Despite significant improvements in the health of the global population, disease outbreaks and pandemics have become increasingly frequent and severe, leading to significant global health and economic impacts. For instance, 44 countries have experienced a tenfold increase in infectious diseases since the beginning of 2022 compared to a pre-pandemic baseline. Climate change, urbanization and global travel are factors contributing to the spread of infectious diseases, with vector-borne diseases like malaria and dengue fever surging in the past decade. The COVID-19 pandemic alone had infected more than 700 million people and caused 7 million deaths as of August 2024, reversing over a decade of gains in life expectancy. Vulnerable populations such as migrants and refugees are most at risk as they receive only limited access to healthcare and experience high rates of exposure to communicable diseases.
GeoAI can track disease spread, identify hotspots and inform public health interventions by combining geospatial data with traditional ML and deep learning techniques. By combining spatial data and maps with demographic, environmental and healthcare information, GeoAI can accurately and effectively visualize epidemiological trends and patterns, while identifying areas with high disease prevalence. For instance, GeoAI can analyse environmental data such as population density, air quality and access to health services to inform disease prevention strategies. Through its ability to efficiently model and predict disease outbreaks such as malaria and dengue fever, GeoAI can also be used to improve public health responses and provide evidence-based support for treatment plans. During the COVID-19 pandemic in 2020, the United Nations Global Pulse initiative explored using GeoAI to track infection rates, model transmission patterns and optimize vaccine distribution in low-resource settings. Elsewhere, scientists relied on social media data as a complementary resource to natural language processing and ML to track the transmission and trajectory of the COVID-19 pandemic in the United Kingdom.
AI for Good in focus: AI medical diagnostics and treatment
AI-powered medical devices and cloud-based data analysis systems are transforming medical diagnostics and screening processes. These technologies enhance the clinical value of medical data and enable earlier detection and intervention for life threatening diseases.
Le Lu, a researcher and Senior Director at Alibaba DAMO Academy, presented on the growing capabilities of AI-powered cancer screening at the 2024 AI for Good Global Summit. He highlighted that today’s AI models can detect the seven deadliest cancers—breast, lung, esophageal, stomach, colon, pancreatic and liver—which together account for 70 per cent of cancer deaths and 50 per cent of new cases globally each year. These AI systems are accessible via AI cloud platforms through public Application Programming Interfaces, broadening global access to advanced diagnostics. Lu also shared case studies where AI detected pancreatic tumours that routine CT scans missed, underscoring their superior diagnostic accuracy.
Tuberculosis, the world’s deadliest infectious disease, disproportionately affects low-income populations, with treatment costs often reinforcing cycles of poverty. AI Diagnostics, winner of the 2024 AI Innovation Factory Africa, developed an AI model that detects tuberculosis by analysing lung sounds. The company created a commercial AI-powered stethoscope and a tuberculosis lung sound training database to provide accurate, low-cost detection of lung abnormalities. This innovative technology lowers detection barriers, reduces patient suffering, speeds up treatment and helps limit disease transmission.
AI for Good in focus: Soft robots for humanity
Soft robotics refers to the coupling of soft material with the force-generating capabilities of rigid structures. These robots have a wide range of uses, including being used in archaeological exploration (e.g., exploring Peruvian ruins) and environmental monitoring (e.g., inspecting salamander habitats). Cameras and sensors placed at the robot’s tip can further enable real-time environmental feedback, enabling autonomous navigation and manipulation using techniques such as reinforcement learning.
On 18 March 2025, the AI for Good platform conducted a webinar with Dr. Allison Okamura, the Richard W. Weiland Professor of Engineering at Stanford University and a founding member of the Stanford Robotics Center. A major breakthrough presented in the webinar was the development of a patient-specific concentric tube robot - a semi-soft, pre-shaped nitinol system designed for minimally invasive surgery. This device outperforms traditional rigid surgical robots by successfully navigating complex anatomies with sub-millimeter precision.
The webinar also introduced novel fabrication techniques, such as ultrasonic welding of thermoplastic polyurethane fabric to create strong yet flexible pneumatic actuators capable of significant elongation via tip eversion. One of the main innovations was the “vine robot” - a pneumatically actuated, everting structure with 2D and 3D steering capabilities and the ability to grow autonomously. The webinar also featured soft haptic technologies, such as 3D-printed pneumatic wearables, which give realistic tactile feedback for teleoperation and social interaction while addressing power efficiency and sensor integration challenges. These improvements showcase the unique ability of soft robots to bridge the gap between rigid and fully soft systems. They also address critical concerns such as material durability, autonomy in unstructured environments and the scalability of pneumatic systems. Looking ahead, researchers are investigating 3D-printed multifunctional materials and AI-driven control systems. The webinar concluded by underlining how soft robotics may help democratize access to key technologies by using low-cost, bioinspired designs.
