The rapid advancement of artificial intelligence and digital health technologies is revolutionizing global healthcare. These innovative tools present unparalleled opportunities to enhance the efficiency and equity of healthcare systems worldwide. But to achieve these goals, it is necessary to overcome substantial challenges, encompassing technological, ethical, and governance issues, particularly in the context of global disruptions such as the United States' withdrawal from the World Health Organization, geopolitical tensions and a weakening of multilateral collaboration.

Today, AI is becoming increasingly embedded in the entire spectrum of medicine. In research and development, AI is massively accelerating drug discovery and enzyme design by predicting protein structures at pace and scale far surpassing human ability. Tellingly, developers of one of the leading AI solutions for this — Google DeepMind — won the Nobel Prize in Chemistry in 2024 for their unprecedented impact on biological research through AI.

Similarly, AI is helping with clinical trial design and analyzing vast biomedical datasets to uncover novel therapeutic targets. For patients, AI supports diagnostics through imaging analysis, predictive analytics and natural language processing. For example, AI-based retinal scans can detect not only eye disorders such as diabetic retinopathy and glaucoma but also aging disorders such as stroke and Parkinson's disease and heart failure using the same retinal images. AI is already being used in low- and middle-income countries for reading chest X-rays to flag suspected tuberculosis and chronic obstructive pulmonary diseases within seconds.

There are also applications to diagnose malnutrition through image analyses and predict high risk pregnancies. People living in low-resource settings, where there are no specialists, can now get accurate diagnoses from specialists sitting far away who study their pathology slides or MRI scans. This has potential for improving health equity globally.

AI is also helping transform promotive and preventive public health functions. It is enhancing disease and risk factor surveillance, improving outbreak prediction, and facilitating decision-making. In South American and many other countries, AI has been routinely forecasting granular short-term trends of air pollution, informing people in advance to take mobility decisions to avoid pollutants, and municipalities to manage/improve air quality.

Additionally, AI is improving health communication by customizing messages and predicting behavioural responses. In low-resource settings, AI is contributing to efficient healthcare system management and providing supportive tools for frontline workers.

While solutions for use cases will proliferate and modulate, it is critical that the ethics of AI always takes primacy while designing frameworks and solutions for use. The WHO's recent guidelines on ethics and governance of AI for healthcare are a step in the right direction. As we forge ahead, we must constantly identify and mitigate ethical issues that are relevant to global health, some of which are mentioned here.

We must invest in creating and curating diverse and representative datasets for eliminating the bias in algorithms and data, and making AI solutions more accurate.

Also, we need to augment humans-in-the-loop systems for guiding AI training and use in fields such as image-based diagnosis, patient counseling, clinical-decision systems and AI-robotic surgery. People should guide healthcare algorithm training, and rather than replacing clinicians, AI should augment them.

Besides, home and community-based AI tools must be designed for offline functionality and low-resource environments — this entails lightweight apps, local data storage, and user interfaces that do not assume literacy or tech fluency — and AI solutions should be built with, not just for, communities. To realize that, it is necessary to involve local engineers, clinicians, and patients in the design and evaluation process.

AI exemplifies the role of the private sector as one of the driving forces in healthcare outcomes. Incubating and promoting private initiatives through policy will play a pivotal role in the impact of AI. Parallelly, we must create frameworks for determining responsibilities and liabilities unique to applications and contexts, and governments need to balance entrepreneurial goals and prowess with social goals and regulatory systems.

Currently, the AI regulatory frameworks of China, the European Union, Japan, the United Kingdom and the United States are recognized as standout, even if nascent, governance architectures. However, there is also a need to devote considerable thought to how we promote "public health AI exchange" and AI as a public good across borders, especially in the current state of geopolitics, because all the public health gains across the world are built on sharing of ideas and talents, collaborations and joint action.

While exploring the power of AI it behooves us to take the wide view of determinants of health rather than limit ourselves to healthcare delivery. AI can help solve problems in mental health, air and water pollution, food security, housing, climate risk adaptation, digital — increasingly recognized as a determinant — and other factors. And global public health must enhance multi-sectoral action through AI for improving lives.

For the first time in human history, we are at the cusp of an era in which a farmer in Tikrit (Iraq), a trader in Guangzhou (China) and a schoolteacher in Lima (Peru) can access the same AI-enabled diagnostics and curative healthcare services at similar cost. It is for society to use scientific temper, ethics and political will to use AI to ensure equity in global health outcomes.

The views don't necessarily reflect those of China Daily. 

If you have a specific expertise, or would like to share your thought about our stories, then send us your writings at opinion@chinadaily.com.cn, and comment@chinadaily.com.cn.