Classification of COVID-19 by using supervised optimized machine learning technique.

Area of Science:

• Medical informatics
• Computational biology
• Machine learning in healthcare

Background:

• COVID-19 poses a significant global health threat with high mortality rates.
• Early identification of COVID-19 symptoms is crucial for effective patient management and reducing mortality.
• Accurate classification of COVID-19 cases remains a challenge for researchers and clinicians.

Purpose of the Study:

• To develop and evaluate a machine learning model for accurate COVID-19 classification.
• To improve the classification performance by employing feature selection and hyperparameter optimization techniques.

Main Methods:

• Utilized the Support Vector Machine (SVM) classifier for disease classification.
• Implemented a hybrid feature selection technique: Minimum Redundancy Maximum Relevance (mRMR) algorithm.
• Optimized classifier performance using a modified cuckoo search algorithm for hyperparameter tuning.

Main Results:

• The proposed model effectively classifies between COVID-19 and normal cases.
• Performance analysis using various metrics demonstrates the efficacy of the hybrid approach.
• Feature selection and hyperparameter optimization significantly enhanced classification accuracy.

Conclusions:

• The integrated machine learning approach offers a promising tool for accurate COVID-19 detection.
• This methodology can aid in early diagnosis, potentially reducing disease severity and mortality.
• Further research can explore broader applications of this technique in medical diagnostics.