AI for a Healthier, More Secure Population

The national security of the United States depends on a strong public health infrastructure, affordable and accessible healthcare, and investment in biomedical research. Maintaining the nation’s position as a world leader in biotechnology and medical research will require significant investment in innovation, including expanding the use of artificial intelligence (AI) to combat biological threats, prepare for future pandemics, and improve quality of life. Recognition of the importance of operationalizing AI for healthcare applications is growing; however, a lack of urgency poses a significant threat to the health of U.S. citizens and the nation’s efforts to create a future where AI is responsibly employed.

The COVID-19 pandemic brought into focus two alternate futures: one future where continued underfunding of the federal information technology and AI infrastructure places America at risk for catastrophic failure during public health emergencies; and another future where we realize the full potential of AI to rapidly develop vaccines based on a virus’ genetic sequence and align critical supplies to disease hot spots, ensuring that citizens across the nation have access to live-saving care. Preparedness for future pandemics is critical for national security and requires a whole-of-government approach where federal civilian agencies include AI as a mission-critical priority and take a central role in advancing AI capabilities.

The National Security Commission on Artificial Intelligence (NSCAI) Final Report emphasizes the importance of committing resources, talent, technology, and partnerships to develop the future of AI for civilian purposes. At the NSCAI’s Global Emerging Technology Summit, leaders from nations around the world joined with the private sector to outline a comprehensive strategy for advancing AI platforms through multilateral cooperation and adherence to democratic values. As this global collaboration takes shape, federal civilian agencies will provide critical leadership and innovation. Winning the global AI competition will drive advancements in biomedical research and development (R&D); therefore, it is imperative that all federal agencies, including civil health organizations, prioritize investment in AI.

Many federal health agencies are piloting AI to test the potential for this technology to automate repetitive processes, improve data accessibility, and inform decision making. These early efforts are critical for developing best practices for the responsible use of AI and for building buy-in for new approaches. Examples of key AI successes include:

• The Centers for Medicare and Medicaid Services (CMS) sponsored the industry-wide AI Health Outcomes Challenge to encourage the use of deep learning and neural networks to predict hospital readmissions and adverse events. The challenge demonstrated the potential of AI to reduce health risks and lower the total cost of unplanned admissions.
• The Agency for Healthcare Research and Quality (AHRQ) invested in the development of a cognitive search engine and chatbot that use machine learning to connect users with the most relevant information about complex medical topics. This effort demonstrates that natural language processing and cognitive search can improve data accessibility and automatically retrieve the most relevant information to generate healthcare insights.
• The U.S. Food and Drug Administration (FDA) developed an AI pilot to screen imported foods and anticipate foodborne risks. This approach uses machine learning and retrospective data to predict which food shipments are more likely to contain products that could lead to foodborne illness outbreaks, demonstrating how AI can be deployed in the field to address mission-critical challenges.
• The National lnstitutes of Health (NIH) National Cancer Institute (NCI) is running a clinical trial to test whether machine learning can help analyze facial imaging, voice, and spoken language for the capture and classification of cancer pain information. The design of this observational study, which includes selecting a sample to include diversity of genders and skin types, illustrates how AI can transform clinical trials and highlights the importance of considering bias in collecting data to inform machine learning algorithms, especially for healthcare applications.

As civilian organizations continue to increase their investment in AI, a coordinated approach that supports efforts to scale and operationalize this technology will be necessary to ensure that civilian agencies are taking the lead in protecting public health and, by extension, national security. Just as the Department of Defense (DOD) created the Joint Artificial Intelligence Center (JAIC), an overarching organization focused on advancing the development and adoption of AI solutions across DOD, federal health agencies require a similar governance structure to coordinate AI efforts and create more efficiencies in large-scale implementations.

In addition to serving as a leader in healthcare innovation, federal health organizations can contribute to the nation’s security agenda, as outlined in the NSCAI Final Report in the following ways:

1. Combat biological threats: The NSCAI Final Report calls for the federal government to “update the U.S. National Biodefense Strategy to include additional AI-enabled biological threats,” including human enhancement. With the advent of AI-enabled genome sequencing and annotation, widespread use of AI for gene editing is not far off and poses a new type of biological threat. In addition to increased funding for the aggregation of genetic data and AI-enabled genomic R&D, creation of a governance framework to enforce the responsible use of AI in genomics is necessary to address emerging ethical issues.
2. Prepare for future pandemics: Collaborative global disease monitoring based on open-source data will enable governments to predict and prepare for pandemics. According to the NSCAI Final Report, “Combining increased transparency and data sharing on disease outbreaks with AI tools” will be critical for outbreak detection. Cross-agency initiatives and coordination with groups in industry and academia focused on disease monitoring and data aggregation, such as the Global.health data science initiative, will be central to efforts to monitor foodborne, waterborne, and environmentally transmitted diseases.
3. Improve quality of life for all citizens: Chronic conditions are the leading causes of death in the United States, and diseases like diabetes, kidney disease, and cancer are a key contributor to the nation’s rising health costs. AI is increasingly used to mine large data sets to identify the best treatment for patients suffering with chronic conditions, but additional investment is required to make personalized healthcare a reality. As the NSCAI Final Report states, “AI technology that can help the elderly live independently longer, assisting in managing health and daily tasks and improving the quality of life” will enhance security, stability, and societal well-being.
4. Reduce the cost of medical care: The escalating cost of medical care is a major challenge for both public health and national security, and investment in AI is one critical way to address out-of-control spending. AI-enabled systems to detect aneurysms, distinguish between cancerous and benign nodules, and capture ultrasound images can reduce time to diagnosis. Machine learning algorithms that predict high-risk patients can identify patients at risk of readmission, leading to reduced levels of disability and a lower risk of mortality. As the applications of AI for clinical care continue to grow, federal guidelines for developing and deploying operational AI solutions will be required—from best practices for the integration of AI within legacy environments to approaches for model testing and recalibration.

In addition to prioritizing these four areas, efforts to support a healthier population start with reducing the barrier to accessing quality care for all citizens and ensuring that the United States maintains its position as a leader in biopharmaceutical innovation, where AI has the potential to transform critical trials, drug discovery, and telemedicine.

The future of healthcare depends on innovation, and AI is a critical enabler for solutions that will protect public health and improve societal well-being, key elements of a national security strategy. Advances in genomics, disease surveillance, clinical care, and biomedical R&D require a renewed focus on creating infrastructure and frameworks for data sharing to drive AI solution development in healthcare. The NSCAI Final Report recommends investing in public-private and global partnerships to “enhance the United States’ position as an international emerging technology research hub,” which will in turn advance national biotechnology, biodefense, and security initiatives. As a trusted AI partner to the U.S. civilian, defense, and intelligence agencies, Booz Allen is committed to partnering with the federal government to ensure that the nation maintains its leading role in advancing responsible and ethical AI for a healthier population.