Keywords
Artificial Intelligence, Global South, Healthcare Capability, Systematic Literature Review, ICT4D, capability approach, conceptual framework, Digital health, Health information systems, ICT4D, information systems
Artificial Intelligence (AI) has the potential to transform the healthcare ecosystem, but further research is needed to understand how it can enhance healthcare capabilities. This study analyzes the literature on AI and healthcare capability using the PRISMA approach, applying specific search keywords and inclusion/exclusion criteria. The findings indicate that AI benefits the healthcare ecosystem, significantly influences health outcomes, and transforms medical practices. However, there is limited literature and a lack of understanding regarding how AI enhances healthcare capabilities. Most studies date from 2019, suggesting that COVID-19 has accelerated the adoption of AI systems in healthcare. This research contributes theoretically by developing a framework that clarifies AI’s role in enhancing healthcare capabilities, serving as a foundational model for future studies. It identifies critical gaps in the literature, especially in the Global South, and encourages exploration in under-researched areas where healthcare professionals can benefit from AI. Additionally, it bridges the gap between AI and healthcare, enriching interdisciplinary dialogue relevant to emerging economies facing financial constraints. Practically, the study provides actionable insights for healthcare practitioners and policymakers in the Global South on leveraging AI to improve service delivery. It sets the stage for empirical research, promoting the testing and refinement of the proposed framework in resource-limited contexts, while raising awareness among healthcare staff, managers, and technology developers about AI’s role in healthcare.
Artificial Intelligence, Global South, Healthcare Capability, Systematic Literature Review, ICT4D, capability approach, conceptual framework, Digital health, Health information systems, ICT4D, information systems
This systematic literature review analyzes diverse sources to enhance understanding of the nexus between artificial intelligence (AI) and healthcare capabilities, particularly within the framework of ICT4D. AI, a technology designed to replicate human cognitive functions (Jiang et al., 2017), has the potential to generate significant value across various sectors (Mikalef et al., 2019). However, its specific contributions to the healthcare ecosystem remain underexplored.
This study highlights a notable gap in AI research, particularly regarding systematic analyses of its relationship with healthcare capabilities. The role of AI-enabled technologies in the healthcare sector is emerging as a crucial area of inquiry (Zhao, 2018), yet published research remains inadequate (Mikalef et al., 2019). Moreover, contributions related to business-oriented AI are minimal, indicating a pressing need for studies focused on AI applications in developing markets, especially in light of present and future health crises (Bharati, 2020). Further research is essential to explore how AI can support sustainable development goals while addressing potential trade-offs (Gupta et al., 2021). Future studies on AI for sustainability should consider various value perspectives to illustrate how AI can deliver immediate and impactful solutions (Nishant et al., 2020).
Consequently, this study aims to review the literature and generate insights on effectively applying AI within the healthcare ecosystem. It specifically addresses the research question: How does artificial intelligence empower healthcare capabilities? To answer this question and fulfill the study’s objectives, the article is organized as follows: it begins with a description of the methodology used to identify the nexus between AI and healthcare constructs in the literature, followed by the results. The discussion and conclusion sections will present the constructed conceptual model.
This study adopts the PRISMA approach to conduct a systematic literature review on AI and healthcare capability, with a particular emphasis on the implications for ICT4D. By utilizing the PRISMA framework, we aim to enhance the rigor, transparency, and relevance of our findings, ultimately fostering the development and implementation of AI-driven healthcare solutions grounded in solid evidence. This approach is vital in understanding how AI can contribute to healthcare improvements, especially in resource-limited settings typical of many developing regions. The PRISMA methodology has been widely employed in systematic literature reviews within the context of AI and healthcare. For instance, Abdolkhani et al. (2022) utilized this method to investigate the impact of digital health transformation, driven by COVID-19, on nursing practice, highlighting the transformative potential of technology in healthcare. Similarly, Choudhury and Asan (2020) applied this technique to explore the role of artificial intelligence in enhancing patient safety outcomes. By implementing the PRISMA approach, this study aims to systematically review the literature on AI and healthcare capabilities, drawing insights that are particularly relevant to the context of ICT4D. This will help identify pathways through which AI can empower healthcare systems, especially in developing regions, thereby promoting equitable access to advanced healthcare technologies.
This systematic literature review explores different reference materials from the PubMed database using different searching keywords. Table 1 shows search keywords and sources of literature this systematic literature study used during the examining process.
