Related Experiment Video
Updated: Jul 27, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.2K
Remora Namib Beetle Optimization Enabled Deep Learning for Severity of COVID-19 Lung Infection Identification and
Amgothu Shanthi1, Srinivas Koppu1
1School of Information Technology and Engineering, Vellore Institute of Technology, Vellore 632014, India.
Sensors (Basel, Switzerland)
|June 10, 2023
Summary
New methods, Remora Namib Beetle Optimization_ Deep Quantum Neural Network (RNBO_DQNN) and RNBO_Deep Neuro Fuzzy Network (RNBO_DNFN), improve COVID-19 detection from lung CT scans. These techniques enhance the accuracy of identifying infectious lung tissues for better patient treatment.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Coronavirus disease 2019 (COVID-19) poses a significant global health challenge.
- Accurate and rapid detection of COVID-19 from medical imaging is crucial for effective patient management.
- Identifying and segmenting infectious lung tissues in CT images presents considerable challenges.
Purpose of the Study:
- To introduce novel optimization and classification techniques for automated COVID-19 detection in lung CT images.
- To enhance the accuracy and efficiency of identifying and classifying COVID-19 related lung infections.
- To improve diagnostic capabilities for COVID-19 through advanced artificial intelligence models.
Main Methods:
- Pre-processing of lung CT images using an adaptive Wiener filter.
- Lung lobe segmentation utilizing the Pyramid Scene Parsing Network (PSP-Net).
- Development and application of Remora Namib Beetle Optimization_ Deep Quantum Neural Network (RNBO_DQNN) and RNBO_Deep Neuro Fuzzy Network (RNBO_DNFN) for classification, with RNBO merging Remora Optimization Algorithm (ROA) and Namib Beetle Optimization (NBO).
Main Results:
- The RNBO_DQNN and RNBO_DNFN models were employed for the identification and classification of COVID-19 lung infections.
- The RNBO_DNFN model achieved maximum testing accuracy.
- Specific performance metrics included True Negative Rate (TNR) of 89.4%, True Positive Rate (TPR) of 89.5%, and another metric of 87.5%.
Conclusions:
- The proposed RNBO_DQNN and RNBO_DNFN techniques demonstrate high potential for accurate COVID-19 detection from lung CT images.
- These advanced methods can aid in the early identification and classification of infectious lung tissues, supporting clinical decision-making.
- The study highlights the effectiveness of combining novel optimization algorithms with deep learning networks for medical image analysis in the context of infectious diseases.
