Intelligent system for COVID-19 prognosis: a state-of-the-art survey.

Area of Science:

• Public Health
• Epidemiology
• Artificial Intelligence

Background:

• The 21st century faces global disturbances, with COVID-19 declared a Public Health crisis by the WHO.
• Traditional statistical and epidemiological models are used for COVID-19 prediction, but face challenges.
• Inadequacy of medical tests hinders effective COVID-19 control and diagnosis.

Purpose of the Study:

• To explore the application of intelligent systems, specifically Machine Learning (ML) and Deep Learning (DL), in addressing COVID-19 outbreak issues.
• To understand the significance and impact of ML and DL in COVID-19 prognosis and diagnosis.
• To analyze data challenges and future research directions in ML-driven COVID-19 prognosis.

Main Methods:

• Review of intelligent systems, including Machine Learning (ML) and Deep Learning (DL), for COVID-19.
• Analysis of the applicability and effectiveness of these methods in disease diagnosis and prediction.
• Examination of data types, data processing challenges, and advanced ML methods for COVID-19 prognosis.

Main Results:

• ML and DL demonstrate suitability for identifying patterns in complex datasets, offering effective solutions for COVID-19 diagnosis.
• Intelligent systems provide a partial resolution to challenges posed by inadequate medical testing.
• The study highlights the growing development of ML methods for COVID-19 prognosis.

Conclusions:

• Intelligent systems like ML and DL are crucial for managing the COVID-19 pandemic, improving diagnostic accuracy, and predicting disease spread.
• Addressing data-related challenges is essential for enhancing the performance of ML models in public health.
• Further research into ML and DL applications is vital for innovating solutions in healthcare and other impacted sectors.