This research performs a scoping (initial) review of nearly a total of 759 studies which are related literature to AI and healthcare capability from PubMed database as per the data available by 22 May 2024. When we limit it to free full text and articles published after 2017, we get a total of 700 papers. This study applies exclusion and inclusion criteria (discussing or reporting the role of AI for healthcare capabilities, published in English language) and came up with 636 irrelevant, 64 not irrelevant articles. Out of the 64 articles, this study drops 9 articles because of being irrelevant, as they are not within the scope of the study. In other words, 64 articles passed to the next screening process. Finally, from 64 articles, this study drops 9 articles due to redundancy. As a result, we perform a systematic review with 46 articles in the domain area of AI and healthcare capabilities (see Table 2).
No. | Reference | Objective | Methods used | Key findings |
---|---|---|---|---|
1. | (Ahuja, 2019) |
|
|
|
2. | (Alhashmi et al., 2019) |
|
|
|
3. | (Asan & Choudhury, 2021) |
|
|
|
4. | (Bohr & Memarzadeh, 2020) |
|
|
|
5. | (Choudhury & Asan, 2020) |
|
|
|
6. | (D’Antonoli, 2020) |
|
|
|
7. | (Davenport & Kalakota, 2019) |
|
|
|
8. | (Fletcher et al., 2021) |
|
|
|
9. | (Floridi et al., 2018) |
|
|
|
10. | (Gama et al., 2022) |
|
|
|
11. | (Gerke et al., 2020) |
|
|
|
12. | (Goralski & Tan, 2020) |
|
|
|
13. | (Gordon, 2021) |
|
|
|
14. | (Guan, 2019) |
|
|
|
15. | (Gupta et al., 2021) |
|
|
|
16. | (Gwagwa et al., 2020) |
|
|
|
17. | (He et al., 2019) |
|
|
|
18. | (Hercheui & Mech, 2021) |
|
|
|
19. | (Iliashenko et al., 2019) |
|
|
|
20. | (Jiang et al., 2017) |
|
|
|
21. | (Keskinbora, 2019) |
|
|
|
22. | (Khullar et al., 2021) |
|
|
|
23. | (Liu et al., 2020) |
|
|
|
24. | (Malik et al., 2021) |
|
|
|
25. | (Matheny et al., 2020) |
|
|
|
26. | (Mikalef et al., 2019) |
|
|
|
27. | (Mikalef & Gupta, 2021) |
|
|
|
28. | (Naik et al., 2022) |
|
|
|
29. | (Nomura et al., 2021) |
|
|
|
30. | (Owoyemi et al., 2020) |
|
|
|
31. | (Paul et al., 2018) |
|
|
|
32. | (Racine et al., 2019) |
|
|
|
33. | (Reddy et al., 2020) |
|
|
|
34. | (Rong et al., 2020) |
|
|
|
35. | (Rosemann & Zhang, 2022) |
|
|
|
36. | (Schonberger, 2019) |
|
|
|
37. | (Secinaro et al., 2021) |
|
|
|
38. | (Shuaib et al., 2020) |
|
|
|
39. | (Strohm et al., 2020) |
|
|
|
40. | (Tran et al., 2021) |
|
|
|
41. | Vinuesa et al. (2020) |
|
|
|
42. | (Wiljer et al., 2021) |
|
|
|
43. | (Wolff et al., 2021) |
|
|
|
44. | (Vinuesa et al., 2020) |
|
|
|
45. | (Yin et al., 2021) |
|
|
|
46. | (Yu et al., 2018) |
|
|
|
The first step in the selection process is to find relevant articles by going through each article’s title and abstract. The complete texts of the remaining papers were examined after the abstracts to determine if they related to the current research project. The papers were also screened for duplicate research or publications, and only those that met the inclusion and exclusion requirements were downloaded and stored for later use. The bibliographies of all the remaining studies were also examined in order to locate additional published works that were not included in the online search that was chosen. Furthermore, a total of 46 papers discussing AI and healthcare capability. Following that, every chosen article was categorized under one of the main themes. As a preferred reporting item for systematic reviews, the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) Flow Diagram in Figure 1 was adopted and used. To generate themes for this study and collect pertinent data for additional analysis, a data extraction sheet was created ( Table 1).
