Defining Agentic AI and Its Transformative Potential in Healthcare
Agentic Artificial Intelligence (AI), a cutting-edge advancement in AI technology, refers to autonomous systems capable of goal-oriented reasoning, dynamic decision-making, and coordinated action with minimal human supervision. Unlike traditional AI models that react to specific prompts or perform isolated tasks, agentic AI perceives its environment, plans multi-step strategies, selects appropriate tools, and executes actions proactively. This paradigm shift is particularly promising in healthcare, where complex workflows demand adaptability and real-time integration of multimodal data such as patient history, imaging, and lab results.
In the context of eye care, agentic AI addresses longstanding challenges like resource constraints and the need for personalized treatment. Singapore, with its rapidly aging population and high prevalence of vision-threatening conditions, stands at the forefront of this innovation. The city-state's myopia epidemic—where up to 80 percent of young adults are affected—combined with rising diabetic retinopathy cases, underscores the urgency for efficient, scalable solutions.
The Lancet Digital Health Publication: A Milestone for Singapore Research
A pivotal publication in The Lancet Digital Health on February 19, 2026, titled "Agentic artificial intelligence in eye care: is clinical autonomy finally within reach?", spotlights Singapore's leadership. Led by researchers from the National University of Singapore (NUS) Yong Loo Lin School of Medicine, Singapore Eye Research Institute (SERI), Singapore National Eye Centre (SNEC), and Duke-NUS Medical School, the paper outlines how agentic AI can orchestrate end-to-end clinical workflows.
The authors, including Ke Zou from NUS and collaborators from SERI and Duke-NUS, argue that ophthalmology's image-rich, protocol-driven nature makes it ideal for agentic systems. These AI agents can handle history-taking, image analysis, diagnosis summarization, treatment planning, and scheduling autonomously, reducing clinician workload by automating routine tasks while flagging complex cases for human review.
Singapore's Ecosystem: Universities Driving AI Innovation in Ophthalmology
Singapore's higher education institutions are central to this revolution. SERI, affiliated with NUS and Duke-NUS, tops global rankings as the number one non-academic institute in ophthalmology per ScholarGPS 2026. Its AI & Digital Health Research Group, headed by Associate Professor Daniel Ting from Duke-NUS, develops tools like SELENA+, an AI system screening for diabetic retinopathy, glaucoma, and age-related macular degeneration used nationwide.
Duke-NUS Medical School, a collaboration between Duke University and the National University of Singapore, integrates clinician-scientists into AI research. NUS's School of Computing partners with SERI on machine learning models, while initiatives like the SingHealth Duke-NUS Artificial Intelligence in Medicine Institute (AIMI) explore agentic AI for diagnostics and workforce augmentation. These universities produce graduates equipped for higher ed jobs in AI-driven eye care.
Key Applications: From Cataract Workflows to Diabetic Retinopathy Screening
The Lancet paper illustrates agentic AI in a cataract surgery workflow: a lead agent (large language model) gathers symptoms and orders tests; vision agents analyze fundus images; a summarizer compiles findings and recommends surgery; a scheduler books procedures and follow-ups. This multi-agent collaboration achieves near-clinician performance with 80 percent accuracy gains over single models, as seen in similar frameworks like EyeAgent.
In Singapore, AI already screens 500,000 diabetics annually via the National Diabetic Retinopathy Screening Programme, detecting referable cases with 90 percent sensitivity. Agentic extensions could predict progression, personalize atropine therapy for myopia control, or triage high-risk patients amid the ophthalmologist shortage—one specialist per 20,000 residents versus global averages.
- Automated triage: Prioritizes urgent cases like retinal detachment.
- Personalized follow-up: Adjusts plans based on real-time data from wearables.
- Subspecialty bridging: Integrates cardiology data for hypertensive retinopathy.
