How a Non-Profit is Working on AI-Induced Tool to Help Asha Workers

Down South, in Kerala, there is a rage among the Asha workers, considered to be the backbone of India's public health system, who have been protesting in the state for months now. While the demands behind the protests include a pay hike, recognition, and benefits, Asha workers also face various hurdles in their day-to-day lives in the operations.

The advent of one of the newest technologies, AI, might improve the functioning, going by Wadhwani AI, an independent non-profit organisation building and deploying AI-powered solutions for underserved communities in the developing world. For the workforce consisting of over 8.46 lakh ASHAs, and their role, which is critical in ensuring newborns thrive, the organisation has come up with 'Shishu Maapan', an AI-powered tool developed by the Indian-American tech entrepreneurs and founders, Romesh and Sunil Wadhwani.

Underscoring how the traditional weight monitoring, which is a key indicator of an infant’s health, remains a challenge due to bulky, unreliable weighing scales and manual record-keeping, the founders have been piloting this technology in Daman and Diu. 

The technology is aiming to reduce the dependence on the heavy weighing scales that are carried across villages or urban settlements while ensuring a minimisation of manual errors for A "precise newborn growth tracking". "During Home Based Newborn Care (HBNC) visits, these workers often encounter challenges such as supply issues, maintenance problems, performance discrepancies of spring balances, and inaccurate and cumbersome weighing scales all of which can lead to inefficiencies and errors. AI-driven solutions like Shishu Maapan address these issues by automating some of the most labor-intensive and error-prone tasks," Dr. Sneha Nikam, Senior Program Manager, Wadhwani AI says. 

For instance, Dr. Nikam continues, "Shishu Maapan allows ASHA workers to simply take a short video (10–15 seconds) of a newborn from an overhead angle. The AI then analyses this video to predict five key anthropometric measurements, mainly weight, length, chest circumference, head circumference, and mid-upper arm circumference. These are critical indicators of a newborn's health and growth, and accurate tracking of them is vital for early identification of referrals and ensuring timely interventions." She adds that with real-time, data-backed insights, ASHA workers can counsel mothers more effectively, enhancing trust within communities.

Wadhwani AI has been working closely with the government ministries—particularly the Ministry of Health and Family Welfare, to identify areas within existing healthcare programs where our AI solutions can complement and strengthen ongoing efforts. 

Talking about the Shishu Maapan, Dr. Nikam says that the idea is to integrate seamlessly into the existing health infrastructure. "Before deploying Shishu Maapan, we collaborated with the state government to gather local
data and contextualise the AI model for the new geography . This ensured that the AI model would function effectively for the deployment geography," she says.

But are we at par with technology? 

Despite the promise of AI in improving healthcare delivery, ASHA workers may still encounter several challenges when using AI-powered tools.The lack of access to technology might seem to be the prominent of all. "In many rural areas, ASHA workers may not have access to basic smartphones with sufficient processing power or cameras of the required quality," Dr. Nikam states. 

Even though Wadhwani AI has designed Shishu Maapan to work with basic smartphones in offline mode, this challenge remains a barrier in some regions where mobile phones may not be available or fully functional.. she notes. 

The cultural barrier does not seem to be tackled so easily, posing another set of problems for the Asha workers. "There is often resistance to technology adoption from health workers who may not be familiar with AI or digital tools. Despite AI tools being user-friendly, some ASHA workers may be skeptical about using technology in their everyday tasks, particularly if they have limited exposure to smartphones or apps," the Program Manager shares while adding that overcoming this skepticism and ensuring proper training for ASHA workers is crucial for the successful adoption of the AI.

Keeping such potential problems along the way, however, the organisation is working to mitigate them beforehand. As per Dr. Nikam, Wadhwani AI is directly coordinating with the public health officials, state governments, and health ministries to ensure the technology aligns with the existing health infrastructure. Further, the team is working on the contextualisation for local needs. "In Shishu Maapan, we ensure that the AI model is fine-tuned with data from a state before deployment within that state. The process helps validate the AI’s model performance and ensure contextualisation before its direct use in the field, thereby enhancing its performance," Dr. Nikam says.

In the rural India where connectivity is still a problem, and since many frontline workers, lack constant access to the internet, Wadhwani AI has designed the Shishu Maapan to function in offline mode. Dr. Nikam states that this will allow the workers to take measurements and receive results without needing to be connected to the internet, and the data can be synchronised with central systems when the worker gains access to internet connectivity.

As of yet, the technology is stil in the nascant stage and has not spread to the other parts of the country.