The world is observing an exponential growth in the field of artificial intelligence (AI) and digital technologies. This has major potential to support maternal and neonatal healthcare in low-resource settings, although the development of these technologies is in its nascent stage. AI solutions targeted to improve perinatal health and impacting clinical care require societal acceptance and integration within the Indian public health ecosystem, supplemented by the trust of involved stakeholders for wide scale deployment. This approach demands engagement from every faction of society: from healthcare professionals to government stakeholders and policy makers1.
A comprehensive and multidisciplinary approach is required to improve the health of mothers and children in low-resource settings. AI can transform and supplement conventional practices in perinatal health, thus improving the accuracy of diagnoses and increasing access to care, ultimately saving lives.
According to the World Health Organisation (WHO), maternal health refers to the health of women during pregnancy, childbirth and the post-partum period, whereas perinatal health refers to health from 22 completed weeks of gestation until 7 completed days after birth representing a vital period for maternal and infant health. A healthy start during the perinatal period is critical and has potential to influence infancy, childhood and adulthood.
As per the SRS Bulletin, India’s maternal mortality rate (MMR) was 97 deaths per 100,000 live births in 2018–20, and the infant mortality rate (IMR) was 35.2 deaths per 1,000 live births in 2019–212. Most maternal deaths occur during the perinatal period where common identified causes such as haemorrhage, infections, hypertensive disorders during pregnancy, pre-existing chronic health conditions, complications from delivery are preventable in nature, and usually are associated with the lack of access to quality health care services. Similarly, the leading causes of neonatal/infant mortality are prematurity and low birthweight which account for 45.5 percent of deaths during the first 29 days of a newborn in India. States with a high population, and belonging to the Empowered Action Group (EAG) with states such as Uttar Pradesh, Bihar, Chhattisgarh, Jharkhand, Rajasthan, Madhya Pradesh, etc., lag behind in terms of progress and contribute significantly towards poor health outcomes. Despite all the challenges faced over the last two decades, considerable progress has been made in perinatal health and one of the flagship and monumental milestones is the Reproductive, Maternal, Newborn, Child, Adolescent Health Plus Nutrition (RMNCAH+N) strategy. This strategy promotes links between various interventions across thematic areas to enhance coverage throughout the lifecycle to improve maternal and child survival in India. The strategy succeeded in establishing an institutional framework and a mechanism of facility-level, need-based planning through concentrated and coordinated efforts focusing on reducing perinatal adverse outcomes (infant mortality rates, maternal mortality rates) and to ensure quality services to pregnant women and children across the country. However, they are still a public health concern, and considerable disparities exist within and between states.
Artificial intelligence offers novel approaches to prediction modelling, diagnosis, early detection, and monitoring in perinatal health to untangle a web of causation for maternal and child health outcomes with potential to accelerate this progress. Given the advancements being made under the ambit of Health System Strengthening (HSS) via Pradhan Mantri – Ayushman Bharat Health Infrastructure Mission (PM-ABHIM) and the Ayushman Bharat Digital Mission (ABDM) program, we are looking at creating one of the largest clinical databases in the world. This could be used for creating predictive analytical models powered by AI.
Artificial intelligence and machine learning algorithms in perinatal health
Machine learning (ML) algorithms and pattern recognition via foundation models, generative AI, and large language models (LLM) are gaining significant traction in the AI space with an anticipated rise in electronic health records and audio-visual data points. Real-time electronic health recording and predictive modelling using AI has the potential to revolutionise public health systems in India. Such tools could have a significant impact on the approaches and pathways required to build predictive algorithms and compute the probability of certain adverse outcomes happening prior to the actual events.
Unlike parametric models in statistics, fewer assumptions are required in modelling via ML3. ML models have the potential to perform better than traditional statistical models because of their ability to deal with nonlinear complex data, multiple interactions between variables, and handle multiple predictors and chains of events simultaneously4.
The way forward and the need for a comprehensive approach
AI-powered technologies have the potential to supplement and improve prenatal diagnosis of birth defects as well as fetal abnormalities. While using these technologies, critical focus areas include i) the availability of routine clinical data that is representative of the population (rich multimodal data of large sample size) and ensuring the quality of such data points, ii) the modification of existing artificial intelligence modalities and using them in perinatal health, and iii) augmenting existing health care decision-making processes via AI.
Such predictive modelling has limitless potential to change the Indian health ecosystem. Besides tabular data, ML pattern recognition algorithms could also be trained using audio-visual data such as images, videos, X-rays, sonograms, echocardiograms, heart sounds, etc. and put to use towards creating population-based screening tools for various healthcare programs.
AI for perinatal health is in its nascent stages. If developed methodically and in collaboration with health specialists, AI research can benefit clinical aspects of perinatal health and even research into applied AI at large. AI can significantly enhance the clinical impact in perinatal healthcare by providing researchers and clinicians with more efficient tools. These advanced AI systems have the capability to predict conditions with greater precision and at an earlier stage compared to existing methods. Consequently, the use of AI in perinatal health care can lead to more accurate and timely interventions, ultimately improving patient outcomes.
Dr Dhwani Almaula, Program Manager, Maternal, Newborn and Child Health (MNCH), Wadhwani AI
(DISCLAIMER: The views expressed are sole of the author and ETHealthworld does not necessarily subscribe to it. ETHealthworld.com shall not be responsible for any damage caused to any person/organisation directly or indirectly)
