What Happened
- Union Minister Dr. Jitendra Singh announced that India's largest pregnancy cohort study — the GARBH-INi initiative — has enrolled 12,000 women and is developing indigenous, AI-driven tools for the early prediction and prevention of preterm births.
- The study has generated over 1.6 million well-characterised biospecimens and more than one million ultrasound images, forming a comprehensive national biorepository.
- GARBH-INi has also established the GARBH-INi-DRISHTI data-sharing platform, enabling wider access for the research community.
- The initiative integrates clinical epidemiology, multi-omics biomarkers, and AI to create personalised risk prediction models tailored to the Indian population.
- Preterm birth — defined as birth before 37 weeks of gestation — is the leading cause of neonatal mortality in India and a major contributor to the global burden of under-five deaths.
Static Topic Bridges
Preterm Birth: India's Public Health Challenge
India has one of the highest absolute numbers of preterm births globally, contributing nearly 3.5 million preterm births annually — the largest national burden in the world. Preterm birth accounts for approximately 35% of all neonatal deaths in India. Complications from prematurity include respiratory distress syndrome, sepsis, and long-term neurodevelopmental disabilities. Despite progress in neonatal care, survival rates for very preterm infants (born before 28 weeks) remain low in low-resource settings across India. Standard risk screening tools developed in high-income countries perform poorly in Indian populations due to differences in nutrition, genetics, and antenatal care patterns — making indigenous AI models an urgent necessity.
- India contributes ~23% of global preterm births annually
- Preterm birth is the leading cause of newborn death in India
- Risk factors in India: malnutrition, anaemia, infections, teenage pregnancy, poor antenatal access
- Existing scoring tools (developed for Western populations) are not validated for Indian contexts
Connection to this news: GARBH-INi's focus on building an India-specific AI-based gestational age and risk model directly addresses the gap between global tools and Indian clinical realities.
GARBH-INi Initiative: Architecture and Science
GARBH-INi (Gestational Age Research in Births — India initiative) was launched in May 2015 as a prospective hospital-based observational cohort study, initially at Gurugram Civil Hospital, Haryana, under the Department of Biotechnology (DBT). The project brings together the THSTI (Translational Health Science and Technology Institute), IIT Madras (Robert Bosch Centre for Data Science and AI), and other research institutions. Its core scientific output has been the Garbhini-GA2 model — a machine learning model for estimating gestational age in the second and third trimesters — published in The Lancet Regional Health Southeast Asia. By training on Indian women's ultrasound data, the model outperforms WHO-standard tools (originally derived from European populations) for Indian pregnancies.
- Lead institution: THSTI, Faridabad (under Department of Biotechnology)
- Study launched: May 2015, Gurugram Civil Hospital
- Key output: Garbhini-GA2 — Indian population-specific gestational age model (published in The Lancet)
- Biospecimen repository: >1.6 million samples; ultrasound image dataset: >1 million images
- Data platform: GARBH-INi-DRISHTI (open-access for researchers)
Connection to this news: The 12,000-women cohort announced by the Minister is not a new launch but a milestone in an ongoing flagship study — signalling maturation of the initiative into a national-scale AI health programme.
Artificial Intelligence in Healthcare: Policy and Potential
India's National Digital Health Mission (ABDM) and the National Health Policy 2017 both identify AI and digital health as priority areas. The Union Budget 2024-25 announced three CoEs (Centres of Excellence) for AI in healthcare, agriculture, and education. In the health domain, AI is being applied for early disease detection (tuberculosis, diabetic retinopathy, cancer screening), drug discovery, and precision medicine. The GARBH-INi-DRISHTI data-sharing platform — built around 12,000 well-characterised patient records — exemplifies federated biorepositories that can accelerate AI model training without compromising patient privacy.
- ABDM (Ayushman Bharat Digital Mission): creates unique Health IDs (ABHA), links health records digitally
- MoHFW's National Health Stack: interoperability layer for digital health data
- Government target: AI-driven diagnostics in 25,000+ Health and Wellness Centres
- Multi-omics integration: combining genomics, proteomics, metabolomics with clinical data for holistic risk models
Connection to this news: GARBH-INi represents a model for how India can build AI health tools grounded in indigenous data — avoiding the "data colonialism" of relying solely on Western-trained algorithms for Indian clinical decisions.
Mother and Child Health: Constitutional and Policy Framework
India's approach to maternal and child health is anchored in the National Health Mission (NHM), specifically the Reproductive and Child Health (RCH) programme. The Janani Suraksha Yojana (JSY) and Pradhan Mantri Surakshit Matritva Abhiyan (PMSMA) are institutional delivery and antenatal care schemes targeting high-risk pregnancies. India's Maternal Mortality Ratio (MMR) has declined from 254 (2004-06) to 97 (2018-20) per 100,000 live births. Neonatal Mortality Rate (NMR) stands at 20 per 1,000 live births (2020), still above the SDG target of 12 by 2030. GARBH-INi's AI tools, once validated, could integrate with PMSMA-linked risk screening at government hospitals.
- MMR target under NHM: <70 per 100,000 live births (SDG Goal 3)
- NMR target under SDG 3.2: <12 per 1,000 live births by 2030
- Janani Suraksha Yojana: conditional cash transfer for institutional delivery; ~10 million beneficiaries annually
- India's NMR (2020): 20 per 1,000 live births — still above SDG target
Connection to this news: The AI tools being developed under GARBH-INi have potential to be integrated into existing government antenatal care infrastructure, directly contributing to NMR reduction and India's SDG commitments.
Key Facts & Data
- GARBH-INi enrolled 12,000 pregnant women (India's largest pregnancy cohort)
- Launched: May 2015, Gurugram Civil Hospital; led by THSTI under Department of Biotechnology
- Biospecimen repository: >1.6 million samples; ultrasound dataset: >1 million images
- Key AI output: Garbhini-GA2 model — Indian-specific gestational age estimation (published in The Lancet Regional Health Southeast Asia)
- India contributes ~23% of global preterm births; ~3.5 million preterm births annually
- Preterm birth causes ~35% of neonatal deaths in India
- Data platform: GARBH-INi-DRISHTI (open-access research sharing platform)
- India's NMR (2020): 20 per 1,000 live births; SDG 2030 target: <12