Detect Breast Cancer At A Much Earlier Stage With This Software Developed By Dr Geetha Manjunath
- IWB Post
- June 19, 2019
One out of two women diagnosed with breast cancer dies within five years in India. So we spoke to Dr. Geetha Manjunath, co-founder & CEO of NIRAMAI, a startup developing a novel breast cancer screening solution.
Dr. Geetha, with more than three decades of experience in the IT industry, is among the pioneers who envisioned integration of the advanced technology with healthcare. “Early detection is key to saving lives,” is something she heavily emphasized on. At NIRAMAI, she is developing a novel software to detect breast cancer at a much earlier stage than traditional methods or self-examination. The core technology of the solution is an artificial intelligence-led diagnostic platform that uses patented thermal image processing and machine learning algorithms for reliable and accurate breast cancer screening.
Dr. Geetha also shares her thoughts on the current industry trends, educational infrastructure and tips on dealing with gender disparity at the workplace. Excerpts from the conversation:
Niramai Health Analytics is a breast cancer screening solution. Could you start by telling us more about it?
Niramai is an acronym for Non-Invasive Risk Assessment with Machine-learning and Artificial Intelligence. It adopts a new method of doing breast cancer screening. The highlight of the screening process is that it is non-contact, non-invasive and privacy aware. It is an entirely radiation-free screening process that can be used to detect early stages of breast cancer. This service can be availed by women of all age groups. It is low cost and also portable, so the entire process can be the set up anywhere, even in rural areas.
While going through your website, one thing in the description struck me – “No one touches or even sees you.” How does the breast examination work then?
The way the entire process works is simple and time-saving. The woman enters a small closed booth in the diagnostic center, and two minutes of imaging is done, while the woman sits on a chair disrobed. What happens during the imaging is that the sensor is at a three feet distance from the woman’s chest, measures the temperature variations on her chest. All of this is done with the technician being outside the booth. Once the imaging is done, within ten minutes, the report is also ready.
So how and where can women avail this service?
Women of all age groups can avail this service at multiple diagnostic centers in Bengaluru, Dehradhun and Pune (as of now). Women can avail the service as a preventive health check up package. It is also available as a separate test, where one can come in, just to get a imaging of their breasts done. We also cater to corporate clients. As the entire service can be set up just about anywhere because of its portability, we have conducted numerous breast screening camps in offices, apartment complexes etc.
Could you tell us more about the image processing technique that Niramai uses – Thermography?
We use a high resolution thermal imaging device, which measures about four lakh data points per person. It records how the temperature differs on a woman’s chest – basically the heat variations. But by just looking at the thermal image output, we cannot identify if there is any anomaly in the health of her breasts. That’s where the AI algorithm on the thermal images detects any abnormality. So it’s all about the use of newer technologies such as advanced computer image processing, AI, machine learning, deep learning to build our diagnostic software. We do have 10 patents filed in the US and 4 have already been granted.
You have brought automation into a personal service such as healthcare. Could you tell us about the measures you take to prevent errors?
From day one, we have adopted the management principle of garbage in, garbage out. So we have always made sure that the data we collect is always error free. We have also invested heavily in training the technicians, so that there is no manual errors in data collection. Next is we have made sure that the software for image processing is up to date, so that we don’t have to spend more time in redoing the image. Even after doing 2500 patients, we have always included the doctors approval at the end of the report before giving the results to the patients. In this way, we are extremely cautious in all stages of the the process.
At Niramai you have integrated IT with the Healthcare sector. How big do you think is this industry currently?
When it comes to tech start ups, it depends on what the founder wants to do with his company. Healthcare in this respect (start up) is not just about making a lot of money. It is more about creating something more useful to the community at large. So it’s safe to say that currently, the market hasn’t picked up much pace. It requires a lot more patience and time, than other start ups. So people who sign up for a healthcare tech start up, need to be in it dedicated for at least 5-10 years.
But yes, the market potential is absolutely huge, the main reason is that healthcare is probably the only domain not touched by the tech start up culture. Specially in counties like India, where there is a huge divide between the number of doctors and patients, the need for tech integrated healthcare services is the need of the hour. Also, the regulatory measures in this respect are not so stringent in India as it is in the US, so India is one of the most ideal markets to pilot many such innovations.
Data Science has become the buzzword these days, but from a layman’s perspective, there is still a lot of ambiguity around this topic. Could you help our readers understand this subject more?
To explain the concept very lucidly, it is the science of collecting and analyzing data. Basically creating new insights and value for the data, given the digitization of the world. There are different layers or stages to data science. It starts with developing an efficient system to capture data. Data needs to be computed by algorithms and read by software. Now, the information that you get can vary – it can be statistical, called descriptive analytics. You can also do something called predictive analytics, by looking at variations in the data and predicting a pattern. Another one is called diagnostic analytics, where you immediately and automatically identify the problem points based on data collected. Prescriptive analytics is another level which works on identifying the prescription for the problem that is already detected.
Considering that Data Sciences is a relatively new concept, what is your take on the educational infrastructure in India for courses in data science?
There is a lot of hype about this advent in technology in India. We need to be very clear about what is it that we are teaching, the syllabus and current industry trends needs to be well balanced. To understand current technological advances involves a lot of mathematical and statistical background, as it is not just another programming language. Some of the institutions, do take this into consideration. But we do see a lot of courses and training programs provided, but I suppose there is no benchmark of excellence that these colleges or courses can follow. The ministry should maybe try and formulate specifications that these courses need to comply, specially when it comes to advanced computer learning skills, so as to fill the skill gap that is there in the industry.
Having being part of the IT industry for so long, could you provide our female readers with a set of dos and don’ts, specially for those starting out now.
I think one of the biggest things is, lack of confidence. It is a major factor holding women back from venturing into the IT field, specially the advanced technological space. They may perform well academically, but then lose out on confidence when it comes to excelling in the industry. Women really need to do to what it takes to make it happen. Sincerely work hard, figure out your strengths and work on making yourself indispensable. Of course, everything depends on what she is passionate about, and giving it her all. Don’t try to be like your peers, find your strength in being the odd woman out in the industry. And try to overcome the internal discomfort that one may face in the team, that is a major thing that a woman needs to achieve.
One in every 12 women have the risk of a breast abnormality. With such alarming statistics, on a concluding note, what advice do you have to give to our women readers?
First of all the awareness needs to increase. Breast cancer is a non-communicable disease, but because of the lifestyle changes, dietary habits, lack of exercise, the type of family planning women do all of this results in radical changes in the way the metabolism works inside the body. And that is why we say prevention is better than cure. The earlier you detect, it is a simple out-patient department procedure. Women should also not have the notion that it won’t or can’t happen to them, I am not saying it will, I am just saying take measures to prevent it, and don’t ignore your breast health. The key is early detection, post which the treatment can be administered immediately and once cured, the woman can live a normal life. The havoc is created when it goes undetected for a long time, and then it becomes too late to do something.
This article was first published on April 25, 2018.