CoChat
Sign Up
Cited by 177Artificial Intelligence in Healthcare and EducationEthics and Social Impacts of AIArtificial Intelligence in Healthcare

Artificial intelligence for global health

Open access
Hosny, Ahmed, Aerts, Hugo J. W. L.·2019·Science
DOI 10.1126/science.aay5189PMC PMC7790340
Open this paper in CoChat — free account

Abstract

Socially responsible technologies promise to help address health care inequalities

Integrity signals

Registry

Crossref · 10.1126/science.aay5189

Cites

15 of 15 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 12 cited sources checked · 0 supported · 0 flagged

  • Premisecitation
    § Artificial intelligence for global health

    Deep learning has produced AI systems whose performance on medical image interpretation and drug compound discovery can approach that of trained professionals.

    “Deep learning, a subset of machine learning based on artificial neural networks, has enabled applications with performance levels approaching those of trained professionals in tasks including the interpretation of medical images and discovery of drug compounds (1).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [1] Topol EJ, High-performance medicine: the convergence of human and artificial intelligence. Nat. Med 25, 44– 56 (2019). 30617339 source

      Not checked in the free run.

  • Interpretationreasoning
    § Artificial intelligence for global health

    AI may help keep national cancer registries current.

    “For example, AI may help maintain up-to-date national cancer registries.”
    Quote in the paperNot a substantial claim
  • Interpretationnone given
    § Artificial intelligence for global health

    Low-cost AI tools on smartphones or portable devices can be used by non-specialist community health workers in off-site settings to address common diseases.

    “The first includes AI-powered low-cost tools running on smartphones or portable instruments. These mainly address common diseases and are operated by non-specialist community health workers (CHWs) in off-site locations, including local centers and households.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: LMICs have unmet basic needs, including shortages of health care workers, limited access to drugs, imaging hardware, and surgical infrastructure.
    • Rests on an assumption: LMICs often lack the technical expertise to operate, maintain, and repair equipment even when it is available.
  • Interpretationnone given
    § Artificial intelligence for global health

    AI analysis of aerial photography and weather patterns can identify potential disease outbreak hotspots in unmapped rural areas.

    “Other applications include identifying hotspots for potential disease outbreaks in unmapped rural areas by utilizing AI-powered analysis of aerial photography and weather patterns, as well as planning and optimizing CHWs’ household visiting schedules.”
    Quote in the paperNot a substantial claim
  • Premisecitation
    § Artificial intelligence for global health

    More than half of all cancer registry activity expenses are labor costs.

    “Automated registry curation, by extracting standard data from free-form text found in radiology and pathology reports, may help reduce labor costs that account for more than 50% of all registry activity expenses (9).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [9] Tangka FKL, Subramanian S, Edwards P, Cole-Beebe M, Maxwell Parkin D, Bray F, Joseph R, Mery L, Saraiya M, Resource requirements for cancer registration in areas with limited resources: Analysis of cost data from four low- and middle-income countries. Cancer Epidemiology. 45 (2016), pp. S50– S58. 27793574 PMC5842435 source

      Not checked in the free run.

  • Classified as probably not a substantial claim. 14
  • May combine two claims. 8
  • The quote only partly states this claim. 1

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationreasoning
    § Artificial intelligence for global health

    AI embedded in equipment design may help nontechnical operators troubleshoot when technicians are unavailable.

    “Although AI may not directly address diagnostic and therapeutic equipment shortages , AI integration into equipment design may help nontechnical operators troubleshoot issues when technicians are scarce.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § Artificial intelligence for global health

    AI applications include diagnosing skin cancer from photographs and diagnosing malaria by analyzing peripheral blood samples.

    “Applications include diagnosing skin cancer from photographic images and analyzing peripheral blood samples to diagnose malaria (7);”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [7] Oliveira AD, Prats C, Espasa M, Zarzuela Serrat F, Montañola Sales C, Silgado A, Codina DL, Arruda ME, I Prat JG, Albuquerque J, The Malaria System MicroApp: A New, Mobile Device-Based Tool for Malaria Diagnosis. JMIR Res. Protoc 6, e70 (2017). 28442456 PMC5424126 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given
    § Artificial intelligence for global health

    AI analysis of maintenance history may help sustain long-term operations, predict failures, and prevent delays in parts and consumables.

