In a groundbreaking collaboration, researchers from the Indian Institute of Technology Madras (IIT Madras) and Translational Health Science and Technology Institute (THSTI) have unveiled an India-specific artificial intelligence (AI) model designed to precisely determine the age of a fetus in pregnant women during the second and third trimesters. This cutting-edge research, conducted as part of the ‘Interdisciplinary Group for Advanced Research on Birth Outcomes – DBT India Initiative’ (GARBH-Ini) program, addresses the challenges faced in accurately estimating gestational age (GA) in the Indian context.
Accurate GA estimation is pivotal for providing optimal care to pregnant women, monitoring fetal well-being, and identifying and managing pregnancy complications. However, in India, late initiation of antenatal care poses a significant challenge, often beginning as late as 14 weeks after gestation. This delay complicates traditional methods like the last menstrual period (LMP), given uncertainties in LMP recall and irregular menstrual cycles.
The ethnic diversity in India further complicates GA estimation, with variations in fetal growth patterns and biometric measurements not accounted for in Western-centric models. Maternal nutrition, health conditions, and genetic factors contribute to diverse fetal growth patterns, necessitating population-specific models to ensure accurate GA assessment.
Introducing Garbhini-GA2: A model tailored for India
To address these challenges, the researchers developed the Garbhini-GA2 model using data from the GARBH-Ini cohort study, a comprehensive examination of clinical data from pregnant women in India. Analyzing this data allowed researchers to identify key parameters influencing GA estimation in late trimesters. Garbhini-GA2 incorporates unique fetal biometry and growth patterns observed in the Indian population.
Comparative evaluations with existing formulas, such as Hadlock and InterGrowth-21st, showcased the superior accuracy of Garbhini-GA2. Metrics including root-mean-squared error, bias, and pre-term birth rates revealed that Garbhini-GA2 significantly reduced median error in GA estimation by more than three times compared to the Hadlock formula. This highlights its superior suitability for the Indian population.
Dr. Rajesh Gokhale, Secretary of the Department of Biotechnology, Government of India, praised the development of population-specific models, emphasizing their commendable outcomes within the GARBH-Ini initiative. He noted that these models are currently undergoing validation across the country.
Once validated, Garbhini-GA2 holds the potential to revolutionize prenatal care in India. Deploying this model in clinics across the nation could enhance the care delivered by obstetricians and neonatologists, reducing maternal and infant mortality rates.
Future prospects and impact on healthcare
This pioneering research, led by Dr. Himanshu Sinha of IIT Madras and Dr. Shinjini Bhatnagar, Principal Investigator of GARBH-Ini at THSTI, is poised to transform prenatal care. Accurate GA estimation is crucial in managing complications like gestational diabetes and preeclampsia, enabling specific monitoring and treatment based on the stage of pregnancy. Moreover, it ensures the reliability of data for research on outcomes such as stillbirth, preterm birth, and fetal growth restriction.
The development and potential deployment of the Garbhini-GA2 model mark a significant leap forward in addressing the unique challenges of gestational age estimation in India. This AI-driven innovation promises to enhance healthcare outcomes for mothers and infants, providing a tailored and precise approach to gestational age determination. As the model undergoes validation, its successful implementation in clinics nationwide could usher in a new maternal and neonatal care era, cementing India’s position at the forefront of innovative healthcare solutions.
