One Day at Bootcamp [All Cities] - Next Two Weekends! RSVP

Demystifying Data Science: Improving Delivery of Patient Care with Analytics

By Emily Wilson • July 05, 2016

Bootcamp graduate Alison Cossette came to Metis with 10 years of healthcare experience and a desire to integrate that with rigorous data science training. At the time of this Q&A, she was a Senior Analyst in Center for Healthcare Management at the University of Vermont Medical Center, working on data projects that directly impact patients' healthcare. She has since moved on and is now a Senior Analyst at Remedy Partners, a company delivering technology and services that enable payers and providers to organize and finance a patient's episode of care. 

Tell me about your background. How did you become interested in data science?

My early career was actually in technical project management. Then after 9/11, I decided I needed to make the world a better place and spent the next 12 years in lymphedema therapy. Lymphedema is a swelling of the limbs most often secondary to cancer treatment. It is a progressive and incurable disease. I spent a lot of my time working with breast cancer patients. After taking a break from working to have my children, I decided that I really wanted to use the more analytical part of my brain, so I started taking CS classes including C++. While I was in school my best friend from college was diagnosed with breast cancer and subsequently lymphedema. It was at this time that I realized I really wanted to be involved in personalized medicine. 80% of breast cancer patients don't develop lymphedema but 100% of them have to take precautions. They are cancer free but not free of the cancer. I wanted to be able to utilize data from wearables to give that 80% their life back. Describe your current role. What do you like about it? What are some challenges?  

I am a Senior Analyst in Center for Healthcare Management at the University of Vermont Medical Center. Currently, I work with the healthcare innovation lab initiative (hiCOlab). We seek to use advanced analytics to improve delivery of patient care. The most rewarding aspect is knowing that the projects I am working on have a direct impact on the health of people in our region. There are two primary challenges in our setting. The first is one of highly-siloed data. As in many hospital settings, we have a number of independent software packages that are utilized, like billing, admitting, electronic health records (EHR), and registration. We do have an enterprise data warehouse, but record linkage is extremely challenging. Data practices can vary greatly depending on the specialty or the needs of a given practice/hospital. The second challenge is in data ownership. While the data we need does exist, getting various departments to give insight into where in the 15,000-data-field it lives, and how it is utilized, can be a stalling point. In your current role, what aspects of data science are you using regularly? 

I work on a wide variety of projects, everything from basic stats analysis & data visualization of survey data to natural language processing (NLP) on survey responses. My largest project right now is in workforce planning. I am charged with determining how many physicians, of which specialties, and in which locations, we need throughout the network. The short-term model is more reporting/projection-based. On a long-term basis, I am working on a deep-learning model based on electronic health record data that will predict diagnoses on an individual patient level. This will, in turn, be the basis for various intervention, planning, and research projects. Do you think the projects you did at Metis had a direct impact on your finding a job after graduation? 

100% yes! When I started with Metis, I had a basic understanding of programming from C++ but no practical Python experience. By the end of Metis, I was a data scientist with 10 years of healthcare experience. I certainly would never have gotten this position without it. The variety of projects we completed was also helpful. As I said, in my current position, it's NLP one day, classification the next, and deep-learning the day after that. With my Metis training, I can smoothly shift from one to the next with ease. What would you say to a current Metis applicant? What should they be prepared for? What can they expect from the bootcamp and the overall experience? I can't tell you what Metis will be for you, but I can tell you what it was for me. Metis is one of the most challenging things I have ever done. During my first project, I said this: "An Ironman triathlon is so much easier than Metis. In Ironman, after 15 hours, you stop - good or bad. You can go 15 hours at Metis and still have something that doesn't work...but you just have to keep going!" My point is that Metis pushed me to find things in myself I didn't know I had. This is an opportunity to be your best self. This is an opportunity to gain experience that will allow you to take on really big problems in life. Be prepared to: work hard, not sleep, fail, make friends, succeed, celebrate, fail again, question, succeed again, and triumph. To get the most out of it, I have two pieces of advice. 1) don't be afraid of messy data (a dictionary is your best friend) and 2) don't let yourself struggle. Ask questions when you need. What's great about Metis is that there is a whole team of people, staff and students alike, who are committed to your success. There are 35 people on your side. When was the last time you could say that? Ultimately, I get to make the world a better place every day because I went to Metis. What more is there to say?

Similar Posts

alumni Eikbkecrb2s9risciv3o
Made at Metis: Waste Analysis + Building Data Science Solutions

By Emily Wilson • August 01, 2018

Take a look around and you might notice some waste. It could come in the form of something physical that you can see or feel or smell. Or it might be invisible to the eye, like wasted time or resources due to lapses in efficiency. Two recent Metis graduates took an interest in these different forms of waste and used data science to come up with ways to minimize negative impact and maximize positive solutions.

alumni 7gu9puqpq9ex1etqcpw8
Made at Metis: Predicting a Pitch in Baseball, Predicting Mode in a Song

By Emily Wilson • October 18, 2018

Read about two recent final projects created by graduates of our data science bootcamp . See what's possible to create in just 12 weeks, including a way to predict which pitches the Cubs should throw to the Cardinals, and how to predict the mode of a hit single on the Billboard charts.

alumni Gs1beeedr5gsdd3flfob
A Look at What You Can Create During a 12-Week Bootcamp

By Emily Wilson • May 14, 2018

As you enter the bootcamp, your computer screen is like a blank canvas, waiting to be covered with the messy excitement of ideas and projects, failures and successes. Projects are at the core of our Data Science Bootcamp curriculum and the philosophy surrounding it. A robust portfolio, demonstrating a wide understanding of tools and theories, along with a solid grasp of needed industry skills, is the central piece of the overall "why our graduates get hired" puzzle. See some final project examples here.