The Human anatomy is a miracle of nature , and to be able to study its functions through Medical Imaging techniques is a miracle of science !

Nikhil Narayan, our next pathbreaker, Research Scientist at Sony Research India, works on strategic initiatives to help build imaging technologies that will make healthcare more reachable and affordable to the society.

Nikhil talks to Shyam Krishnamurthy from The Interview Portal about his defining moment during his internship that led him to a PhD in Medical Imaging at Nanyang Technological University, Singapore and his vision to solve real problems in Healthcare using imaging technologies.

For students, it is remarkable to see Medical Imaging technologies such as CT Scan and Ultrasound improving the lives of millions of people everyday. If you want to work at the intersection of engineering and healthcare to create disruptive technologies for the real world, this is the career for you!

Nikhil, tell us about Your background?

I was born in Bangalore into a family of Doctors. My Grandfather was a General Practitioner, my Father is a Paediatrician while my mother is a homemaker. I did my schooling and undergraduate studies in Bangalore. The house where I grew up was right under the glide slope of HAL airport (now closed) and I spent my childhood days constantly gazing at the skies watching planes land and take off. I was hooked to aviation from then on. With this I picked up the hobby of collecting scaled plane models. This is an expensive hobby and to buy new plane models, I had made a deal with my parents that every time my academic score went beyond a certain threshold, I would get a new plane model as a reward. This interest in aviation kindled my desire to learn more about the mechanics of operation and I was drawn towards maths and science, physics in particular, during my school days. I still remember that one week when I was in class 9 where I attempted to recreate a tiny turbo-jet with whatever little scrap material I found in my house. To cut a long story short, it was in flames before it could even start. But that failure ignited this fire in me to learn more technically. 

What did you study?

I secured a very good Karnataka-CET rank and managed to get a seat in Electronics and Communications engineering at P.E.S. Institute of Technology (now PES University) in Bangalore. For my Post Graduation, I enrolled for a Masters in Science degree with specialisation in Signal Processing at Nanyang Technological University, Singapore. A couple of years after obtaining my Master’s degree, I obtained my PhD degree in Medical Imaging from Nanyang Technological University, Singapore.

What made you choose such an offbeat, unconventional and uncommon career?

By the time I entered 10th standard, I had made up my mind to pursue engineering. But having come from a background of Doctors, my parents did not know which field of engineering would help me get to where I wanted to. So my mother started asking relatives who were engineers about the best engineering degree. I guess she forgot to mention aviation to them, as the most common answer she got was Electronics and Communication Engineering. Given that the internet was still in its infancy back then and trust in relatives was higher than Googling for information (I think google wasn’t even there back then. I remember the answer to the quiz question: “which is the largest search engine in the world?” was , I believed that a degree in Electronics would take me to the aviation world. In retrospect, even though I am not in aviation, I am glad that I made the decision to pick up Electronics as my major for undergraduate studies.

Tell us about your career path

As mentioned previously, I had adopted a goal oriented approach towards academics right from a very young age (goal being, to beat my previous best to get a new plane model as reward). When I made up my mind to pursue a degree in Electronics and Communication with the help of my parents and friends, I managed to gather information regarding the best college in Bangalore and the cut off CET rank for the same. With the goal set, I was laser focused on obtaining a rank higher than the cut-off. I had a diary where I used to clock the number of hours and topics I had studied everyday. Weekends were reserved to test the knowledge gained during the week. A disciplined approach to studies ensured that I got the exact cut-off score required to make it to PESIT.

