CNN is a strong utility that is used for classifying images. Built on layers organized by three dimensions. The neurons of a specific layer do not attach to a full set of neurons to the later layer. The output is a diminished vector of probability scores to produce a model of COVID-19 classification.

University of California, Santa Cruz

Sensitivity and specificity of CT scans varies with COVID-19, therefore cannot be the determinant factor in diagnosing infection. Produce a medical image using multiple X-ray measurements from different angles of the patient's chest.

CT Scans

Singh, D., Kumar, V., Vaishali, & Kaur, M. (2020). Classification of COVID-19 patients from chest CT images using multi-objective differential evolution–based convolutional neural networks. European Journal of Clinical Microbiology & Infectious Diseases, 39(7), 1379-1389. doi:10.1007/s10096-020-03901-z

Classification of COVID-19 patients from chest CT images using multi-objective differential evolution–based convolutional neural networks

- Chest CT Imaging Features of COVID-19 Patients from Study #1

CT Imaging Features of COVID-19 Patients

Unlike pathogenic laboratory testing which produces a significant amount of false-negatives while still being a time-consuming method, Machine Learning models  can display radio-logical and image patterns via CT chest scans. 3D CT volume can detect the virus's transmission by using a lung-region 3D CT scan which is segmented and fed into a 3D deep neural network to predict the infection rate.

Large-Scale Analysis of CT Scans: 

Coronavirus pneumonia tends to affect the entire lung as opposed to other viral pneumonia affecting small parts. As per respiratory clinicians and radiologists, the 3 features  of coronavirus pneumonia are ground-glass opacities, consolidation, and pleural effusion seen in the chest of the CT scans.

COVID-19 pneumonia vs. other viral pneumonia

Convolutional neural networks

- 103 patients with lung CTs were chosen for the study
- The CT scans from 107 healthy patients were taken (control group)
- Figures show the relationship between the volume of blood vessels and its cross-sectional area, where the area of a circle is measured by taking a slice of the blood vessel
- The volume of blood vessels in healthy patients was lower compared to COVID-19 patients
- However, there is a slight increase in blood vessel volume in healthy patients when the cross section area of the blood vessels is between ~1.25 and 3.5 mm^2
- The increase in blood vessel volume in COVID-19 patients could be due to inflammation in the blood vessels to increase oxygen uptake by the lungs


Learn Before

Related