A Hybrid Deep Neural Approach for Segmenting the COVID Affection Area from the Lungs X-Ray Images

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• COVID-19 severity prediction is crucial for patient management.
• Image processing of lung X-rays aids in COVID-19 severity identification.
• Traditional neural networks struggle with accuracy due to image complexity.

Purpose of the Study:

• To develop a novel Chimp-based Adaboost Severity Analysis (CbASA) model.
• To accurately predict COVID-19 severity using lung X-ray images.
• To improve classification accuracy compared to existing methods.

Main Methods:

• Implemented a novel CbASA model in MATLAB.
• Utilized lung X-ray images for model testing.
• Preprocessed images to remove noise in the initial hidden layer.
• Performed feature extraction, segmentation, and severity classification.

Main Results:

• The CbASA model effectively removed noise from X-ray images.
• Feature extraction, segmentation, and classification were successfully performed.
• The CbASA model achieved superior classification outcomes compared to other models.

Conclusions:

• The novel CbASA model demonstrates high accuracy in COVID-19 severity prediction.
• CbASA offers an effective approach for analyzing lung X-ray images for disease severity.
• This method shows promise for improving diagnostic capabilities in medical research.