The PubMed search yielded 46 relevant publications after quality assessment, covering a seven-year period from 2017 to 2024. These findings underscore the critical role of AI in the healthcare industry, particularly within the context of ICT4D. AI is designed to mimic human cognitive functions (Jiang et al., 2017) and replicate human thinking capabilities (Mohapatra & Kumar, 2019). Its capability is defined as “the ability of a firm to select, orchestrate, and leverage its AI-specific resources” (Mikalef & Gupta, 2021), highlighting the potential for organizations to mobilize technological innovations and optimize resources in ways that can enhance healthcare delivery in both developed and developing regions. AI is recognized as a transformative technology with the capacity to revolutionize the healthcare ecosystem (Schonberger, 2019), creating significant enabling impacts on health outcomes (Goralski & Tan, 2020). The integration of AI technologies is gradually transforming medical practices (Yu et al., 2018), facilitating notable advancements in diagnostics and treatment. Additionally, AI enhances healthcare operations and delivery processes, streamlining tasks and augmenting human roles across a variety of responsibilities (Bohr & Memarzadeh, 2020). This transformation is especially pertinent for developing economies, where AI can play a pivotal role in improving access to quality healthcare and addressing systemic challenges.
AI has the ability to drastically change healthcare and speed up medical research (D’Antonoli, 2020). AI will continue the principal enabler and driver to the transformation of healthcare to precision medicine (He et al., 2019). AI accomplishes healthcare enterprise management, assistance in diagnosis, and keeping a healthy lifestyle (Iliashenko et al., 2019). AI performs treatment recommendations and diagnosis, administrative activities, adherence and patient assignation (Davenport & Kalakota, 2019). AI solutions can enhance efficiency, healthcare quality, and diagnostic correctness (Hercheui & Mech, 2021). Technologies in the health ecosystem such as AI can resolve gaps in quality health and reach underserved communities (Paul et al., 2018). AI-enabled systems have the capability to advance access to quality health challenges in economically developing countries (Kalyanakrishnan et al., 2018). AI-enabled systems play a role in diagnosis, predicting the spread of diseases and customizing treatment paths (Secinaro et al., 2021).
AI can transform healthcare by turning big data of patients into actionable information, accelerating health responses, improving public health surveillance, and producing leaner and faster R&D (Raghupathi & Raghupathi, 2014). Analysis, diagnosis, and treatment of medical diseases are being transformed by AI algorithms (Wiljer et al., 2021).
Healthcare sectors show service improvement due to the recent advancement in ICT specifically e-health is emerged by the vital contribution of IT; and to improve excellent healthcare delivery systems in any country of the world, it is essential to execute solutions such as e-health (Azeez & Van, 2019). The traditional model of medicine is completely changed due to AI technologies and this technology significantly enhanced medical services level, and assured human health in numerous features (Liu et al., 2020). AI-enabled systems are progressively becoming an integral part of all our lives and are vital in the next-generation healthcare ecosystem (Bohr & Memarzadeh, 2020). AI significantly impacts sectors such as public health management, genomics, medical diagnostics, treatment personalization, drug development, supportive health services (Shuaib et al., 2020). AI has the capability to mimic human cognitive functions and can be functional to numerous categories of healthcare data such as unstructured and structured data (Jiang et al., 2017). The expansion of novel AI systems of machine learning (ML) changed the exercise of medicine by refining diagnosis and treatment accuracy across numerous specializations (Ahuja, 2019).
In healthcare industries, AI-enabled systems are augment physicians which are capable of caring for the upcoming medicine practice (Ahuja, 2019). AI can help physicians by automating clinical documentation and image analysis, assisting by virtual observation, diagnosis and patient outreach (Murali and Jayadevan, 2019). Increased health outcomes are observed using AI-based tools for many remote monitoring applications in heart failure, migraine, and diabetes management (Jeddi & Bohr, 2020). AI-enabled systems have the capability to improve the issue of quality health in developing nations (Kalyanakrishnan et al., 2018). AI can be applied in biomedicine because of the suitability of AI to resolve biomedical problems, and the continuous progress of AI itself (Rong et al., 2020). The incorporation of AI-based solutions to medical services ranging from appointment scheduling via intelligent chatbots to risk profile-based insightful diagnosis, intricate surgeries guided by intelligent robots, and mentoring services that described health goals and discussed sustainable solutions towards achieving desired goals through lifestyle changes (Guha, 2021).
In healthcare; descriptive (the most widely used which focus on event quantifying that already happened, and able to perceive trends and other insights based on the event data), predictive (it uses data from descriptive to make predictions about the future), and prescriptive (expands the purpose of predictive AI, detects trends and suggests possible treatments) are the three (3) wide groups for the uses of AI (Paul et al., 2018). AI-based systems are also valuable in epidemiological demonstrating of Covid-19 pandemic, and to guess needs of healthcare infrastructure, human resource requirements in future when the disease spreads, that help health agencies in adopting suitable control and prevention strategies (Malik et al., 2021). Cancer, stroke, neurology and cardiology are the major disease areas that use AI tools (Jiang et al., 2017).
