Hybrid model for early identification post-Covid-19 sequelae.

Area of Science:

• * Computer Science
• * Medical Informatics
• * Public Health

Background:

• * COVID-19 diagnosis presents challenges due to varied symptom presentation, with up to 80% of cases being asymptomatic or mild.
• * Severe cases require intensive care, highlighting the need for accurate and timely diagnosis.
• * Traditional diagnostic methods can be enhanced by leveraging advanced data analysis techniques.

Purpose of the Study:

• * To optimize the diagnostic process for SARS-CoV-2 (COVID-19) using Artificial Intelligence (AI).
• * To develop and evaluate a hybrid AI model combining Machine Learning and multicriteria decision support methods.
• * To improve the accuracy of COVID-19 diagnosis through enhanced clinical observation and data analysis.

Main Methods:

• * Application of AI techniques including Machine Learning algorithms, Neural Networks, and Naïve Bayes.
• * Development of a hybrid model integrating AI with multicriteria decision support (Verbal Decision Analysis methodology).
• * Analysis of health professional data for attribute engineering and knowledge discovery.

Main Results:

• * AI techniques demonstrate significant potential in optimizing the analysis of patient data for COVID-19 diagnosis.
• * The proposed hybrid model aims to enhance diagnostic accuracy by exploring the predictive power of specific data points.
• * Attribute engineering effectively addresses the reduced number of observable signs and symptoms in COVID-19 patients.

Conclusions:

• * AI and Machine Learning offer powerful tools for improving the diagnostic accuracy of COVID-19.
• * The hybrid AI approach shows promise for more precise clinical diagnosis, aiding healthcare professionals.
• * Further research into AI-driven diagnostic models is crucial for managing infectious diseases like COVID-19.