Big Idea

Early warning system

With the help of AI and machine learning, Niramai is saving countless lives, one breast scan at a time

The world has transformed beyond imagination over the past few decades. Man has achieved unthinkable feats — from self-driving cars and 3D printing a two-storeyed building to gene editing and augmented reality. But despite much progress, we are still defeated by our very own bodies. Particularly, uncontrollably and abnormally multiplying cells, what we call cancer. In 2018 alone, breast cancer — the second most common variant — claimed over 87,000 lives. 

Geetha Manjunath, a computer scientist from IISc, Bengaluru, suffered the loss of a close cousin to breast cancer in 2013. The shock of it led her to hunting for solutions and eventually founding Niramai, short for Non-Invasive Risk Assessment with Machine Intelligence — a company that has developed tech to detect breast cancer. There are detection technologies already, such as mammograms, but those radars were not catching every blip.

Young tech 

Manjunath tells us that over 50% of the women diagnosed with breast cancer are below the age of 50, but mammography, or simply put, an x-ray scan of a breast, gives accurate results only for women above 40-45 years of age. That’s where the trouble starts. 

Geetha Manjunath Co-founder, NiramaiIn younger patients, the tissue density in the breast is high, and the fibroglands are more densely packed. Dense tissue shows up as white sections in a mammogram and cancerous tissue is difficult to spot through that. Manjunath’s cousin wasn’t even 40 years old when she succumbed to the disease. Hence, she was determined to figure out a solution, and she roped in Nidhi Mathur, an IIM-B alumnus and her former colleague at HP. In July 2016, they opened Niramai.

“We wanted to do something that doesn’t involve x-rays,” says Manjunath when she explains why t


You don’t want to be left behind. Do you?

Our work is exclusively for discerning readers. To read our edgy stories and access our archives, you’ve to subscribe