Artificial Intelligence in Healthcare: Historical Development, Benefits and Increasing Access for Underserved Populations

The history of AI in healthcare can be traced back to the mid-20th century when researchers began exploring the potential of computer systems to mimic human intelligence. Early efforts focused on developing algorithms that could assist in medical diagnosis and treatment planning [1].

1960s-1970s: Early beginnings: One of the earliest AI systems, MYCIN, was developed in the 1970s at Stanford University. MYCIN was designed to diagnose bacterial infections and recommend antibiotics. It utilized rule-based expert systems, which laid the groundwork for future AI applications in healthcare [2].

1980s-1990s: Growth and development: The 1980s saw the proliferation of expert systems in healthcare. These systems used a set of rules to mimic the decision-making process of human experts. Examples include INTERNIST-1 and its successor, CADUCEUS, which were used for diagnosing complex medical conditions [3]. The late 1980s and 1990s marked the advent of machine learning techniques. Neural networks and early forms of pattern recognition started to be applied to medical imaging and diagnostics [1].

2000s: Advancements in data and algorithms: The widespread adoption of Electronic Health Records (EHRs) provided a wealth of data that AI systems could leverage. Machine learning algorithms began to be used for predictive analytics, identifying patient risk factors, and improving clinical decision support. The Human Genome Project, completed in 2003, generated large datasets that spurred the development of AI algorithms for genomics and personalized medicine [4].

2010s-Present: Deep learning and big data: The emergence of deep learning revolutionized AI in healthcare. Convolutional Neural Networks (CNNs) and other deep learning models achieved remarkable success in image recognition, making them highly effective for analyzing medical images like X-rays, MRIs and CT scans [5]. Advances in Natural Language Processing (NLP) enabled AI systems to process and analyze vast amounts of unstructured clinical data, such as physician notes and research articles, facilitating better information retrieval, clinical documentation and decision support. AI techniques, including machine learning and deep learning, have been increasingly used in drug discovery and development, significantly accelerating the process of identifying potential drug candidates [4].

Enhanced diagnostic accuracy: AI algorithms, particularly those based on deep learning, have demonstrated high accuracy in diagnosing diseases from medical images. This reduces diagnostic errors and leads to earlier detection of conditions like cancer, improving patient outcomes [5].

Personalized treatment plans: AI can analyze patient data, including genetic information, to tailor treatment plans to individual patients. This approach, known as precision medicine, enhances the efficacy of treatments and reduces adverse effects [6].

Streamlined clinical workflows: AI-powered tools automate routine administrative tasks, such as scheduling, billing, and documentation. This reduces the administrative burden on healthcare providers, allowing them to focus more on patient care [7].

Predictive analytics: AI systems can predict patient deterioration, hospital readmissions and other critical events by analyzing historical and real-time data. This enables proactive interventions, improving patient safety and outcomes [8].

Virtual health assistants: AI-driven chatbots and virtual health assistants provide patients with 24/7 access to medical information, symptom checking, and appointment scheduling. This improves patient engagement and access to care [7].

Telemedicine and remote monitoring: AI-powered telemedicine platforms and remote monitoring tools enable healthcare delivery in remote and underserved areas. Patients can receive medical consultations, monitor chronic conditions and access specialist care without the need for travel [9].

Resource optimization: AI can optimize the allocation of healthcare resources, ensuring that underserved populations receive timely and adequate care. Predictive analytics can identify areas with high disease prevalence and direct resources where they are most needed [10].

Mobile health (mhealth) solutions: AI-driven mobile health applications provide health education, preventive care and disease management support to individuals in underserved communities. These solutions are particularly valuable in regions with limited access to healthcare infrastructure [11].

Language translation and accessibility: NLP algorithms facilitate real-time language translation and communication between healthcare providers and patients who speak different languages. This improves healthcare accessibility for non-Englishspeaking populations [7].

Affordable diagnostic tools: AI has led to the development of cost-effective diagnostic tools that can be deployed in lowresource settings. For example, AI-powered portable ultrasound devices and diagnostic apps enable healthcare workers to perform accurate diagnostics in the field [10].

The evolution of Artificial Intelligence (AI) in healthcare reflects a remarkable journey from rudimentary expert systems to advanced deep learning models, significantly transforming various aspects of medical practice. This discussion integrates key insights from recent literature to contextualize the benefits, challenges and future directions of AI in healthcare.

Historical perspective and technological progress

Davenport T et al. [12] emphasize the transformative potential of AI in healthcare, noting that the technology has moved beyond early expert systems to more sophisticated applications. The historical development outlined in our manuscript aligns with this view, highlighting milestones such as the development of MYCIN and the growth of machine learning in the 1980s and 1990s. Davenport T et al. [12] argue that the integration of AI into healthcare is not just an incremental improvement but a paradigm shift that enhances diagnostic accuracy, personalizes treatment and streamlines clinical workflows. This view underscores the evolutionary trajectory of AI, from its nascent stages to its current state, marked by deep learning and big data analytics.

