Last-Mile Delivery of Medical Supplies

Access to medical care cannot be taken for granted. There are still societies around the world where people are forced to make long and dangerous journeys to reach it. According to the World Bank’s Rural Access Index (RAI), in rural Africa only 34% of the population lives within 2 km of an all-season road, compared to over 90% in East Asia. In hard-to-reach areas, this lack of access translates directly into higher supply costs, longer delivery times, and preventable deaths.

The need for life-saving deliveries of blood and medical supplies continues to grow. Many infants, children and pregnant or breastfeeding women are dying from illnesses that could be prevented with medicines. In South Sudan, preventable diseases such as malaria and pneumonia account for around 75% of child deaths, and 1 in 10 children die before their fifth birthday (UNICEF South Sudan; UN IGME, 2023).

SORA Technology works to address these challenges by establishing new infrastructure in the sky. Using drone air mobility to enable safe, reliable, on-demand transport of goods between multiple locations, SORA focuses on emergency small-lot deliveries of pharmaceuticals, vaccines, and specimens that are highly urgent or require temperature control. This contributes to the realisation of Universal Health Coverage (UHC) in Asia and Africa. Improved access to medicines in developing countries also benefits the wider world by enabling timely diagnosis and treatment, and reducing the risk of infectious disease spread.

In July 2025, SORA Technology conducted a two-day field demonstration on Leyte Island, the Philippines, applying its drone and AI-based vector control approach — developed through malaria control operations in Africa — to dengue prevention for the first time. The project was implemented in partnership with Help.NGO, a UN-registered international humanitarian organisation with extensive experience in disaster response and technology-enabled field operations.

Drones conducted aerial surveys to map potential breeding water bodies across the target area. AI models then analysed environmental indicators to identify sites at highest risk of supporting Aedes mosquito populations, the primary vectors of dengue. The demonstration was designed to test whether SORA’s approach could be effectively adapted from the malaria context in Africa to the dengue context in Southeast Asia.

The collaboration with Help.NGO reflects a deliberate strategy of pairing technical innovation with established humanitarian networks and local field expertise. SORA intends to build on this demonstration toward broader deployment across the Philippines and other countries in the region where dengue, chikungunya, and other neglected tropical diseases pose growing public health threats — risks that climate change is expected to intensify.

AI-Assisted Flood and Infectious Disease Risk Prediction in Nairobi, Kenya

Climate change is intensifying flood risk across African cities, with severe consequences for public health. In informal settlements with poor drainage infrastructure, heavy rainfall creates pools of stagnant water that drive outbreaks of waterborne diseases including typhoid and cholera. Conventional public health responses are reactive — triggered after disasters occur rather than in advance of them.

Between September 2024 and March 2025, SORA Technology completed a proof of concept in Nairobi, Kenya, deploying an AI and satellite-based flood management system to test whether disease outbreak risk could be predicted and acted on before it materialised. The project was conducted under the Tokyo Metropolitan Government’s King Salmon Program, which supports innovative startups expanding into global cities, and received additional backing from Japan’s Ministry of Economy, Trade and Industry.

The system generated predictions across eight indicators, covering flood hotspot locations, estimated affected populations by sub-county, impacts on roads and hospitals, and projected patient numbers and medicine demand for diarrheal diseases and typhoid. When validated against actual flooding events in Nairobi in April 2024, the flood hotspot prediction model achieved an accuracy rate of 69%, with overall damage trends aligning closely with observed outcomes.

The key operational advance was timing. Where information had previously only been available after a disaster, the system demonstrated the ability to generate risk assessments approximately one week in advance — providing local governments and healthcare providers with lead time to pre-position supplies and plan responses.

SORA Technology is exploring expansion of the platform to additional cities in Kenya and other countries in Africa, building on its existing operations in Ghana, Kenya, and Mozambique.

Drone and AI-Assisted Larval Source Management in Ghana

Malaria remains one of the leading causes of illness and death in sub-Saharan Africa. In Ghana, larval source management (LSM) — the targeted treatment of mosquito breeding sites with larvicide — is recognised as a complementary vector control strategy, but its wider adoption has been constrained by the high cost and labour intensity of conventional manual mapping and treatment.

In 2024, SORA Technology conducted a comparative field trial across eight sub-districts in Ghana’s Kwaebibirem District, in partnership with the University of Ghana Business School, the Noguchi Memorial Institute for Medical Research, and Ghana’s National Malaria Elimination Program. Four sub-districts received SORA’s drone and AI-assisted approach; four continued with standard manual LSM as controls.

Drones mapped each area one to two days before larvicide application, generating high-resolution imagery of potential breeding sites. An AI model then classified each water body by larval risk, directing field teams via mobile devices to treat only high-risk sites. The results were significant: drone mapping identified 3.61 times more breeding sites than manual scouting, labour requirements were cut by approximately half, and larvicide efficiency improved nearly threefold — all without compromising malaria case trends relative to the control group.

The findings, published in PLOS ONE in February 2026, provide the first peer-reviewed evidence that integrating drone mapping and AI risk classification into LSM operations can substantially reduce resource inputs without undermining vector control effectiveness.