Ritu Jha-
For a 16-year-old Intel Science winner convincing his parents has been like pitching the “Shark Tank” team because he says, he needs to win them over to fund his projects since most of the time they don’t work out.
“I always have wild ideas, and it’s like “Shark Tank,” but they are always supportive of me in either way. I have to ask for money and investment whether my ideas work on not, and I’m lucky to have them,” said Krithik Ramesh, a high school junior of Greenwood Village, Colorado, who won the $75,000 Gordon E. Moore Award, a top prize at the Intel International Science and Engineering Fair (ISEF), a program of Society for Science & the Public held this past month for developing a machine-learning technology for orthopedic surgeons.
“It is always overwhelming, and all are smart and have passion for science and so much fun to interact with them,” said Indian- American Ramesh on winning the world’s largest international pre-college science competition. The competition featured more than 1,800 young scientists selected from 423 affiliate fairs in more than 80 countries, regions and territories.
Ramesh, in an interview with indica, shared how his prior background in aviation and mechanical engineering helped him to take the challenge of working on the Intel competition and innovating an idea into biomedical engineering.
When asked on getting into biomedical engineering, he said, “Even people were surprised when I did this and that I understand.”
Explaining why biomedical he said that medicine and artificial intelligence are converging, and these are upcoming fields and are an upcoming opportunity as well.
His project was based on using machine learning and computer vision. The award-winning project helps orthopedic surgeons achieve greater accuracy for screw placement during spinal surgery.
Based on Ramesh’s tests, this method has the potential to decrease operating time by at least 30 minutes, reduce physical therapy recovery time by four weeks and diminish the negative side effects associated with traditional medical imaging.
He said the most interesting thing for him was that a lot of underlining engineering principles he learned from his previous work in aviation and apply pretty well on the mechanics of the spine and structural object and autonomous skill compliance system.
When asked what inspired him to work on spine-related research, Ramesh said that while he was reading a couple of medical journals in which the University of Minnesota Department of Orthopedic Surgery was talking about the placement of pedicle screws, the kind that goes into your spine during surgery, was only 76 percent accurate, which is relatively alarming on how precise the surgery should be.
Also, the idea came while playing a video game, where players dance along with the motion on the screen. “I was thinking, ‘What if I apply motion capture system, for the radiology of spine and ultimately improve the performance?” said Ramesh, who borrowed an augmented reality device and tried to apply the same to the spine, the camera used during surgery.
Ramesh said he started working on the project in 2018 and then finished it at the end of January this year.
But there were many failures before winning the big prize. Ramesh said he used to work at a couple of radiologist and doctors outside the school environment, and he was accepted after making a lot of cold calling. Often he worked on his project until 3 a.m.
“To get into research on the spine you have to do an extensive amount of research and did independently,” he said.
“There were a lot of failures in the beginning, and a lot of things I tried to do did not necessarily work and needed the right algorithm, too, a lot of time,” he said, adding that to reached the result he used surgical data from 34 patients and more than 2000 images and employed that data and analysis to access data about how viable algorithm was.
“And that data was validated by medical experts and that ultimately helped me how to do my project. There was a lot of viability,” said Ramesh, who said his project is the first of its kind, and this is novel and complicated research that needed many algorithms.
Ramesh, who loves experiment projects, said his next project is to study fetal cardiac abnormalities and fetal surgery in first and second trimesters using cardiogram ultrasound.
