Table 1.

Chronological Roots of AI in Healthcare.

Year	Root	Summary
1950s	Expert Systems	Expert systems were among the earliest roots of AI in healthcare. These systems aimed to capture the knowledge and expertise of human experts in specific domains, including healthcare. By codifying expert knowledge into rules and algorithms, expert systems could provide diagnostic and decision-making support, aiding healthcare professionals in making accurate and timely assessments and recommendations.
1960s	Machine Learning	Machine learning emerged as a foundational root of AI in healthcare in the 1960s. Machine-learning algorithms enabled computers to learn from data and improve their performance without explicit programming. In healthcare, machine-learning techniques have been used for tasks, such as pattern recognition, classification, and prediction. Machine-learning models can analyze large volumes of patient data and extract valuable insights, contributing to personalized medicine and clinical decision making.
1970s	Natural Language Processing	Natural language processing (NLP) has its roots in the 1970s, focusing on enabling computers to understand and interact with human language. In healthcare, NLP techniques have been utilized to extract information from clinical narratives, electronic health records (EHRs), medical literature, and patient-generated data. NLP has facilitated information extraction, sentiment analysis, clinical coding, and the development of conversational agents for healthcare applications.
1980s	Image Analysis	Image analysis became a significant root of AI in healthcare in the 1980s. Computer vision and image processing techniques were applied to medical imaging modalities, such as X-rays, CT scans, MRIs, and pathology slides, enabling the automated interpretation, segmentation, and detection of abnormalities. AI algorithms have enhanced medical imaging analysis, aiding in early disease detection, diagnosis, and treatment planning in fields such as radiology and pathology.
1990s	Robotics	Robotics started making an impact on healthcare in the 1990s, combining AI with mechanical devices to perform various medical tasks. Robotic systems have been developed for surgical procedures, rehabilitation, assistive care, and remote telemedicine applications. By incorporating AI algorithms, robotic systems can enhance precision, dexterity, and automation in healthcare, leading to improved outcomes, reduced invasiveness, and increased accessibility to medical services.
1990s	Data Mining	Data mining, or knowledge discovery from databases, became a key root of AI in healthcare in the 1990s. With the growth of electronic health records and the accumulation of vast numbers of healthcare data, data-mining techniques were applied to uncover hidden patterns, relationships, and insights. Data mining has contributed to population health management, disease surveillance, predictive modeling, and the identification of risk factors in healthcare.
1990s	Decision Support Systems	Decision support systems (DSS) emerged as a root of AI in healthcare in the 1990s, aiming to assist healthcare professionals in making informed decisions. DSSs incorporate AI techniques, such as rule-based systems, probabilistic models, and machine learning, to provide evidence-based recommendations, clinical guidelines, and alerts. DSSs have facilitated diagnosis, treatment planning, medication management, and improved patient safety in healthcare settings.
2000s	Knowledge Representation	Knowledge representation, which involves capturing and organizing knowledge in a structured format, has been a fundamental root of AI in healthcare. Various knowledge-representation techniques, such as ontologies, semantic networks, and knowledge graphs, have been applied to represent medical knowledge, clinical guidelines, and domain-specific information.