“We wanted to address all issues that hindered early detection of breast cancer”

September 7, 2019 0 By FM

Breast cancer is the most common cancer among Indian women in terms of incidence as well as mortality. Its prevalence in younger age-groups in India is higher than the global average. With 40% to 50% deaths, the country continues to have one of the lowest survival rates. The main reason for the low survival rate, despite the potential of today’s new treatments to even cure breast cancer, is clearly lack of early detection. Most of the patients are diagnosed at the third or fourth stage, by when treatment options are limited. Access to mammography, the current gold standard for breast cancer screening, is poor as providing it entails high costs for equipment and experienced radiographers. Even if it is accessible in the tier-1 and tier-2 cities, it is recommended once every 2 years and only to women above 45 years because it cannot identify tumours effectively for younger women. Clinical examination, the other common method, can typically detect tumours only after they are large enough to be palpable. Another big dilemma is the aversion to mammography even for the aged, as it is a painful experience.   

In this context, only an innovation addressing the issues, like cost, ease of use and accessibility, can make adequate impact in this market. Geetha Manjunath, Chief Executive Officer, Niramai Health Analytix, one of the world’s top 100 AI start-ups and the lead brain behind a most sought-after breast cancer detection solution, says in Straight Talk with the editor CH Unnikrishnan that her team’s primary and most important challenge while developing the solution was to address all the real issues that hindered early and more specific detection of cancer. Edited excerpts:

Mammography is the gold standard and the most used screening solution for breast cancer worldwide. But what made you to think about a different concept?

No doubt, Mammography is the most used screening method worldwide at present. But the context of our thinking was different as the priority focused in our project were early detection as well as making it suitable for the young. Early diagnosis is very critical in saving the life of a cancer patient. So it is all the more important in India as the susceptible age group in the country is much younger.  In this situation, we wanted a solution that can resolve not only these issues, but also to address other issues such as affordability and accessibility. Mammography is not recommended for women under 45 and it also uses X-rays for scanning, which can make women more susceptible to cancer if screened multiple times. At the same time, false positives with traditional methods and non-targeted treatments have been the other big challenge that worsens the situation often.

Thermal imaging as a technique used in medical diagnosis has been around for many years. But it never progressed into a fully reliable one. But your innovation, taking this as a base and enhancing it with AI analytics and other cross-verification tools, has really made it a revolutionary solution for breast cancer detection. Can you elaborate on the technology? 

As you rightly mentioned, it works on thermal imaging technology. Our solution—Thermalytix, is a computer-aided diagnostic engine powered by artificial intelligence (AI). It uses a high-resolution thermal sensing device and a cloud-hosted analytics solution for analysing the thermal images. This SaaS solution uses big data analytics, artificial intelligence and machine learning for reliable, early and accurate breast cancer screening. In simple terms, it acts like a thermometer kept a little far away from the object (the woman) to measure different temperature values at the
chest and the high-resolution thermal sensing camera attached to it captures almost 400,000 thermal images of the person based on the variation in thermal values on the entire chest. These huge data from the individual is analysed using AI algorithms to detect the tumour and then it is subjected to comparative analysis using other diagnostic tools (adjunct modalities) to ensure accuracy. This way, it avoids false positives almost fully.   

The cancer screening software that uses machine intelligence over thermography images thus enables an easy to use and portable solution that requires minimal human supervision. Certainly, the other biggest advantage is universal accessibility. In a way, we wanted to address all the issues that hindered early detection of breast cancer, crucial to increasing survival rate, by combining traditional and new-generation technologies.     

Early detection is key to ensure better survival rates. How effective is your screening method this way?  

It can detect cancer at a much earlier stage than traditional diagnostic methods and self-examination, as it detects cancer long before a lump can be felt.

Breast cancer screening using this method can detect tumours 5 times smaller than what clinical exam can detect. It is also non-contact, painless and free of any radiation, apart from being low-cost. So, women of all age groups can undergo frequent screening without any side-effects. As it enables an early detection of cancer, that too without triggering any of the challenges associated with the already established methods, it helps medical intervention at a preventable stage and helps increase the survival rate.

What is its accuracy and sensitivity?

We have established in all our trials so far that the sensitivity is over 90%. It has also achieved 27% higher accuracy as compared to mammography and about 70% higher positive predictive value
than a visual interpretation of thermography. What we have successfully proved in this method is that it is effective in 32% more patients walking into hospitals with pain or discomfort in their breasts. With this method, we were able to detect cancer lesions as small as 4 mm with no palpable lumps. In fact, we have even detected cancers without lumps as our target is not exactly the lump but the abnormality.   

How big is the number of subjects who have undergone trials and routine screening so far?

Our cancer screening tool has been tested on more than 9,000 women in more than 30 settings, including hospitals, diagnostic centres as well as screening camps, in the last one year. 3 clinical trials, comparing Thermalytix with the current standard of care, have been presented and published in peer- reviewed conferences and journals. The results from these clinical trials prove the very high accuracy of this method, which is comparable to or perhaps even better than mammography.