Current applications and benefits

Bajwa J et al. [13] discuss how AI is transforming medical practice by improving diagnostic accuracy and personalizing treatment plans. Our findings support this view, demonstrating that AI, particularly deep learning models, has significantly enhanced diagnostic precision and efficiency. For example, advancements in medical imaging, as detailed by Esteva A et al. [5] have enabled AI to achieve dermatologist-level accuracy in skin cancer classification. Bajwa J et al. [13] also highlight the role of AI in streamlining administrative tasks, a benefit echoed in our discussion on the reduction of administrative burdens through AI-powered tools. Moreover, the benefits of AI extend to predictive analytics and virtual health assistants, which have become crucial in managing patient care and engagement. Rajkomar A et al. [8] highlight how predictive models can foresee patient deterioration and hospital readmissions, allowing for timely interventions. Bajwa J et al. [13] further elaborate on how virtual health assistants improve patient access to care and information, aligning with our findings on the role of AI in enhancing patient engagement.

Expanding access to underserved populations

Jiang F et al. [14] provide a comprehensive overview of AI’s potential to address healthcare disparities, a theme central to our discussion on increasing access for underserved populations. AI-driven telemedicine and remote monitoring tools have proven effective in delivering healthcare to remote and underserved areas, as discussed by Keesara S et al. [9]. These technologies enable patients to access medical consultations and manage chronic conditions without the need for extensive travel, addressing significant barriers to care. Additionally, AI’s role in resource optimization and mobile health solutions is critical for improving healthcare delivery in low-resource settings. Our manuscript reflects this perspective, emphasizing how AI can optimize resource allocation and support mobile health applications, which are particularly valuable in regions with limited healthcare infrastructure [11].

Challenges and future directions

While the advancements in AI are promising, several challenges remain. Davenport T et al. [12] discuss the ethical and regulatory issues associated with AI in healthcare, including data privacy concerns and the need for transparent algorithms. Our discussion acknowledges these challenges and calls for ongoing research to address ethical considerations and ensure equitable AI integration. Bajwa J et al. [13] highlight the importance of continued innovation and adaptation in AI technologies to meet the evolving needs of healthcare. They advocate for interdisciplinary collaboration to overcome barriers and drive AI advancements. This aligns with our concluding thoughts on the future of AI in healthcare, emphasizing the need for further research and innovation to enhance health outcomes and address disparities.

Artificial intelligence has made significant strides in healthcare, from early expert systems to advanced deep learning models. The benefits of AI in patient care and for providers are substantial, enhancing diagnostic accuracy, personalizing treatments, streamlining workflows and enabling predictive analytics. Furthermore, AI plays a crucial role in increasing access to healthcare for underserved populations, both in the US and globally, through telemedicine, resource optimization, mobile health solutions, language translation and affordable diagnostic tools. The future of AI in healthcare promises even greater advancements, driving innovation and improving health outcomes worldwide.

1. Sutton RS, Barto AG (2018) Reinforcement learning: An introduction. (2^nd edn), MIT Press, Cambridge, USA.
2. Shortliffe EH (1977) MYCIN: A knowledge-based computer program applied to infectious diseases. Proceedings of the ACM Annual Conference, pp. 66-69.
3. Miller RA, Pople HE, Myers JD (1982) INTERNIST-1, an experimental computer-based diagnostic consultant for general internal medicine. New England Journal of Medicine 307(8): 468-476.
4. Topol EJ (2019) High-performance medicine: The convergence of human and artificial intelligence. Nature Medicine 25(1): 44-56.
5. Esteva A, Kuprel B, Novoa RA, Ko J, Swetter SM, et al. (2017) Dermatologist-level classification of skin cancer with deep neural networks. Nature 542(7639): 115-118.
6. Krittanawong C, Zhang H, Wang Z, Aydar M, Kitai T (2017) Artificial intelligence in precision cardiovascular medicine. Journal of the American College of Cardiology 69(21): 2657-2664.
7. Reddy S, Fox J, Purohit MP (2019) Artificial intelligence-enabled healthcare delivery. Journal of the Royal Society of Medicine 112(1): 22-28.
8. Rajkomar A, Dean J, Kohane I (2019) Machine learning in medicine. New England Journal of Medicine 380(14): 1347-1358.
9. Keesara S, Jonas A, Schulman K (2020) Covid-19 and health care’s digital revolution. New England Journal of Medicine 382(23): e82.
10. Mbunge E (2020) Integrating emerging technologies into COVID-19 contact tracing: Opportunities, challenges and pitfalls. Diabetes & Metabolic Syndrome: Clinical Research & Reviews 14(6): 1631-1636.
11. Watanabe A (2019) Telemedicine: Current and future perspectives. Journal of General and Family Medicine 20(3): 120-125.
12. Davenport T, Kalakota R (2019) The potential for artificial intelligence in healthcare. Future Healthc J 6(2): 94-98.
13. Bajwa J, Munir U, Nori A, Williams B (2021) Artificial intelligence in healthcare: Transforming the practice of medicine. Future Healthc J 8(2): e188-e194.
14. Jiang F, Jiang Y, Zhi H, Dong Y, Li H, et al. (2017) Artificial intelligence in healthcare: Past, present and future. Stroke and Vascular Neurology 2(4): 230-243.