    “By analyzing historical maintenance data, AI may also help sustain long-term operations, predict failures, and avoid delay on parts and consumables.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § Artificial intelligence for global health

    Ethical concerns about AI in health care include weakened patient data privacy, greater tension between care and profit, and a third party in the patient-doctor relationship that changes expectations of confidentiality and responsibility.

    “Ethical concerns about the use of AI in health care include undermining patient data privacy protections and exacerbating the existing tension between providing care and generating profit, as well as introducing a third party into the patient-doctor relationship, which changes expectations of confidentiality and responsibility (10).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [10] Char DS, Shah NH, Magnus D, Implementing Machine Learning in Health Care — Addressing Ethical Challenges. New England Journal of Medicine. 378 (2018), pp. 981– 983. PMC5962261 29539284 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given
    § Artificial intelligence for global health

    AI may enable non-specialist primary care physicians to read diagnostic radiology and pathology images and to refer to specialists only when needed.

    “AI may allow non specialized primary care physicians to perform specialized tasks including reading diagnostic radiology and pathology images, only referring to specialists if necessary.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: In oncology, a shortage of subspecialists may force oncologists to treat tumors across multiple anatomical sites, leading to lower-quality care.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationreasoning
    § Artificial intelligence for global health

    Patient-facing AI applications may offer guidance on lifestyle and nutrition, symptom self-assessment, and advice during pregnancy or recovery, which could let patients manage their health and reduce pressure on health systems.

    “With increasing smartphone penetration, patient-facing AI applications may guide lifestyle and nutrition, allow symptom self-assessment, and provide advice during pregnancy or recovery periods—ultimately allowing patients to take control of their health and reducing the burden on limited health systems.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: LMICs have unmet basic needs, including shortages of health care workers, limited access to drugs, imaging hardware, and surgical infrastructure.

  • May combine two claims.

    Premisecitation
    § Artificial intelligence for global health

    Population health AI can help public agencies understand cause-and-effect relationships, allocate scarce resources, and slow epidemics.

    “The third application area relates to population health and allows public agencies to realize cause-and-effect relationships, appropriately allocate the often limited resources, and ultimately mitigate the progression of epidemics (8).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [8] Wahl B, Cossy-Gantner A, Germann S, Schwalbe NR, Artificial intelligence (AI) and global health: how can AI contribute to health in resource-poor settings? BMJ Glob Health. 3, e000798 (2018). PMC6135465 30233828 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Interpretationnone given
    § Artificial intelligence for global health

    In radiotherapy, semi-automated treatment planning may accelerate delivery, increase patient intake, and allow more focus on clinical nuances without extra staff.

    “In radiotherapy, for example, semi-automation of the treatment planning process may speed up treatment delivery, increase patient intake, and allow greater focus on the clinical nuances of patient management—all without requiring additional personnel.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: In oncology, a shortage of subspecialists may force oncologists to treat tumors across multiple anatomical sites, leading to lower-quality care.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given
    § Artificial intelligence for global health

    It remains unclear whether population health AI applications can be translated into effective long-term health policies.

    “Although these applications may prompt immediate actionable interventions, their translation into effective long-term health policies remains unclear.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § Artificial intelligence for global health

    The global burden of disease has moved from infectious to chronic noncommunicable diseases, which indicates that health conditions in LMICs and HICs are converging.

    “Health conditions in LMICs and HICs are rapidly converging, as indicated by the recent shift of the global disease burden from infectious diseases to chronic noncommunicable diseases (NCDs, including cancer, cardiovascular disease, and diabetes) (2).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [2] Mayor S, Non-communicable diseases now cause two thirds of deaths worldwide. BMJ. 355, i5456 (2016). source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationnone given
    § Artificial intelligence for global health

    AI can plan and optimize the household visiting schedules of community health workers.

    “Other applications include identifying hotspots for potential disease outbreaks in unmapped rural areas by utilizing AI-powered analysis of aerial photography and weather patterns, as well as planning and optimizing CHWs’ household visiting schedules.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation
    § Artificial intelligence for global health

    HICs and LMICs share challenges such as physician burnout from work stress, inefficient clinical workflows, inaccurate diagnostic tests, and rising hospital-acquired infections.

