Possibly, and due to poor eating habits and unhealthy lifestyle, many viruses are transmitted to human people. Such is the case, of the novel coronavirus SARS-Cov-2, which has expanded of exponential way, practically, to whole world population. For this reason, the enhancement of real microscopic images of this coronavirus is of great importance. Of this way, one can highlight the S-spikes and visualizing those areas that show a high density, which are related to active zones of viral germination and major spread of the virus. The SARS-Cov-2 images were captured from nasopharyngeal samples of Cuban symptomatic individuals (RT-PCR positives for SARS-CoV-2) and processed via scanning electron microscopy. However, many times these microscopic images present some blurring problems, and the S-spikes do not look well defined. Therefore, the aim of this work is to propose new computational methods to carry out enhancement and segmentation of SARS-Cov-2 high-resolution microscopic images. The proposed strategy obtained very satisfactory results, and we validated its performance, together with specialist physicians, on a set of 1005 images. Due to the importance of the obtained results, this first work will be addressed to the application of the proposed algorithm. A second paper will deeply analyze the theory related to these algorithms.
Keyphrases
- sars cov
- deep learning
- convolutional neural network
- high resolution
- respiratory syndrome coronavirus
- optical coherence tomography
- machine learning
- electron microscopy
- physical activity
- primary care
- mental health
- endothelial cells
- mass spectrometry
- metabolic syndrome
- palliative care
- type diabetes
- coronavirus disease
- drug induced
- single molecule
- tandem mass spectrometry
- single cell