In a significant breakthrough, researchers from the Indian Institute of Technology Madras (IIT Madras) and the Translational Health Science and Technology Institute (THSTI) have successfully developed the first India-specific Artificial Intelligence (AI) model for accurately determining the age of a foetus during the second and third trimesters of pregnancy. This innovative model, named ‘Garbhini-GA2,’ aims to address the limitations of existing Western population-based formulas, potentially revolutionizing pregnancy care in India.
Accurate estimation of ‘Gestational Age’ (GA) is crucial for providing appropriate care to pregnant women and ensuring precise delivery dates. Existing models, designed for Western populations, often lead to errors when applied to the diverse growth patterns of foetuses in the Indian population. The ‘Garbhini-GA2’ model, developed and validated using Indian population data, significantly reduces estimation errors, thereby enhancing the quality of care provided by obstetricians and neonatologists.
The key benefits of ‘Garbhini-GA2’ include:
- Precision for Indian Population: Unlike existing models developed for Western populations, ‘Garbhini-GA2’ accurately estimates foetal age for the Indian population, minimizing errors by almost three times.
- Improved Maternal and Infant Care: The enhanced accuracy of this GA model has the potential to improve the care delivered by healthcare professionals, leading to a reduction in maternal and infant mortality rates in India.
Welcoming the research, Dr Rajesh Gokhale, Secretary, Department of Biotechnology (DBT), Government of India, expressed appreciation for the commendable outcome of the ‘Interdisciplinary Group for Advanced Research on Birth Outcomes – DBT India Initiative’ (GARBH-Ini) programme.
The study, led by Dr Himanshu Sinha of IIT Madras and Dr Shinjini Bhatnagar of THSTI, was published in the prestigious international peer-reviewed journal Lancet Regional Health Southeast Asia. The validation of these population-specific models for estimating gestational age is currently underway across the country.
Highlighting the importance of the study, Dr Himanshu Sinha emphasized IIT Madras’ commitment to solving healthcare problems at the grassroots level in India. The collaboration with clinical partners, particularly THSTI, showcases the utilization of advanced data science and AI/ML techniques to develop tools that predict unfavourable birth outcomes.
Dr Shinjini Bhatnagar, Principal Investigator of the GARBH-Ini programme, underscored the critical role of improved GA accuracy in the broader goals of reducing adverse pregnancy outcomes. The study exemplifies the synergy between clinicians and data scientists, ensuring that technological advancements are not only technically sound but also clinically relevant and seamlessly integrated into healthcare workflows.
The ‘Garbhini-GA2’ model, developed using genetic algorithm-based methods, outperformed existing models such as Hadlock and INTERGROWTH-21st in accuracy during the second and third trimesters. Once validated in pan-India cohorts, this model holds the potential to significantly enhance pregnancy care outcomes, ultimately reducing maternal and infant mortality rates across the country.
This groundbreaking research was conducted in collaboration with Gurugram Civil Hospital, Safdarjung Hospital, Christian Medical College Vellore, and Pondicherry Institute of Medical Sciences, with funding from the Grand Challenges India programme of the Biotechnology Industry Research Assistance Council, DBT, Government of India, and additional support from the Robert Bosch Centre for Data Science and Artificial Intelligence (RBCDSAI), IIT Madras, and the Centre for Integrative Biology and Systems Medicine (IBSE), IIT Madras.
