A Fully Automatic Deep Learning System for COVID-19 Diagnostic and Prognostic Analysis
read more
Citations
Prediction models for diagnosis and prognosis of covid-19: systematic review and critical appraisal
The Role of Imaging in the Detection and Management of COVID-19: A Review
Adoption of Digital Technologies in Health Care During the COVID-19 Pandemic: Systematic Review of Early Scientific Literature.
Leveraging Data Science to Combat COVID-19: A Comprehensive Review
The ensemble deep learning model for novel COVID-19 on CT images.
References
Densely Connected Convolutional Networks
Feature Pyramid Networks for Object Detection
Clinical Characteristics of 138 Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China.
Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia.
Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study.
Related Papers (5)
Frequently Asked Questions (19)
Q2. What future works have the authors mentioned in the paper "A fully automatic deep learning system for covid-19 diagnostic and prognostic analysis" ?
To avoid time-consuming lesion annotation by radiologists, automatic lesion segmentation models [ 17, 24 ] were used in further studies. In the future, the authors will use a generative adversarial network to convert CT images of different slice thickness into CT images with a unified slice thickness, which may further improve the diagnostic performance of the DL system.
Q3. What are the main computational formulas in the DL model?
The main computational formulas are convolution, pooling, activation and batch normalisation as defined in supplementary methods S3.
Q4. What were the main characteristics of the DL system?
Area under the receiver operating characteristic (ROC) curve, accuracy, sensitivity, specificity, F1-score, calibration curves and Hosmer-Lemeshow test were used to assess the performance of the DL system in diagnosing COVID-19.
Q5. What is the proposed DL system for COVID-19?
The proposed DL system includes three parts: automatic lung segmentation, non-lung area suppression, and COVID-19 diagnostic and prognostic analysis.
Q6. What is the role of RT-PCR in COVID-19?
fast diagnosis and finding high-risk patients with worse prognosis are very helpful for the control and management of COVID-19.
Q7. What is the common type of COVID-19?
For patients with COVID-19, bilateral lung lesions consisting of ground-glass opacities were frequently observed in CT images [6–8].
Q8. How many patients were included in the COVID-19 dataset?
In the COVID-19 dataset, 1266 patients were finally included who met the following inclusion criteria: 1) RT-PCR confirmed COVID-19; 2) laboratory confirmed other types of pneumonia before December 2019; 3) have non-contrast enhanced chest CT at diagnosis time.
Q9. What is the main reason for using a chest CT dataset for auxiliary training?
using a chest CT dataset for auxiliary training (pre-training) enables the DL model learn features that are more specific to chest CT images.
Q10. Why did WANG et al. use the 3-dimensional bounding box of lung as?
Since lesions can be distributed in many locations in lungs, and automatic lesion segmentation may not guarantee very high precision.
Q11. How many patients were included in the CT-EGFR dataset?
In the CT-EGFR dataset, 4106 patients with lung cancer were finally included who met the following criteria: 1) EGFR gene sequencing was obtained; and 2) non-contrast enhanced chest CT data obtained within 4 weeks before EGFR gene sequencing.
Q12. What is the DL model used to predict the probability of the input patient being infected?
After an iterative training process in the COVID-19 dataset (supplementary methods S5), the COVID-19Net can predict the probability of the input patient being infected with COVID-19; this probability was defined as DL score in this study.
Q13. What was the first step in the transfer learning process?
In the second training process, the authors transferred the pre-trained COVID-19Net to the COVID-19 dataset to specifically mine lung characteristics associated with COVID-19.
Q14. What was the auxiliary training process used to train the COVID-19Net?
In this auxiliary training process, the authors trained the COVID-19Net to predict EGFR mutation status (EGFR-mutant or EGFR wild-type) using the lung-ROI [11].
Q15. What is the DL system's role in COVID-19?
Through training in this large CT-EGFR dataset, the DL system learned hierarchical lung features that can reflect the associations between chest CT image and micro-level lung functional abnormality.
Q16. What is the important component of the DL system?
Suspicious lung area discovered by the DL system Through the DL visualisation algorithm [22, 23], the authors are able to visualise the lung area that draws most attention to the DL system.
Q17. What is the purpose of this study?
In this study, the authors proposed a novel fully automatic DL system using raw chest CT image to help COVID-19 diagnostic and prognostic analysis.
Q18. What is the description of the DL system?
The good diagnostic and prognostic performance of the DL system illustrates that DL could be helpful in the epidemic control of COVID-19 without adding much cost.
Q19. What is the way to diagnose COVID-19?
In recent studies, radiological findings demonstrated that computed tomography (CT) has great diagnostic and prognostic value for COVID-19.