Meeting Report
Harnessing the power of artificial intelligence for disease-surveillance purposes
The COVID-19 pandemic accelerated the development of AI-driven tools to improve public health surveillance and outbreak management. While AI programs have shown promise in disease surveillance, they also present issues such as data privacy, prejudice, and human-AI interactions. This session of the WHO Pandemic and Epidemic Intelligence Innovation Forum examines the use of Artificial Intelligence (AI) in public health by collecting the experience of key global health organizations, such the Boston Children's Hospital, the Global South AI for Pandemic & Epidemic Preparedness & Response (AI4PEP) network, Medicines Sans Frontières (MSF), and the University of Sydney.
Executive Impact: Key Metrics & Projections
Artificial intelligence is revolutionizing public health surveillance. The following key metrics highlight the transformative potential and projected impact of AI adoption in this critical domain.
0%
Reduction in Outbreak Detection Time
0%
Improvement in Diagnostic Accuracy
0%
Efficiency Gain in Data Processing
0x
Expected ROI from AI Investments
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
The world of healthcare is filled with time-consuming tasks. AI models can operate behind the scenes, assisting in the delivery of care and facilitating clinical decision-making across various domains, from radiology and pathology to the intensive care unit. For instance, advances in voice and language processing are revolutionizing physician-patient conversations. These interactions now flow more naturally, with the dialogue captured, processed, and integrated into electronic medical records.
In addition to operational and administrative applications, AI can harness clinical data to improve healthcare research and delivery. By analysing vast clinical datasets, AI algorithms can quickly identify potential new drug candidates. This significantly reduces the research and development timeline, potentially leading to faster access to life-saving medications.
45%
Potential reduction in drug discovery timelines with AI.
In the past two decades, the team at Boston Children's Hospital has focused on applying natural language processing models to specific areas of clinical medicine. In 2023, the team conducted a landscape analysis of the current applications of AI in infectious disease surveillance. One identified interesting example is the way the Greek government employed AI to enhance surveillance and effectively deploy resources for disease control during the COVID-19 pandemic.
Enterprise Process Flow
Incoming Travellers
→
AI Categorization (Low, High Risk)
→
Prioritize for PCR Testing
→
Collect New Data
→
Improve Predictive Model
This innovative use of AI showcases its potential to optimize resource allocation in disease control efforts. Boston Children's Hospital itself has utilized AI to study how epidemics impact communities, providing accurate local and regional forecasts for the hospital.
AI in Action: Boston Children's Hospital Forecasts
Description: During the winter 2022-2023, when pediatric infections surged and disrupted hospital operations in the US, AI tools helped to better understand when infections were occurring. By combining expert forecasts with data on different viruses circulating in the country, the AI models generated accurate local and regional forecasts for the hospital.
Impact: Provided clinicians and planners with valuable insights days or even a week in advance of an impending wave of infections, such as Respiratory Syncytial Virus and influenza.
Learnings: AI tools can effectively identify and differentiate pathogens within noisy data sets, promoting a One Health approach.
The transformative impact of recent AI advancements, particularly the availability of large language models and related tools, has prompted a re-evaluation of information mining practices from traditional and untraditional data sources. For many years, the team at Boston Children's Hospital has been mining online open data sources and extracting insights from unstructured data to gain valuable insights and inform public health decisions.
| Feature | Traditional Methods | AI/LLM-Powered Methods |
|---|---|---|
| Data Source Processing |
|
|
| Classification Systems |
|
|
| Information Extraction Speed |
|
|
| Bias Mitigation |
|
|
The emergence of large language models like ChatGPT has revolutionized this landscape. The near future will rewrite established practices because it will soon be possible to input an article into these tools and generate summaries with a level of detail and classification with a speed never achieved before. This transformation will not only enrich historical information but open vast new opportunities for data exploration and the extraction of additional attributes.
Clear frameworks should be put in place so that AI solutions follow fundamental values that include accountability, fairness, transparency, reliability, ethics, security, inclusivity, and sustainability, forming the foundation for effective solutions that cater to communities' needs and respect. First and foremost, it is crucial to consider the ethical implications and balance the benefits of AI with safeguarding people's rights.
Global.health
A collaborative data science initiative harnessing AI for disease surveillance, anonymized COVID-19 data.
It's crucial to recognize that large language models cannot be regarded themselves as generating 'correct answers'. The inherent nature of these models is generative, even when capable of producing imaginative content. Their effectiveness is heavily reliant on the quality and accuracy of the data used to train them. If the training data does not faithfully represent reality, the model's outputs may not be genuinely useful.
Advanced ROI Calculator: Quantify Your AI Advantage
Input your organization's details to see a personalized estimate of potential savings and reclaimed hours through optimized AI implementation.
Annual Savings
$0
Annual Hours Reclaimed
0
Implementation Roadmap: Your Path to AI Transformation
Our structured approach ensures a seamless and effective integration of AI into your enterprise, maximizing impact and minimizing disruption.
Phase 1: Discovery & Strategy Alignment
Engage stakeholders, identify critical surveillance needs, assess existing infrastructure, and define clear AI integration objectives. This phase includes a comprehensive data audit and ethical review.
Phase 2: Pilot Development & Data Integration
Develop a proof-of-concept AI model for a specific use case (e.g., early outbreak detection), integrate relevant data sources, and establish secure data pipelines with robust privacy controls.
Phase 3: Model Refinement & Validation
Rigorously test and validate the AI model's performance against historical data and real-world scenarios. Iterate on model design, address biases, and ensure interpretability and transparency.
Phase 4: Deployment & Capacity Building
Deploy the AI solution in a controlled environment, train public health professionals on its use, and establish continuous monitoring and maintenance protocols. Foster multidisciplinary collaboration.
Phase 5: Scaling & Continuous Improvement
Expand AI integration across broader surveillance efforts, explore new applications, and continuously refine models based on feedback and evolving public health needs. Implement governance frameworks for ongoing responsible AI use.
Ready to Transform Your Enterprise with AI?
Book a personalized strategy session with our AI experts to discuss your specific needs and how we can help you achieve your goals.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat