A novel multiple instance learning framework for COVID-19 severity assessment via data augmentation and self-supervised learning

Li, Zekun; Zhao, Wei; Shi, Feng; Qi, Lei; Xie, Xingzhi; Wei, Ying; Ding, Zhongxiang; Gao, Yang; Wu, Shangjie; Liu, Jun; Shi, Yinghuan; Shen, Dinggang

doi:10.1016/j.media.2021.101978

Cited by 47 publications

(40 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although a recently automatic diagnosis of COVID-19 pneumonia has been introduced with the deep learning methods, there are still many unsolved challenges in the automatic and analysis of CT images related to this epidemic. The lack of public and large datasets is an important challenge for this application so that a very recent approach used data augmentation 35 or applied GAN 12 , or its extension, Cycle-GAN (CGAN) 13 to reduce the gap of needs to the lots of data. Also, the main problem of existing public datasets is that all taken images are labeled based on the patient.…”

Section: Discussion and Future Workmentioning

confidence: 99%

COVID-19 early detection for imbalanced or low number of data using a regularized cost-sensitive CapsNet

Javidi

Abbaasi

Atashi

et al. 2021

Sci Rep

View full text Add to dashboard Cite

With the presence of novel coronavirus disease at the end of 2019, several approaches were proposed to help physicians detect the disease, such as using deep learning to recognize lung involvement based on the pattern of pneumonia. These approaches rely on analyzing the CT images and exploring the COVID-19 pathologies in the lung. Most of the successful methods are based on the deep learning technique, which is state-of-the-art. Nevertheless, the big drawback of the deep approaches is their need for many samples, which is not always possible. This work proposes a combined deep architecture that benefits both employed architectures of DenseNet and CapsNet. To more generalize the deep model, we propose a regularization term with much fewer parameters. The network convergence significantly improved, especially when the number of training data is small. We also propose a novel Cost-sensitive loss function for imbalanced data that makes our model feasible for the condition with a limited number of positive data. Our novelties make our approach more intelligent and potent in real-world situations with imbalanced data, popular in hospitals. We analyzed our approach on two publicly available datasets, HUST and COVID-CT, with different protocols. In the first protocol of HUST, we followed the original paper setup and outperformed it. With the second protocol of HUST, we show our approach superiority concerning imbalanced data. Finally, with three different validations of the COVID-CT, we provide evaluations in the presence of a low number of data along with a comparison with state-of-the-art.

show abstract

Section: Discussion and Future Workmentioning

confidence: 99%

COVID-19 early detection for imbalanced or low number of data using a regularized cost-sensitive CapsNet

Javidi

Abbaasi

Atashi

et al. 2021

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Multiple informatic approaches have been applied to the integration of population level, molecular and imaging data associated with SARS-CoV-2 infection. These include conventional interactome networks, associations with abundance trajectories with viral proteins, neural networks, deep and self-supervised learning, as well as machine learning of digital data [8,10,[16][17][18][19][20]. Each type of informatic approach has its advantages and disadvantages based on a priori knowledge, power and structure of the dataset, as well as the ability to develop de novo knowledge.…”

Section: Discussionmentioning

confidence: 99%

Bayesian Model Infers Drug Repurposing Candidates for Treatment of COVID-19

et al. 2021

View full text Add to dashboard Cite

The emergence of COVID-19 progressed into a global pandemic that has functionally put the world at a standstill and catapulted major healthcare systems into an overburdened state. The dire need for therapeutic strategies to mitigate and successfully treat COVID-19 is now a public health crisis with national security implications for many countries. The current study employed Bayesian networks to a longitudinal proteomic dataset generated from Caco-2 cells transfected with SARS-CoV-2 (isolated from patients returning from Wuhan to Frankfurt). Two different approaches were employed to assess the Bayesian models, a titer-center topology analysis and a drug signature enrichment analysis. Topology analysis identified a set of proteins directly linked to the SAR-CoV2 titer, including ACE2, a SARS-CoV-2 binding receptor, MAOB and CHECK1. Aligning with the topology analysis, MAOB and CHECK1 were also identified within the enriched drug-signatures. Taken together, the data output from this network has identified nodal host proteins that may be connected to 18 chemical compounds, some already marketed, which provides an immediate opportunity to rapidly triage these assets for safety and efficacy against COVID-19.

show abstract

“…Zhang et al ( Zhang, et al, 2020 ) used Light Gradient Boosting Machine (LightGBM) and Cox proportional-hazards (CoxPH) regression models based on quantitative CT lung-lesion features and clinical parameters to assess clinical stages (severe, non-severe). Li et al ( Li, et al, 2021 ) proposed a deep multiple instance learning (MIL) model with instance-level attention to jointly classify the bag and weigh the instances in each bag, so as to distinguish the severe instances from non-severe instances. He et al (Kelei He, et al, 2021 ) also proposed a synergistic learning framework for severity assessment (severe, non-severe) in 3D CT images.…”

Section: Related Workmentioning

confidence: 99%

“…Given that manual severity assessment could be a labour-intensive work for front-line healthcare workers, providing computer-aided clinical support for automatic severity assessment is highly desired. Several computer-aided methods ( Aboutalebi et al, 2021 , Ali and Budka, 2021 ; Kelei He et al, 2021 , Lessmann et al, 2021 , Li et al, 2021 , Lizzi et al, 2021 , Zhang et al, 2020 ) have been proposed recently. However, these studies have some obstacles.…”

Section: Introductionmentioning

confidence: 99%

A deep-learning-based framework for severity assessment of COVID-19 with CT images

Zhao

Chen

et al. 2021

Expert Systems with Applications

Self Cite

View full text Add to dashboard Cite

Millions of positive COVID-19 patients are suffering from the pandemic around the world, a critical step in the management and treatment is severity assessment, which is quite challenging with the limited medical resources. Currently, several artificial intelligence systems have been developed for the severity assessment. However, imprecise severity assessment and insufficient data are still obstacles. To address these issues, we proposed a novel deep-learning-based framework for the fine-grained severity assessment using 3D CT scans, by jointly performing lung segmentation and lesion segmentation. The main innovations in the proposed framework include: 1) decomposing 3D CT scan into multi-view slices for reducing the complexity of 3D model, 2) integrating prior knowledge (dual-Siamese channels and clinical metadata) into our model for improving the model performance. We evaluated the proposed method on 1301 CT scans of 449 COVID-19 cases collected by us, our method achieved an accuracy of 86.7% for four-way classification, with the sensitivities of 92%, 78%, 95%, 89% for four stages. Moreover, ablation study demonstrated the effectiveness of the major components in our model. This indicates that our method may contribute a potential solution to severity assessment of COVID-19 patients using CT images and clinical metadata.

show abstract

A novel multiple instance learning framework for COVID-19 severity assessment via data augmentation and self-supervised learning

Abstract: Highlights A novel weakly supervised learning framework for COVID-19 severity assessment A multiple instance learning model with virtual bag-based augmentation A novel self-supervised pretext task to aid the learning process

Cited by 47 publications

References 39 publications

COVID-19 early detection for imbalanced or low number of data using a regularized cost-sensitive CapsNet

COVID-19 early detection for imbalanced or low number of data using a regularized cost-sensitive CapsNet

Bayesian Model Infers Drug Repurposing Candidates for Treatment of COVID-19

A deep-learning-based framework for severity assessment of COVID-19 with CT images

Contact Info

Product

Resources

About