    “Both contexts also face similar challenges, such as physician burnout due to work-related stress (3), inefficiencies in clinical workflows, inaccuracies in diagnostic tests, and increases in hospital-acquired infections.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [3] Rotenstein LS, Torre M, Ramos MA, Rosales RC, Guille C, Sen S, Mata DA, Prevalence of Burnout Among Physicians: A Systematic Review. JAMA. 320, 1131– 1150 (2018). 30326495 PMC6233645 source

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given
    § Artificial intelligence for global health

    AI tools may give specialists expert knowledge across multiple subspecialties.

    “AI tools may also help provide specialists with expert knowledge across multiple subspecialties.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning
    § Artificial intelligence for global health

    Better data collection in LMICs is essential for population health AI applications.

    “Improving data collection in LMICs is central to these applications.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: LMICs have unmet basic needs, including shortages of health care workers, limited access to drugs, imaging hardware, and surgical infrastructure.

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based). The quote only partly states this claim.

    Interpretationreasoning
    § Artificial intelligence for global health

    AI could help advance the UN Sustainable Development Goals, including the target of cutting premature NCD mortality by one third by 2030.

    “AI has the potential to fuel and sustain efforts toward these ambitious goals.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Most health care AI development serves HICs, where most research takes place.
    • Rests on an assumption: There is little discussion of what AI can contribute to medical practice in LMICs.
    • Rests on an assumption: LMICs have unmet basic needs, including shortages of health care workers, limited access to drugs, imaging hardware, and surgical infrastructure.

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).

    Interpretationnone given
    § Artificial intelligence for global health

    An existing mHealth application for remote patient-physician communication can be enhanced with an AI chatbot that triages patients before consultation.

    “An existing mHealth application for patient-physician remote communication can be enhanced with an AI chatbot to triage patients prior to the consultation.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).

    Interpretationnone given
    § Artificial intelligence for global health

    An AI application may use the deployment channels of existing digital technologies, making it almost immediately deployable.

    “An AI application may also use the deployment channels of existing digital technologies, making it almost readily deployable.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: The marginal cost of serving one additional user of an existing AI software service is very small, which gives AI economic scalability.

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationnone given
    § Artificial intelligence for global health

    AI could become a socially responsible technology with inherent equity.

    “AI could emerge as a socially responsible technology with inherent equity.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning
    § Artificial intelligence for global health

    These digital health scaling efforts could offer lessons for similar AI applications.

    “These efforts could provide learning opportunities for similar digital AI applications.”
    Quote in the paperNot a substantial claim
19 of 19

  • The paper relies on this without support.

    Most health care AI development serves HICs, where most research takes place.

    “Not surprisingly, most AI developments in health care cater to the needs of high-income countries (HICs), where the majority of research is conducted.”

  • The paper relies on this without support.

    There is little discussion of what AI can contribute to medical practice in LMICs.

    “Conversely, little is discussed about what AI can bring to medical practice in low- and middle-income countries (LMICs), where workforce shortages and limited resources constrain the access to and quality of care.”

  • The paper relies on this without support.

    LMICs have unmet basic needs, including shortages of health care workers, limited access to drugs, imaging hardware, and surgical infrastructure.

    “Despite these similarities, more basic needs remain unmet in LMICs, including health care workforce shortages, particularly specialist medical professionals such as surgical oncologists and cardiac care nurses. Patients often face limited access to drugs, diagnostic imaging hardware (ultrasound, x-ray), and surgical infrastructures (operating theaters, devices, anesthesia).”

  • The paper relies on this without support.

    LMICs often lack the technical expertise to operate, maintain, and repair equipment even when it is available.

    “When equipment is available, LMICs often lack the technical expertise needed to operate, maintain, and repair it.”

  • The paper relies on this without support.

    In oncology, a shortage of subspecialists may force oncologists to treat tumors across multiple anatomical sites, leading to lower-quality care.

    “This is particularly important in oncology, where lack of subspecialists may force an oncologist to manage tumors across multiple anatomical sites, and thus deliver care of inferior quality owing to the constantly varying scope of services.”

  • The paper relies on this without support.