AI can enable the accomplishment of the sustainable development goals (SDGs) (Vinuesa et al., 2020). Healthcare is one of the sectors that potentially benefited from AI (Smith & Neupane, 2018). In terms of patient care, diagnostics, and mentoring and support services, AI has the ability to unleash a new transformation (Guha, 2021). AI bargains significant opportunities to reduce costs, improve patient and clinical team outcomes, and stimulate people’s health (Matheny et al., 2020), and healthcare institutions should be accountable for AI-related medical faults (Khullar et al., 2021). AI can affect almost every aspect of the healthcare sectors from detection to prediction and prevention (Wiljer and Hakim, 2019). AI-enabled systems practice is growing at an unprecedented speed in the healthcare industry comprising surgical operations, triage or screening, disease diagnosis, and risk analysis (Yin et al., 2021). The health risks of patients can be identified through AI-enabled systems, as a result, AI has the potential to influence patient safety results (Choudhury & Asan, 2020).
AI can transform the way companies do business (Mikalef et al., 2019). AI can produce value in four different ways namely automation, decision support, marketing and innovation (Mikalef et al., 2019). AI has a role in risk management (Bhattacharya & Ghosh, 2007), and asset management (Bartram et al., 2020) which can generate values for different sectors. AI can also be used to enhance the judgment and decision-making of humans in a stream termed amplified intelligence (Zheng et al., 2017). Nowadays, AI is being deployed by many creative occupations to support innovation projects such as biomedical applications and AI is being used by designers to help in design and creativity (Heer, 2019). Some of the studies which are focusing on AI and healthcare are presented hereunder in tabular form.
The findings revealed that AI has the potential to revolutionize the healthcare ecosystem. In recent years, there has been a significant increase in interest in integrating AI into healthcare systems, with numerous studies examining the advantages and applications of this technology. Consequently, this study identifies and analyzes literature focused on the use of AI to enhance healthcare capabilities. The researcher delimited the search by utilizing various keywords in the PubMed database to retrieve research papers published in English since 2017. The review process adhered to the PRISMA framework (Page et al., 2021) and concentrated specifically on the intersection of AI and healthcare capability.
Based on the systematic review of 46 articles in the domain of AI and healthcare capabilities, the findings indicate that AI generates profound enabling influences on health and transforms medical practices. However, the investigation also reveals a limited understanding of how AI effectively enhances healthcare services, highlighting a nascent body of literature in this field. Following the review, the study identifies key themes emerging from the AI and healthcare literature and synthesizes these findings into a cohesive framework.
This research contributes to the body of knowledge on AI in healthcare by providing insights into the role of AI in enhancing healthcare capabilities. Additionally, it offers a framework for future empirical testing. The study underscores the need for more in-depth literature reviews and empirical research, particularly in the healthcare ecosystems of developing economies, where AI can play a crucial role in improving service delivery and addressing systemic challenges.
In summary, AI serves as a crucial enabler and transformative force within the healthcare sector, particularly through the lens of ICT4D (Information and Communication Technologies for Development). Despite its potential, there is a significant gap in research systematically analyzing the intersection of AI and healthcare capabilities. For instance, there remains a shortage of published studies specifically focused on AI (Mikalef et al., 2019), while access to advanced technologies remains a challenge for many countries (United Nations [UN], 2018). Additionally, research contributions related to business applications of AI are limited (Bharati, 2020), highlighting the need for further exploration of AI’s role in addressing current and future health crises (Bharati, 2020). This study addresses these gaps by examining the existing literature with a specific focus on AI and healthcare capability, aiming to elucidate how AI technologies can enhance healthcare systems, particularly in under-resourced contexts. Adopting the PRISMA framework as outlined by Page et al. (2021), the review process systematically identifies and synthesizes key themes within the literature on AI and healthcare. The findings culminate in a framework that serves as a roadmap for future research, emphasizing the critical role of AI in strengthening healthcare capabilities and promoting equitable access to health technologies globally.