During my undergraduate studies, a friend of mine invited me to take part in this science competition organised by Intel called the “Intel Scholar Program”. My friend and I decided to come up with a new Image Processing algorithm for the competition. The idea for the competition stemmed from a course that we had chosen as one of the electives in our 6th semester which was on Digital Signal Processing architectures. I distinctly remember asking the professor if there was a way to compress an image beyond the compression that we get by using JPEG compression method. The professor asked us to look up the literature on Image Processing to see if that is indeed possible. We took it up as a challenge and started researching on different ways to compress an image and found out that it is possible to compress an image to 1/4th of its original size but restoration of the image back to its original size proved to be a challenge as the inverse to the quadratic problem had many solutions. We proposed to solve this by directly interpolating the adjacent values instead of solving the inverse problem and managed to achieve good reconstruction accuracy. Three rounds of competition later, our team was in top twelve in India and we got a chance to present our work to top researchers in academia and research. 

This was the beginning of my journey in research on Computer Vision algorithms. I wanted to explore more in Image Processing, so I enrolled in the Masters course in Signal Processing at NTU Singapore with specialisation in Image Processing. The problem statement that I selected (or rather that was given by my professor Anamitra Makur) for my Masters thesis was in the field of Image Steganography. Image Steganography is the technique of hiding data or inserting invisible watermarks within an image. The problem statement that I had chosen was to restore a tampered image back to its original form by recovering the information about the image that is hidden as a watermark. In order to do this, the original image is scrambled, compressed and inserted back as an invisible watermark. So when one part of the image is tampered, the information about the tampered part before the attack is stored elsewhere in the image as a watermark. Thus, by retrieving this information, the original image is reconstructed. The biggest challenge was the restoration part, as for faithful reconstruction of a compressed image, the compression factor has to be as low as possible. But at low compression factor, the size of the image becomes too big and moreover, the watermark is no longer invisible and appears as noise in the image. I struggled to solve this problem for 8 months. There was only 3-4 weeks left for the final submission and I still hadn’t found a way to solve this problem. Then one day ( the date was 25th May 2010 ) I had an epiphany when I was traveling on the metro (MRT in Singapore). The journey typically takes about an hour from the place where I lived to the University and I had dozed off half way into the journey.  I dreamt of that class in my Undergraduate course on communications theory where we were introduced to channel noise and recovery of signals. I woke up with a jolt and an idea to use the knowledge from communication theory to Image Processing. I hurried to my lab and re-wrote the code to include an encoder and decoder system to add redundancies while maintaining the compression at high levels. The result was astonishing, I could achieve a Peak Signal to Noise (PSNR) of above 27dB which is an indicator of a good quality restoration of the tampered image.  The research work I carried out for my masters thesis was published in a top conference and thus boosted my confidence to pick up a career in Research and Development.

As soon as I completed my Masters degree, I managed to secure an internship at Siemens Healthcare in Bangalore. I applied for the position by asking a relative of mine to forward my resume to the team. I had two rounds of interviews for the internship, one of which was technical and another was about my project work. This internship introduced me to the fascinating field of Medical Imaging. I was so pumped-up that within three months of joining as an intern, I developed a fully functional algorithm for CT scans and filed two invention disclosures.

My short exposure to Medical Imaging at Siemens motivated me to seriously consider a career in the Healthcare domain. Thus, I started applying for a PhD in Medical Imaging and as luck would have it, a Professor at NTU had recently secured a grant for a Medical Imaging project and contacted me. I was thrilled and applied for the position which was offered to me after two rounds of interview by the lab which was funding the project. I got a scholarship for my PhD which provided me with a monthly stipend for 4 years.

My PhD was in Ultrasound Imaging and this led to my next stint as a researcher at Samsung Research Institute India – Bangalore (SRI-B) in the Ultrasound division. The algorithms that I developed at SRI-B are now a part of some of the premium Ultrasound machines that Gynaecologists use globally.

What were the challenges you faced during your career? How did you address them?

The journey from undergraduate studies to where I am today was not an easy one, it was riddled with challenges at every stage starting from when I graduated.

I graduated in 2009 and when I graduated, recession had hit global economies, and companies that had offered jobs during placements withdrew the offers. I had two options back then, pray and search for jobs or pursue higher studies. I picked the latter as this would ensure that I had a better degree when the world gets out of recession. 

