Skip to main content
SearchLoginLogin or Signup

Diagnosis and Prediction Model for COVID-19 Patient’s Response to Treatment based on Convolutional Neural Networks and Whale Optimization Algorithm Using CT Images (lay summary)

This is a lay summary of the article published under the DOI: 10.1101/2020.04.16.20063990

Published onJun 05, 2023
Diagnosis and Prediction Model for COVID-19 Patient’s Response to Treatment based on Convolutional Neural Networks and Whale Optimization Algorithm Using CT Images (lay summary)
·

Egyptian researchers propose a computer model to help diagnose and treat COVID-19

Researchers created an experimental computer program to try to diagnose early COVID-19 more quickly and with higher accuracy. They also developed a model to predict how well a patient might recover with treatment.

SARS-CoV-2, the respiratory virus that causes COVID-19, spread quickly in 2020 to cause the global COVID-19 pandemic.

The reverse transcription polymerase chain reaction (RT-PCR) test is one of the most accurate tests to diagnose COVID-19. But, it can be inaccurate in the early stages of infection. 

Computed tomography (CT) scans seem to be more accurate in diagnosing COVID-19 early, but the scans need to be analysed by human experts. Since hospitals were often overwhelmed by large numbers of COVID-19 patients, medical staff did not have the resources to treat them all.

The authors of this study wanted to use a technology called “machine learning” to help frontline workers identify COVID-19 patients from CT scans, so that they could use scarce resources to treat patients most likely to recover. 

They wanted to use real COVID-19 patient data to teach a computer program how to diagnose COVID-19 from CT scans, and to predict if a patient will respond well to treatment. They also wanted to increase the speed and accuracy of making COVID-19 diagnoses using CT scans.

These researchers fed their computer program patient data (such as CT scans, RT-PCR test results, and full blood count) so that it could learn to diagnose COVID-19 and recommend treatment, all based on the previous experience of human experts who diagnosed and treated those patients. 

The researchers said their program is up to 97% accurate and can make a prediction from a CT scan in about 20 seconds. 

They also said their program makes more accurate predictions in shorter times than other researchers were able to achieve using different data sets. 

Other research groups have looked into diagnosing COVID-19 using CT scans and machine learning. Unfortunately, in those studies the machine models were not able to explain their diagnosis the way a human expert would; they often simply provide the diagnosis itself. In this study, researchers tried to solve this problem by highlighting what pieces of the CT scan the computer uses to make its diagnosis. 

The researchers planned to test this model on more data to confirm their results. They also planned to make a new model that can predict how ill COVID-19 positive patients are. They say this will help doctors decide how best to treat a patient.

With further research, this Egyptian research group’s idea can help many doctors manage limited resources when treating COVID-19 patients. But, their technique would require a relatively high level of infrastructure, such as CT machines and advanced computers, which may not be available in many areas of Africa.

Abstract

The outbreak of coronavirus diseases (COVID-19) has rabidly spread all over the world. The World Health Organization (WHO) has announced that coronavirus COVID-19 is an international pandemic. The Real-Time Reverse transcription-polymerase Chain Reaction (RT-PCR) has a low positive and sensitivity rate in the early stage of COVID-19. As a result, the Computed Tomography (CT) imaging is used for diagnosing. COVID-19 has different key signs on a CT scan differ from other viral pneumonia. These signs include ground-glass opacities, consolidations, and crazy paving. In this paper, an Artificial Intelli-gence-inspired Model for COVID-19 Diagnosis and Prediction for Patient Response to Treatment (AIMDP) is proposed. AIMDP model has two main functions reflected in two proposed modules, namely, the Diagnosis Module (DM) and Prediction Module (PM). The Diagnosis Module (DM) is proposed for early and accurately detecting the patients with COVID-19 and distinguish it from other viral pneumonias using COVID-19 signs obtained from CT scans. The DM model, uses Convolutional Neural Networks (CNNs) as a Deep learning technique for segmentation, can process hundreds of CT images in seconds to speed up diagnosis of COVID-19 and contribute in its containment. In addition, some countries haven’t the ability to provide all patients with the treatment and intensive care services, so it will be mandatory to give treatment to only responding patients. In this context, the Prediction Module (PM) is proposed for predicting the ability of the patient to respond to treatment based on different factors e.g. age, infection stage, respiratory failure, multi-organ failure and the treatment regimens. PM implement the Whale Optimization Algorithm for selecting the most relevant patient’s features. The experimental results show promising performance for the proposed diagnosing and prediction modules, using a dataset with hundreds of real data and CT images.


Disclaimer

This summary is a free resource intended to make African research and research that affects Africa, more accessible to non-expert global audiences. It was compiled by ScienceLink's team of professional African science communicators as part of the Masakhane MT: Decolonise Science project. ScienceLink has taken every precaution possible during the writing, editing, and fact-checking process to ensure that this summary is easy to read and understand, while accurately reporting on the facts presented in the original research paper. Note, however, that this summary has not been fact-checked or approved by the authors of the original research paper, so this summary should be used as a secondary resource. Therefore, before using, citing or republishing this summary, please verify the information presented with the original authors of the research paper, or email [email protected] for more information.

Connections
1 of 7
A Reply to this Pub
Diagnosis and Prediction Model for COVID-19 Patient’s Response to Treatment based on Convolutional Neural Networks and Whale Optimization Algorithm Using CT Images
Description

AbstractThe outbreak of coronavirus diseases (COVID-19) has rabidly spread all over the world. The World Health Organization (WHO) has announced that coronavirus COVID-19 is an international pandemic. The Real-Time Reverse transcription-polymerase Chain Reaction (RT-PCR) has a low positive and sensitivity rate in the early stage of COVID-19. As a result, the Computed Tomography (CT) imaging is used for diagnosing. COVID-19 has different key signs on a CT scan differ from other viral pneumonia. These signs include ground-glass opacities, consolidations, and crazy paving. In this paper, an Artificial Intelli-gence-inspired Model for COVID-19 Diagnosis and Prediction for Patient Response to Treatment (AIMDP) is proposed. AIMDP model has two main functions reflected in two proposed modules, namely, the Diagnosis Module (DM) and Prediction Module (PM). The Diagnosis Module (DM) is proposed for early and accurately detecting the patients with COVID-19 and distinguish it from other viral pneumonias using COVID-19 signs obtained from CT scans. The DM model, uses Convolutional Neural Networks (CNNs) as a Deep learning technique for segmentation, can process hundreds of CT images in seconds to speed up diagnosis of COVID-19 and contribute in its containment. In addition, some countries haven’t the ability to provide all patients with the treatment and intensive care services, so it will be mandatory to give treatment to only responding patients. In this context, the Prediction Module (PM) is proposed for predicting the ability of the patient to respond to treatment based on different factors e.g. age, infection stage, respiratory failure, multi-organ failure and the treatment regimens. PM implement the Whale Optimization Algorithm for selecting the most relevant patient’s features. The experimental results show promising performance for the proposed diagnosing and prediction modules, using a dataset with hundreds of real data and CT images.

Comments
0
comment
No comments here
Why not start the discussion?