In conclusion, the integration of AI into healthcare is becoming increasingly essential as organizations strive to enhance their operations and decision-making capabilities amid rapid technological advancements. This study systematically addresses the existing gap in literature regarding the interplay between AI and healthcare capabilities, offering a theoretical framework that elucidates how AI can empower these capabilities. By focusing on the specific context of the Global South, the research highlights critical gaps in understanding and encourages further exploration into areas where healthcare professionals can leverage AI to improve outcomes. The study contributes to the ICT4D discourse by emphasizing the potential of AI to foster development in financially constrained environments, thereby enriching interdisciplinary dialogue around technology’s role in enhancing healthcare delivery. It provides practical insights for healthcare practitioners and policymakers in these regions, equipping them with knowledge to effectively utilize AI for better service delivery. Moreover, the research sets a foundation for empirical studies, advocating for the testing and refinement of the proposed framework within resource-limited contexts. It aims to elevate awareness among healthcare staff, managers, and technology developers about the transformative role of AI in healthcare. Given that the review is limited to literature published in English since 2017, it underscores the need for more comprehensive research that includes diverse linguistic and cultural perspectives, ultimately enriching the understanding of AI’s potential in various healthcare settings.
Dereje was a lecturer at the University of Gondar and a part-time lecturer at various institutions. He previously served as an ERP project coordinator, senior business analyst, and IT specialist in Addis Ababa. He earned a B.Sc. in Information Science from Jimma University and an M.Sc. in Information Systems from Addis Ababa University. Currently, he is a PhD candidate in Information Systems at Addis Ababa University. In 2022, he completed a Higher Diploma Program at the University of Gondar and became certified in Global Sustainable Leadership in 2024. He holds numerous certificates from prominent technology firms, including Cisco and IBM, in areas such as AI, Data Science, AWS, Linux, Agile Project Management, and Cybersecurity. He has participated in summer schools and workshops, including “Data Science 2023” at the University of Rwanda and “Climate Justice” at the University of Kenyatta. As an ambassador for Applied Machine Learning Day (AMLD) Africa-Ethiopia, he has two publications and ongoing research in Artificial Intelligence, Information Systems Security, and Digital Transformation.
No data are associated with this article.
Figshare: Artificial Intelligence (AI) and Healthcare Capabilities: A Systematic Literature Review. DOI: https://doi.org/10.6084/m9.figshare.27794112.v2 (Ferede, 2024).
The project contains the following reporting guidelines data:
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Figshare: Artificial Intelligence (AI) and Healthcare Capabilities: A Systematic Literature Review. DOI: https://doi.org/10.6084/m9.figshare.27794112.v2 (Ferede, 2024).
The project contains the following reporting guidelines data:
• PRISMA_2020_checklist
• Figure 1. Literature Review ProcessAuthor’s Representation Using the PRISMA Approach
• Figure 2. Theoretical Framework (A Road Map for Future Research on AI in Healthcare Capability) Author’s Elaboration Based on Existing Literatures
Data are available under the terms of the Creative Commons Attribution 4.0 International license (CC-BY 4.0).
Views | Downloads | |
---|---|---|
F1000Research | - | - |
PubMed Central
Data from PMC are received and updated monthly.
|
- | - |
Are the rationale for, and objectives of, the Systematic Review clearly stated?
Yes
Are sufficient details of the methods and analysis provided to allow replication by others?
Partly
Is the statistical analysis and its interpretation appropriate?
Partly
Are the conclusions drawn adequately supported by the results presented in the review?
Yes
If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)
Partly
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Multimedia, AI
Are the rationale for, and objectives of, the Systematic Review clearly stated?
No
Are sufficient details of the methods and analysis provided to allow replication by others?
No
Is the statistical analysis and its interpretation appropriate?
Partly
Are the conclusions drawn adequately supported by the results presented in the review?
No
If this is a Living Systematic Review, is the ‘living’ method appropriate and is the search schedule clearly defined and justified? (‘Living Systematic Review’ or a variation of this term should be included in the title.)
No
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: AI in healthcare
Alongside their report, reviewers assign a status to the article:
Invited Reviewers | ||
---|---|---|
1 | 2 | |
Version 1 03 Jan 25 |
read | read |
Provide sufficient details of any financial or non-financial competing interests to enable users to assess whether your comments might lead a reasonable person to question your impartiality. Consider the following examples, but note that this is not an exhaustive list:
Sign up for content alerts and receive a weekly or monthly email with all newly published articles
Already registered? Sign in
The email address should be the one you originally registered with F1000.
You registered with F1000 via Google, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Google account password, please click here.
You registered with F1000 via Facebook, so we cannot reset your password.
To sign in, please click here.
If you still need help with your Facebook account password, please click here.
If your email address is registered with us, we will email you instructions to reset your password.
If you think you should have received this email but it has not arrived, please check your spam filters and/or contact for further assistance.
Comments on this article Comments (0)