CDSS for Early Recognition of Respiratory Diseases based on AI Techniques: A Systematic Review

• Published in: Wireless personal communications 2023-07, Vol.131 (2), p.739-761
• Main Authors: Ali, Syed Waqad, Asif, Muhammad, Zia, Muhammad Yousuf Irfan, Rashid, Munaf, Syed, Sidra Abid, Nava, Enrique
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Asthma; Communications Engineering; Computer Communication Networks; Decision support systems; Decision trees; Engineering; Health services; Machine learning; Medical research; Networks; Respiratory diseases; Signal,Image and Speech Processing; Signs and symptoms; Support vector machines
• ISSN: 0929-6212; 1572-834X
• DOI: 10.1007/s11277-023-10432-1

Respiratory diseases such as Asthma, COVID-19, etc., require preventive and precautionary measures. Due to the lack of medical treatment for the masses, researchers are currently focusing on clinical decision support systems (CDSS). CDSS for respiratory diseases utilizes Machine Learning (ML) techniques to classify the symptoms into a possible diseases. This approach not only grasps the attention of researchers worldwide but also assists medical doctors in the early diagnosis of the disease. In this review paper, PRISMA guidelines are used to conduct a detailed overview of the early detection of respiratory diseases using ML techniques are identified. Among various ML techniques, Artificial Neural Networks (ANN), Support Vector Machine (SVM), Decision Tree (D-Tree), Logistic Regression (LR), K Nearest Neighbor (KNN), Random Forest (RF), and AdaBoost are discussed. Then respiratory diseases are identified whose CDSS are available with the ML techniques and possible future direction for its improvement. Furthermore, the tools and ML techniques are compared with each other to enhance the researcher’s clarity for future use. The paper concluded with the future direction of the ML in the successful implementation of the CDSS in the field of respiratory disease.