Stuti Khandwala

Stuti  Khandwala
MIT EECS | Philips Undergraduate Research and Innovation Scholar
Advisor: Regina A. Barzilay
Department: EECS
Areas of Research: Medical Devices
Years: 2021-2022
Research Project Title:

Future Risk Assessment of Smoking-Related Cancers in Smokers

abstract:It has been documented that, given how common smoking and the use of tobacco is in the United States and in the world, more than 40% of all cancers diagnosed in the US are linked to these factors. More than 34 million smokers in the US alone, and over a 16 million adults have a smoking-related disease1, many of them being cancers, not just lung, bronchus, and trachea but also that of the larynx, oropharynx, esophagus, bladder, kidney, pancreas, colorectum, cervix and liver. But many cancer deaths can be prevented by early diagnosis and timely preventive care of the patients. My project involves reviewing single low-dose chest computed tomography (LDCT) scan images to predict the future smoking-related cancer risk using an extension of the deep learning algorithm Sybil, which was developed using a subset of National Lung Screening Trial data which has radiologists' annotations for lung cancer diagnosis. Finally, I would see if other metadata associated with the datasets can be used to predict the grade of cancer someone has. Ultimately, such a model will aid doctors in better treating their patients as it reduces the risk of missed interval cancers and reduces the number of follow-up visits while making the treatment plan cater well to individual needs.

Participating in superUROP gives me the ability to have long term ownership on a project, something that I cannot have normally in class or via a UROP, something that I yearn to have and am fortunate to have in my undergrad. I have had numerous research experiences in computational and biological labs at MIT. I get excited by computational techniques applied to biology, and this project allows me to do that in a favourite direction, neuroscience.