This Hyderabad couple uses ML technology to fight deadly neonatal disease

The start-up is also developing an automated peritoneal dialysis device in which the dialysate solution is inserted through the abdomen and performance parameters are monitored using IoT

India has one of the world's highest incidences of Neonatal Sepsis, a deadly bacterial blood-stream infection.

According to a Lancet review paper published in 2018, India has around 17,000 cases per 1,00,000 live births while the case fatality rate of sepsis among newborns ranges between 25 per cent to 65 per cent.

Avyantra, a Hyderabad-based start-up, has built a machine-learning platform to aid early risk assessment of neonatal Sepsis and treatment decisions, including antibiotic therapy. The risk score is based on algorithms developed from original data collected from hospitals.

Hyderabad-based couple Hyma Goparaju and KVKLN Rao, both with more than a decade of experience in healthcare technology, founded the start-up in 2017. 

The couple had lost a family member to complications arising from the disease a decade ago and was shocked to find it was even more critical in babies. Doctors usually administer empirical antibiotics in such cases and it becomes dangerous and even fatal if the babies become antibiotic-resistant. "A lot of times entrepreneurs don't need a personal trigger to solve societal issues. But I don't know what else could have been a bigger trigger for us to solve this problem," said Goparaju.

The start-up was also born out of the "culmination" of Rao's 20-year-plus experience of working with business analytics and healthcare technology in the IT industry and Goparaju's learnings from her PhD in business management.

"Sepsis is very hard to detect and diagnose in neonates (aged from 0-28 days) especially in rural India due to lack of medical care and specialists," said Goparaju. "Now, the algorithms developed would help doctors and health workers in early assessment and ensure quicker treatment."

The machine learning-powered PreSco platform sends out a risk score that ranges between 0 and 10 (low to high risk), and helps the doctor or health worker assess whether the baby needs greater medical attention and more antibiotics.

Non-invasive and observable data points, such as the mother's blood pressure, Group B Strep (GBS) infections and baby parameters such as appearance, distension, skin colour, urine output, are taken as inputs to arrive at the score. Invasive parameters like blood C-reactive protein (CRP), white blood cells (WBC) counts are also checked. 

The start-up is also developing an automated peritoneal dialysis device in which the dialysate solution is inserted through the abdomen. The company aims to provide cost-effective dialysis. "It uses the internet of things to continuously monitors patient performance parameters that come to use in rural areas. We are hoping to complete product development by the end of this year," Goparaju said.

Avyantra and another Hyderabad-based enterprise, StaTwig, were among the eight start-ups globally to receive the UN arm's crypto fund last month.

Eight technology start-ups in developing economies received investment from the Unicef's Cryptocurrency Fund (CryptoFund) last week to solve local and global challenges.

The crypto fund will invest 125 Ether coins worth nearly $30,000 each in these eight firms – from seven countries – to develop prototypes, pilot, or scale their technologies over six months. The eight firms are: Afinidata, Avyantra, Cireha, Ideasis, OS City, StaTwig, Somleng and Utopic.

Goparaju is also a fiction writer and has authored two books — The Abonadoned Daughter which is a fictional take on female infanticide set in British India. Her first book, The Withering Banyan, revolves around a turbulent father-son relationship and many of the members are suffering from a genetically transferred brain disorder.

The health-tech firm will use cryptocurrency to enhance the platform's technological capabilities and work on additional features.

Enable GingerCannot connect to Ginger Check your internet connection
or reload the browserDisable in this text fieldEditEdit in GingerEdit in Ginger×