SPAQ

SOIK Corporation

SPAQ

Community-led population sensing and AI-enabled maternal risk detection in fragile settings

日本、沖縄県中頭郡読谷村

Japan

For-profit, including B-Corp or similar models

Each year, an estimated 287,000 women die from pregnancy- and childbirth-related causes — approximately one every two minutes (WHO). Nearly 70% of these deaths occur in sub-Saharan Africa. In the Democratic Republic of Congo, maternal mortality remains among the highest globally — roughly 100 times higher than in Japan or many European countries. The fundamental challenge is not solely access to services, but the failure of early risk detection systems. In fragile and remote settings, antenatal care coverage is inconsistent, pregnancies go unrecognized, and blood pressure or infection screening is irregular. Preventable complications — including hypertensive disorders, hemorrhage, and ectopic pregnancy — are therefore detected only once they escalate into life-threatening emergencies. A deeper structural gap lies in population-level sensing and data translation. Early warning signals first emerge within communities, yet reporting systems remain fragmented, paper-based, delayed, and prone to underreporting. Information fails to translate into timely decision-making. As a result, health systems operate reactively rather than anticipatorily. What is critically missing is a reliable, community-embedded population sensing and prediction framework that transforms frontline signals into early intervention — before harm occurs.

SPAQ is a smartphone-based maternal health platform designed for remote and resource-constrained settings. It enables standardized antenatal screening and longitudinal pregnancy tracking — from community detection to facility visits, referrals, and care completion. At the community and primary care level, SPAQ supports structured workflows including blood pressure measurement, urine protein testing, pregnancy testing, symptom screening, basic AI-assisted ultrasound guidance, and referral tracking. The app functions fully offline, with guided navigation and local-language audio support, enabling non-specialist frontline workers to deliver consistent, high-quality care. All screening data are digitized and securely synchronized to the cloud, allowing longitudinal tracking across community and facility levels. SPAQ is built on a sensing-to-prediction-to-action architecture. The sensing layer is operational and captures real-time population-level frontline data. Building on this foundation, we have developed a structured risk modeling and decision-support framework using our proprietary Impact Attribution Model (IAM), which formalizes the relationships between screening accuracy, behavior change, treatment pathways, and survival impact. This modeling foundation is complete, and integration into SPAQ’s Action Engine — enabling task guidance and resource prioritization dashboards — is currently underway. SPAQ transforms routine screening into predictive, data-driven maternal health management in fragile settings.

1. Pregnant women in remote rural and displaced settings Our primary beneficiaries are pregnant women living in hard-to-reach villages and refugee camps. SPAQ has conducted over 25,000 structured screenings in rural and displacement contexts. By aggregating frontline data and applying population-level sensing, the system identifies women at elevated risk of complications or mortality before deterioration occurs. This enables earlier referral, timely treatment, and prevention of avoidable maternal deaths. 2. Community health volunteers and midwives at small primary care facilities SPAQ equips frontline workers with standardized digital screening workflows and risk insights, designed to provide structured risk information and predictive signals. This design allows prioritization of high-risk pregnancies once fully implemented, supporting efficient allocation of limited staff and supplies and strengthening quality of care. 3. Government and public health authorities SPAQ generates rich, real-time, field-based data reflecting local epidemiology and service gaps. These insights support evidence-based policy decisions, budget allocation, and emergency response planning. We have also conducted data interpretation workshops with government stakeholders, ensuring that data translates into actionable decisions and system-level improvements.