Polypilot product mascot

Introducing PolyPilot:

Our AI-Powered Mentorship Program

Start your trial today

Learn More
Go to Polygence Scholars page
Krishnaveni Parvataneni's cover illustration
Polygence Scholar2022
Krishnaveni Parvataneni's profile

Krishnaveni Parvataneni

BASIS Independent Silicon ValleyClass of 2024Santa Clara, California



  • "How could AI assist in the prediction and search for risk factors of anorexia nervosa?" with mentor Shaan (Sept. 27, 2022)

Project Portfolio

How could AI assist in the prediction and search for risk factors of anorexia nervosa?

Started Feb. 1, 2022

Portfolio item's cover image

Abstract or project description

Eating Disorders include such disorders as Anorexia Nervosa, Bulimia Nervosa, and Binge Eating Disorder. Patients with anorexia nervosa suppress their hunger, exercise too much, have low Body Mass Indices, and are dangerously thin. Bulimia Nervosa is characterized by binge-purge cycles, in which a patient eats more than is medically recommended in a very short period of time, only to remove the extra calories through a purging process, like throwing up or excessively exercising. Binge eating disorder includes unusual eating habits in which patients eat large amounts of food in a short period of time. However, patients do not purge these extra calories, causing patients with binge eating disorder to grow obese quickly and have high Body Mass Indices. Patients with binge eating disorder have comorbidities like Gastroesophageal Reflux Disease (GERD), where stomach acid shoots up burning the esophagus and sometimes the mouth.

My research focuses on creating an integrated Artificial Intelligence solution to finding eating disorder (such as anorexia nervosa, bulimia nervosa, or binge eating disorder) prevalence based on the responses to a questionnaire developed from the National Longitudinal Survey of Adolescent to Adult Health (ADD Health) database. Analysis centered around question H3GH8, "Have you ever been told by a doctor that you have an eating disorder, such as anorexia nervosa or bulimia?". By using Recursive Feature Elimination, I was able to decrease the number of variables used by the model from the original number in thousands to a number just below 100. My research has been sent to the International Youth Neuro Association, and will be published in their November 2022 issue.