    Most HIC-based health care AI applications are only at proof-of-concept stage and need clinical validation in prospective trials.

    “Most are at the proof-of-concept stage and require further demonstration of utility through clinical validation in prospective trials.”

  • The paper relies on this without support.

    AI methods are often uninterpretable, which makes failures hard to predict and results hard to assess critically.

    “The underlying methods are often uninterpretable, making it difficult to predict failures and critically assess results.”

  • The paper relies on this without support.

    AI training data are almost entirely from HICs, so models are skewed toward particular diseases, demographics, and geographies.

    “Data used to train AI models are almost entirely collected within HICs, and models are hence skewed toward certain diseases, demographics, and geographies.”

  • The paper relies on this without support.

    Varying statistical analysis and quality control introduce errors and systematic biases into AI models, limiting their generalizability in different contexts.

    “With varying degrees of statistical data analysis and quality control, errors and systematic biases are introduced into models, thereby limiting their generalizability, especially when deployed in different contexts.”

  • The paper relies on this without support.

    Medical malpractice and liability rules for health-related algorithmic decision-making have not yet been formulated.

    “From a regulatory perspective, medical malpractice and liabilities in health-related algorithmic decision-making are yet to be formulated.”

  • The paper relies on this without support.

    Nearly all health care AI tools perform a single task and cannot fully replace health professionals.

    “Nearly all AI tools in health care are single-task applications, and so they are incapable of fully substituting for health professionals.”

  • The paper relies on this without support.

    The marginal cost of serving one additional user of an existing AI software service is very small, which gives AI economic scalability.

    “Although sizable initial investments may be required, the marginal cost of providing an existing AI software service to one more user is minuscule, giving it economical scalability.”
12 of 12

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • AI diagnostic tools developed in high-income countries may recommend treatment plans that are very expensive or not available in LMICs.

    “There is also a potential risk that AI diagnostic tools developed in high-income countries may recommend treatment plans (e. g., medication, surgery) that are very expensive or not available in LMICs”

    From SHIFTing artificial intelligence to be responsible in healthcare: a systematic review (2022) · cites this paper as Hosny and Aerts, 2019

  • AI could predict and anticipate the spread of pathologies or vulnerabilities within certain groups or communities, and thus allow for more effective interventions in LMICs.

    “Alami et al. (2020) argue that AI could predict and anticipate the spread of pathologies or vulnerabilities within certain groups or communities, and thus allow for more effective interventions in LMICs”

    From SHIFTing artificial intelligence to be responsible in healthcare: a systematic review (2022) · cites this paper as Hosny and Aerts, 2019

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. T. Lancet, Lancet 393, 1478 (2019).

Cited by

Sorted by citations

Related

Authors
Hosny, A., & Aerts, H. J. W. L.
Affiliations
Brigham and Women's Hospital (US)
ORCID
Venue
Science
Publisher
American Association for the Advancement of Science (AAAS)
Published
2019-11-22
Volume
366
Issue
6468
Pages
955-956
ISSN
  • 0036-8075 (ISSN-L)
  • 1095-9203
DOI
10.1126/science.aay5189
PMC
PMC7790340
OpenAlex
W2990353513
Cited by
177
References
14
Citations by year
2026: 27 · 2025: 39 · 2024: 24 · 2023: 27 · 2022: 21 · 2021: 20 · 2020: 18 · 2019: 1
Type
journal article
Language
en
Open access
Licence
www.sciencemag.org/about/science-licenses-journal-article-reuse (vor)
Locations
  • Science (publishedVersion)
  • PubMed (publishedVersion)
  • Europe PMC (PubMed Central) (acceptedVersion)
  • PubMed Central (acceptedVersion)
  • Research Publications (Maastricht University) (submittedVersion, other-oa)
Funders
  • National Institutes of Health
  • National Cancer Institute (U01CA190234, U24CA194354)
Topics
  • Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
  • Social Sciences › Safety Research › Ethics and Social Impacts of AI
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
Keywords
artificial intelligence, global health, health care inequality, health equity
MeSH terms
Clinical Decision-Making, Artificial Intelligence, Delivery of Health Care, Developing Countries, Humans, Public Health, Global Health

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-07.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-07.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review