Today’s young people are changing the way we think about what they’re capable of. They’re catalysts for nationwide demonstrations. They bring home Olympic medals. They run successful businesses, publish books, give TED Talks and even earn the Nobel Peace Prize.

And then there’s Kavya Kopparapu.

This senior at Thomas Jefferson High School for Science and Technology in Herndon, Virginia, invented a life-saving device. It began when she fell in love with the power of computing, building KNX projects with her brother and reading Scientific American with her cereal. A National Center for Women and Information Technology workshop sent her home in a fever pitch to learn programming, and in no time, she taught herself Java, HTML, Python, C.

Her fire led her to establish the Girls Computing League to hold coding workshops for underprivileged kids. Her brother and fellow high schoolers hold titles ranging from chief technology officer to chief information officer and chief strategy officer.

Kopparapu, of course, is the founder and CEO while competing in and placing in international science fairs for everything from cybersecurity to artificial intelligence.

And then she turned it up yet another notch.

In June 2016, inspired by her grandfather’s health condition in India, she asked an important question: What if there were a way for local doctors to diagnose diabetic retinopathy? She knew the statistics weren’t encouraging: Of 415 million diabetics worldwide, one-third will develop retinopathy. Fifty percent will be undiagnosed. Half of patients with severe cases will go blind in five years.

And she was also aware that, while doctors certainly make visits in India’s villages and slums, the number of patients far outnumber available ophthalmologists.

So she began her research online and emailed her specific questions to ophthalmologists, computational pathologists, biochemists, epidemiologists, neuroscientists, physicists and gurus in the machine learning field. Her next step: founding another company. Eyeagnosis, which uses a 3D printer to create a camera lens that slides over a smartphone, and, through the corresponding app, diagnoses photos of an eyeball.

She “simply” employed Microsoft’s convolutional neural network off-the-shelf program, ResNet-50, and trained it to spot retinopathy, using 34,000 scans available through the National Institutes of Health.
By November 2017, not quite six months after her initial curiosity, she mailed her prototype to Aditya Jyot Eye Hospital in Mumbai for field testing.

To date, the Eyeagnosis has accurately diagnosed all five of the patients doctors have used it on.  
 But Kopparapu isn’t resting on her laurels. In May, she keynoted at the O’Reilly Artificial Intelligence Conference in New York on her latest invention, GlioVision, a project that earned her finalist status in the Regeneron Science Talent Search – known as the junior Nobel Prize.

 GlioVision uses GPU-accelerated (graphics processing unit) deep learning to fight glioblastoma, the deadliest form of brain cancer. The software instantly detects and interprets genetic information from a biopsy slide, skipping the long analysis typically involved in determining glioblastoma treatment.

 The hope is that doctors will then be able to use this information to predict how fast a tumor will grow and what specific drugs and treatments it will respond to. Today, she’s working with a pathologist at the Georgetown University Medical Center to test her invention’s performance on patient data.

Kopparapu says her goal is to be part of a world where she is not known as a girl who happens to be a computer scientist, but instead as a computer scientist who happens to be a girl. Part two of that goal – making the world a better place. That work will continue from Harvard in the fall. 

Was there a teacher or mentor who influenced you or spurred your entrepreneurial thinking?

Over the years, I’ve had the fortune of having several amazing teachers: ones who would stay after school to help me finish a science project, ones who would give up their lunchtime to answer the burning question I wasn’t able to ask during class, ones who would accept invitations to attend events on weekends on their own time.

But I think the teacher who has significantly influenced my entrepreneurial thinking is Mark Hannum, my AP physics teacher and neuroscience lab director. Mr. Hannum teaches at the intersection of many fields: physics, neuroscience, biology and mathematics. He always promoted an interdisciplinary mindset, which led me to the integrative thinking that led to projects such as Eyeagnosis and GlioVision. He also emphasized a mindset of learning, that any subject was within my reach if I had the determination to learn it.

What was your experience with computer science and coding curriculum at school?

At Thomas Jefferson High School, a STEM magnet, we have access to a plethora of computer science classes, from AP computer science to mobile app and web development to artificial intelligence to parallel computing.
Many of my computer science teachers had Ph.D.s in computational science and still actively publish in the fields of computer science and applied mathematics. Whenever attempted a research project, I had the support of teachers with vast experience in the field.

What are adults missing when it comes to getting girls and underrepresented students involved in computer science and STEM? What fueled your interest?

I think the key to solving the problem is creating a community of support. The most discouraging factor to girls and minorities in computer science is not a lack of passion or interest in the subject, but rather the lack of others in the field who are similar in background. That’s one of the primary aims of Girls Computing League: to foster an inclusive environment of support that students may rely on as they progress through their computer science studies.

My interest in computer science was mainly based on the fact I could take technology and combine it with any other interest that I had to make a difference. My major interest has always been biology and medicine, and integrating this interest with computer science has greatly expanded the scope of my work.

Girls Computing League workshops actually follow this philosophy: that the best way of teaching computer science is relating it to a previous personal interest. Our website development and mobile app development workshops are intentionally open-ended, allowing students to make websites showcasing their individual interests.

How can students who don’t have access to a lot of resources and teacher support still learn computer science?

If you have access to a computer (either a personal one, at school or at a public library), then you have access to countless resources for computer science.

What are some good ways for a student or even a teacher who doesn’t have a lot of comfort with computer science to get started?  

Last year, Girls Computing League held teacher professional development workshops in collaboration with the Tiger Woods Foundation. It was especially rewarding because most of the teachers who attended were not actually STEM teachers – they were English teachers, history teachers, music teachers and art teachers who wanted to find a venue to incorporate computer science into their curriculum.

The biggest advice I have for anyone looking into learning computer science is to just dive into it. There are many resources online; Girls Computing League has a resource center and there are Youtube videos, Coursera courses and tutorial websites.

A lot of people believe that computer science is too complicated to self-learn, but it’s actually the easiest skill to pick up because of the plethora of information available online.   

How can we as a society do a better job of tapping into the brilliance of young people?

We need to do a better job of giving students access to resources. My personal philosophy is this: since only about 40 percent of public schools offer computer science, we’re potentially missing out on 60 percent of future innovation.

As a middle schooler, I could have never imagined the world of opportunity available in computer science had I not been introduced to it in eighth grade.

Which of the ISTE Standards for Students best describes you? Why?

I think Global Collaborator best describes me and my view on the world.

At the scientific level, I’m at the intersection of two fields – medicine and computer science. I believe this interdisciplinary thinking will advance all of science and technology because it’s the area we are only just now starting to look at.

What’s next for you? And what real-world problem do you hope to solve next?

As I continue to work on Eyeagnosis and GlioVision, I aim to start a medical informatics company. Under the umbrella of a company, I will be able to produce and sell the medical innovations I have been working on, allowing more patients to have access to medical innovation.

After all, what is the purpose of research if not to help humanity?

Julie Phillips Randles is a freelance writer and editor with 30 years of experience writing about education policy, leadership, curriculum and edtech.