients; respiratory failure (57.8%), septic shock (36.7%), and acute respiratory distress syndrome (26.6%) were the most common complications in patients who died. Many treatments were more frequent in deceased patients, such as antibiotic therapy (88.1% versus 53.7%), glucocorticoid treatment (70.6% versus 11.0%), intravenous immunoglobin treatment (36.6% versus 4.9%), invasive mechanical ventilation (62.3% versus 3.8%). Antivirals, antibiotics, traditional Chinese medicines and glucocorticoid treatment may significantly increase the survival time of fatal cases. Quantitative computed tomography imaging results were correlated with biochemical markers.

show abstract

“…This section consists of existing state-of-the-art transfer learning classifiers such as Vgg-16, Vgg-19, AlexNet, ResNet-50, and inception v3, which were applied to achieve the clinical purpose of classification of healthy and COVID-19-infected radiographs. All of these classifiers were trained on the ImageNet (ILSVR) database, which contains thousands of variant kinds of objects used to train and measure the classification performance of the models [48]. Vgg-16 [50] is an open-source framework and is mainly used in a different variety of research contexts [51].…”

Section: Transfer Learning Classifiersmentioning

confidence: 99%

BDCNet: multi-classification convolutional neural network model for classification of COVID-19, pneumonia, and lung cancer from chest radiographs

2022

View full text Add to dashboard Cite

Globally, coronavirus disease (COVID-19) has badly affected the medical system and economy. Sometimes, the deadly COVID-19 has the same symptoms as other chest diseases such as pneumonia and lungs cancer and can mislead the doctors in diagnosing coronavirus. Frontline doctors and researchers are working assiduously in finding the rapid and automatic process for the detection of COVID-19 at the initial stage, to save human lives. However, the clinical diagnosis of COVID-19 is highly subjective and variable. The objective of this study is to implement a multi-classification algorithm based on deep learning (DL) model for identifying the COVID-19, pneumonia, and lung cancer diseases from chest radiographs. In the present study, we have proposed a model with the combination of Vgg-19 and convolutional neural networks (CNN) named BDCNet and applied it on different publically available benchmark databases to diagnose the COVID-19 and other chest tract diseases. To the best of our knowledge, this is the first study to diagnose the three chest diseases in a single deep learning model. We also computed and compared the classification accuracy of our proposed model with four well-known pre-trained models such as ResNet-50, Vgg-16, Vgg-19, and inception v3. Our proposed model achieved an AUC of 0.9833 (with an accuracy of 99.10%, a recall of 98.31%, a precision of 99.9%, and an f1-score of 99.09%) in classifying the different chest diseases. Moreover, CNN-based pre-trained models VGG-16, VGG-19, ResNet-50, and Inception-v3 achieved an accuracy of classifying multi-diseases are 97.35%, 97.14%, 97.15%, and 95.10%, respectively. The results revealed that our proposed model produced a remarkable performance as compared to its competitor approaches, thus providing significant assistance to diagnostic radiographers and health experts.

show abstract

Deep Regression via Multi-Channel Multi-Modal Learning for Pneumonia Screening

Cited by 24 publications

References 29 publications

Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning

Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning

Multidimensional Evaluation of All-Cause Mortality Risk and Survival Analysis for Hospitalized Patients with COVID-19

BDCNet: multi-classification convolutional neural network model for classification of COVID-19, pneumonia, and lung cancer from chest radiographs

Contact Info

Product

Resources

About