A team from University College of Dentistry at the University of Lahore conducted a cross-sectional survey among 451 medical and dental clinicians in Pakistan. Employing the General Attitude towards Artificial Intelligence Scale and a self-formulated readiness questionnaire, they quantified practitioners’ positive and negative perceptions, familiarity, and confidence in operating AI systems to facilitate informed AI adoption in resource-constrained settings.
Key points
Surveyed 451 public and private medical/dental practitioners in Pakistan using GAAIS and a custom readiness tool.
Positive attitude mean score was 3.6±0.54; negative attitude mean score was 2.8±0.71 on a 5-point Likert scale.
Dental practitioners showed significantly higher confidence in AI operation (38.4% vs. 29.8%, p=0.047) and willingness for AI in diagnosis (68.5% vs. 57%, p=0.004).
Why it matters:
This study underscores critical practitioner readiness and ethical considerations necessary to guide successful AI integration in resource-limited healthcare systems.
Q&A
What is the GAAIS scale?
Why reverse-code negative items?
How do statistical tests support findings?
What barriers exist in LMIC AI adoption?
Read full article
Academy
Artificial Intelligence Adoption in Resource-Limited Healthcare Settings
Artificial intelligence (AI) in healthcare refers to computer systems and algorithms designed to perform tasks such as diagnosis, treatment planning, and administrative workflows. In resource-limited settings—where infrastructure, funding, and trained personnel may be scarce—AI can help expand access, improve accuracy, and streamline clinical processes.
Key Concepts
- Machine Learning: A subset of AI where systems learn patterns from data to make predictions or decisions without explicit programming.
- Deep Learning: Uses neural networks with multiple layers to analyze complex datasets such as medical images.
- Decision Support: AI tools can suggest diagnoses or treatment options, assisting clinicians in making data-driven decisions.
Barriers to Adoption
- Digital Infrastructure: Limited internet connectivity and hardware impede AI deployment in rural clinics.
- Data Scarcity: Lack of standardized electronic health records hinders model training on local patient populations.
- Training Gaps: Clinicians often lack formal education on operating AI tools and interpreting outputs ethically.
- Cost Constraints: High upfront investment in AI platforms and maintenance may be prohibitive for public hospitals.
Strategies for Successful Integration
Capacity Building: Offer workshops, online modules, and mentorship programs to improve clinicians’ digital literacy and confidence in AI operation.
Ethical Guidelines: Develop policies on data privacy, informed consent, and bias mitigation to build trust among practitioners and patients.
Pilot Projects: Start with small-scale deployments (e.g., AI-assisted radiology in one clinic) to demonstrate value and generate local data for model refinement.
Public–Private Partnerships: Collaborate with technology providers, government agencies, and academic institutions to co-fund infrastructure and training initiatives.
Impact on Patient Care
When effectively implemented, AI can accelerate turnaround for diagnostic imaging, predict disease risks from clinical data, and personalize treatment plans based on patient characteristics. In resource-limited settings, these gains translate to earlier interventions, reduced medical errors, and more equitable access to high-quality care.
Conclusion: By addressing infrastructure, training, and ethical considerations, resource-constrained healthcare systems can harness AI’s potential to strengthen service delivery, improve outcomes, and support practitioner readiness.
