SIGLA, the System for Intelligent Growth and Learner Anthropometry, uses computer vision to support the screening of learners for stunting, thinness, overweight, and obesity. It is designed to reduce the manual burden on school personnel while preserving teacher-led and school health workflows. Its public value lies in enabling faster and better-supported learner health monitoring, provided that model performance, privacy safeguards, and human review mechanisms are validated before scaling.
SIGLA supports EDCOM II reform priorities related to learner welfare, inclusion, and school-level support systems. It also aligns with Q-BEDP priorities on learner health and well-being, PREXC requirements for school health and nutrition workflows, and Philippine Development Plan goals on health and human capital development.
In this project, ECAIR works closely with BLSS-SHD, which serves as the process owner and primary partner office, while OUOPS and the EdTech Hub are involved in discussions on implementation and sustainability. ECAIR's role is to develop, validate, and govern the computer vision workflow before any broader rollout.
The NCR pilot focused on improving the model by collecting additional training data to enhance its robustness. The team has met the performance threshold defined by the process owner, allowing it to proceed with further field testing and continued model improvement. The next step is a proposed 10-school calibration cohort to support this continued field testing and model improvement.