Singapore's National Push: Synapxe and HealthX Championing Agentic AI
Synapxe, Singapore's national healthtech agency, is piloting agentic AI through HealthX challenges, targeting patient engagement in specialties like ophthalmology. Platforms like Enigma Health, a SingHealth Duke-NUS spin-off, deploy sovereign agentic AI for workflow optimization, partnering with Roche and ST Engineering.
Under the National AI Strategy 2.0, S$1 billion invests in AI governance, with eye care prioritized due to 26 percent myopia in Primary 1 students (down from higher rates). SERI's SERI-IHPC Joint Lab with A*STAR advances trustworthy agentic models, ensuring fairness across ethnic groups.
Photo by Thomas Kinto on Unsplash
Challenges: Computational Demands, Ethics, and Safety in Deployment
Despite promise, agentic AI faces hurdles. High compute needs strain edge devices; fragile multi-agent coordination risks errors in emergencies; emergent behaviors demand verifiable governance. The Lancet highlights privacy safeguards and fairness audits essential for equitable care in diverse Singapore.
Ethical concerns include bias in training data—SERI mitigates via multi-ethnic datasets—and liability for autonomous actions. Singapore's Health Sciences Authority (HSA) regulates AI as medical devices, with AIMI focusing on explainability.
| Challenge | Solution |
|---|---|
| Compute Intensity | Edge-cloud hybrids |
| Coordination Fragility | Standardized protocols |
| Ethical Bias | Diverse datasets, audits |
Case Studies: AI Successes Paving Way for Agentic Systems
Singapore's track record bolsters agentic adoption. SELENA+ screens 1.2 million retinal images yearly, reducing false positives by 20 percent. Duke-NUS AI predicts chronic kidney disease from fundus photos with 94 percent accuracy. Myopia AI tools flag high-risk children early, preventing 20 percent progression to high myopia (>-5D).
EyeAgent, a multimodal agent, boosts junior ophthalmologists' accuracy by 18.5 percent, mirroring agentic potential. SERI's papilledema detection and African DR validation extend impact globally.
Stakeholder Perspectives: Clinicians, Patients, and Policymakers
Ophthalmologists at SNEC view agentic AI as a "teammate," easing burnout amid 15 percent annual case growth. Patients appreciate faster access—AI triage cuts wait times 30 percent. Policymakers, via MOH, fund RIE2030 with S$25 billion for AI health, emphasizing sovereignty.
Experts like Prof Wong Tien Yin (SERI) stress human-in-loop for trust. Surveys show 88 percent clinician acceptance for assistive AI.
Future Outlook: Autonomous Eye Care and Workforce Transformation
By 2030, agentic AI could handle 50 percent routine Singapore eye care, freeing specialists for surgery. Integration with telemedicine and wearables enables predictive care—e.g., alerting to glaucoma spikes. Universities like NUS train AI ophthalmologists via higher ed career advice programs.
Global implications: Export SERI models to Asia-Pacific, addressing 2.2 billion vision-impaired worldwide.
Implications for Singapore Higher Education and Research Careers
NUS and Duke-NUS lead agentic AI curricula, blending medicine, computing, and ethics. Graduates pioneer at SERI, attracting university jobs in AI health. Research funding surges—Anlayze NIRF rankings penalize retractions, boosting quality.
Prospective students: Pursue Singapore academic opportunities for cutting-edge projects.
Photo by Samuel Regan-Asante on Unsplash
Actionable Insights: Adopting Agentic AI in Practice
- Clinicians: Pilot multi-agent tools for triage; train via SERI workshops.
- Researchers: Collaborate on Duke-NUS AIMI grants.
- Institutions: Implement HSA-compliant governance.
- Patients: Engage AI-screened programs for early detection.
For career growth, visit Rate My Professor for faculty insights.
Conclusion: Singapore Pioneering Autonomous Eye Care
The Lancet publication cements Singapore's role in agentic AI for eye care autonomy. With NUS, Duke-NUS, and SERI driving innovation, expect scalable, equitable vision health. Explore higher ed jobs, university jobs, and career advice to join this revolution. Share your thoughts in comments below.
