Shaan K
- Research Program Mentor
Doctor of Medicine (MD) candidate at University of Connecticut
Expertise
Machine Learning, Medicine, Web Design or Podcast Creation, Large Dataset Analysis/Data Science, Mental Health, Psychiatry
Bio
Hi! I am a 4th year medical student at the University of Connecticut. Prior to medical school, I did a 4 year combined Bachelor of Science/Master's of Science dual degree in Molecular & Cell Biology and Structural Biology, Biochemistry, and Biophysics respectively. During that time, I did molecular dynamics simulation research for my thesis. As a medical student, I've interned at the investment firm Canaan, created websites for medical education (such as medschoolanki.com), produced and launched podcasts (such as Admissions Dissected), and engaged in a lot of machine learning and data science projects, with a focus on research within psychiatry and neuroscience. Between my 3rd and 4th years of medical school, I worked in the BrainLENS Lab at UCSF/UConn using machine learning and fMRI scans to predict if children had dyslexia or not. I'm passionate about finding creative solutions to problems, whether that's through research and scholarly work or through dissemination of information through engaging platforms (such as podcasting or creating websites). I love mentoring and helping students find an avenue to pursue what they are passionate in. I spend my downtime watching sports (basketball, tennis, and football), cooking, and watching movies.Project ideas
Machine Learning in Your Area of Interest
In this project, we would identify an area of interest that you have and a dataset of interest to you. We would then define any outcomes you'd be interested in predicting and then create machine learning models to see how well we can predict those outcomes. As an example, if you were interested in predicting cancer risk, we could use the SEER (Surveillance, Epidemiology, and End Results Program) dataset, identify our specific outcome of interest, and then create models to predict that outcome.
Creating a Website or Podcast
If the student has an area of interest that does not have much information about it on the internet, we could explore how best to present that information on either a website or podcast. Once we decide which avenue to use, we would go through all of the steps necessary to create a website or podcast all the way through launch of the website or podcast.
Sports Analytics
Using publicly available data, a student would choose their sport and area of interest. We would then use data science tools to analyze that area of interest and potentially make predictions. As an example, if the student is interested in predicting which NBA players may become All Stars, we would use the Basketball Reference dataset and tools such as machine learning to predict which players may be selected to become All Stars.