Then when I got into my Masters program, I was given a research problem statement that was left unfinished by a PhD student as the person felt that it could not be improved anymore. I took up the challenge and until the very end I was in the same state as the PhD student, but a constant review of state-of-the-art literature related to the problem statement gave me new ideas, which upon implementation, not only improved the performance but also helped me get a publication in one of the top international conferences in the world.

The two challenges above are some of the many that I have faced in the last decade or so in my career.

How did you get your first break (after PhD)?

In January of 2016, I had come back to Bangalore after submitting my PhD thesis and a couple of weeks after my arrival, I met an ex-colleague (who is a friend and a mentor to me) of mine for a cup of coffee. He told me that the Ultrasound team at Samsung R&D Institute in Bangalore was on the lookout for a researcher and told me that I would be a good fit for the same. So I forwarded my CV to him asking to refer me to the team. A couple of weeks later after 3-4 rounds of interview, I landed the job at SRI-B as a Tech Lead in the Ultrasound team of Health and Medical Equipment division.

Tell us what you do currently?

I work at Sony Research India as an Assoc. Principal Research Scientist. My job involves setting the strategy for healthcare for the next 10 years at Sony. This requires me to be up to date with the recent trends in both technology and market needs. I need a lot of leadership and negotiation skills to be in the position that I hold now. The position also demands that I have good soft skills such as effective communication and presentation (technical and non technical). I am also incharge of increasing the patent and publication portfolio of the company. This means that I must always be thinking out of the box and come up with innovative solutions for real world problems. The thrill of coming up with strategic initiatives motivates me to perform above and beyond in my job. This is what I love about my job.

My work will help Sony to create technology that will make healthcare more reachable and affordable in India and at times such as these (with the pandemic and all), it will greatly benefit the society at large.

What was your most memorable work so far?

My most memorable work is my PhD research work where I challenged the status quo and came up with algorithms that were way ahead of their time.

What is the benefit of your work to society?

I have been involved in developing algorithms for Healthcare for about 10+years now and all of my solutions have made it to products that are helping people in various ways.

My PhD is in Ultrasound imaging and one of the major uses of Ultrasound imaging is in Obstetrics and Gynaecology field of medicine. The female reproductive system consists of a pair of Ovaries that has many follicles in it. Each follicle houses an egg, which when released during the menstrual cycle, and if a sperm is present, will fertilise to form a zygote which in turn will develop into a foetus. When the hormone levels change or when there is a disorder called as a Poly-Cystic Ovarian Syndrome/Disorder (PCOS/PCOD), the follicles fail to grow and release the egg thus leading to infertility. The solution for infertility is a procedure called In-Vitro Fertilization, where the hormones are injected so that the follicles grow in size and when they reach a desired size, the gynaecologist will insert a needle into the follicle, extract the egg and fertilise it in a petridish to form a zygote. The zygote is inserted into the Uterus for further development into foetus.

The entire procedure of IVF is carried out with the help of Ultrasound Imaging where the ultrasound images are used to continuously monitor the number and growth of the follicles. The images are also used as guides to insert the needle for egg extraction and insertion of zygote into the Uterus. I developed the algorithms that can automatically count the number of follicles and measure the size of the follicles in real time when the Ovary is being scanned. All of my algorithms are integrated with premium ultrasound machines of Samsung and are used multiple times everyday to treat infertility in women.

Your advice to students?

  • Never give up!
  • Marks and scores aren’t everything. Common sense takes you a long way.
  • It is important to be both street smart and book smart. Use your smartness wisely.
  • You get what you want provided you work towards it. There is no free food in this world.
  • Be bold and do not shy away from challenges. What doesn’t kill you only makes you stronger.

Future plans?

My future plan is to grow up the career ladder and eventually become a CTO of a reputed global company.

What happened to aviation you ask? I am still pursuing it as a hobby 🙂