A tool to catch cancer early

In early 2016, IT researcher Geetha Manjunath lost two close family members to breast cancer. These tragic events came as a huge shock for Manjunath.

“They were both relatively young. I was struck by the impact the disease has on the patient and the caregivers and began to study more about breast cancer. I began to study the global breast cancer scene more closely.”

In the course of her research, she stumbled upon methods such as thermography and many new techniques. All the research eventually resulted in Manjunath co-founding Niramai, a health-tech startup that uses artificial intelligence (AI) to detect cancer in the early stages, with her colleague, Nidhi Mathur. The main draw was of using non-invasive, radiation-free, and painless methods to detect cancer, especially in the early stages.

In the last two years, Niramai has over 30 installations at hospitals and diagnostic centres across 10 cities. The firm has nine US patents and a Canadian patent and is the only Indian company to figure in the latest edition of the 2019 cohort of AI 100 Startups in the world by global business data intelligence platform CB Insights. It has received regulatory clearances from the Ministry of Health and Family Welfare, Drug Control General of India (DCGI), and Central Drugs Standard Control Organisation (CDSCO).

In 2017, they raised seed funding led by pi Ventures, with participation from Ankur Capital, Axilor Ventures, 500 Startups and Flipkart co-founder Binny Bansal. The company has raised about $6 million in a recent round led by Japanese VC firm Dream Incubator, Beenext and other investors.

What do they do?

Early detection was a key tool in ensuring that the patient could be cured completely.

“In many cases, the cancer is detected at a later stage and it is too late to help, as patients wait for symptoms to appear such as a lump or a discharge of liquid before heading to the hospital. There was a lack of a regular preventive screening method for breast cancer detection. Moreover, mammograms and ultrasounds were expensive and in cases of women under the age of 40, not very effective in identifying tumours.” says Manjunath.

The firm uses Thermalytix, a computer-aided diagnostic engine that is powered by artificial intelligence. It uses a high-resolution thermal sensing device and an analytics solution on cloud for analysing the thermal images. It can give results in quick time, detect smaller tumours and costs much lesser than full-fledged mammograms. “It has a thermal sensor to detect temperature variations, and the findings are analysed using artificial intelligence, to detect any abnormalities and to state if it requires immediate attention. It does not require any touch as well,” adds Manjunath.

How does it work?

Manjunath explains the process, “A thermal imager is placed at some distance from the patient. The sensor captures images and temperatures of the chest. There is no contact, and there is a screen between the person and the technician. It’s like a changing room experience. It works out cheaper both for the patient and the hospitals as well and is completely painless.”

It took them about six months to build the software and AI algorithms. It was first set up at the BMS Hospital in Bengaluru in 2017. The device can detect over 4 lakh temperature points per person, Manjunath states. “The platform can detect multiple thermal patterns and combinations, which then gives a health score. If the score is high, the person needs a follow-up test. “