Spotlighting Our Engineering Team: The Experts Accelerating Our Data (Rohan Pathak)

Author:

Verana Health

We’re shining a spotlight on a few of our Engineering team members at Verana Health. This team of more than 40 bright minds is responsible for the development of sophisticated pipelines and platforms that allow us to ingest and curate a high volume of de-identified healthcare data that can help accelerate research and improve patients’ lives.

In this blog series, we spotlight Rohan Pathak.

Q: How would you explain your role?

A: As a software engineer, my role involves taking machine learning (ML) models that have been developed from unstructured clinical notes in electronic health records and integrating them into a production environment, allowing predictions to be made on a large scale. This process also involves automating how often these models are run and ensuring they are in sync with the latest data to derive the most accurate insights. In addition, I assist our clinical abstraction team by supporting the workflow used to label datasets, which contributes to the creation of accurate training data for our ML models. More recently, I’ve been involved in data engineering efforts that facilitate the calculations of quality measures for MIPS (Merit-based Incentive Payment System).

Q: What attracted you to the field of Engineering?

 I became interested in data science and ML when I first started reading about its use across a variety of domains, ranging from helping basketball teams strategize in the NBA to assisting medical professionals in finding patterns within medical images

A: I became interested in data science and ML when I first started reading about its use across a variety of domains, ranging from helping basketball teams strategize in the NBA to assisting medical professionals in finding patterns within medical images. More recently, what attracted me to the engineering field was the desire to understand the bigger picture of the predictive model lifecycle, starting with data ingestion and ending with the deployment of models in a production environment.

Q: What attracted you to Verana Health?

A: The mission of advancing medical innovations and elevating the standard of care for patients by leveraging healthcare data is what attracted me to Verana Health. Additionally, the prospect of working with talented engineers and collaborating with data scientists and clinicians offered an appealing opportunity. 

Q: What inspires you to do this work?  

A: I got my start in data science in the healthcare technology industry. Since then, I’ve been inspired by the goal of having an impact on patient care. From trying to reduce readmission rates in hospital systems to implementing algorithms that extract information from medical text data, the challenge of improving outcomes while adapting to emerging technologies is highly motivating. 

Q: What is the most rewarding part of your job?

A: The most rewarding part of my job is when I’m able to work with others to automate a process that used to be done manually, such as the end-to-end processes of running an ML model on new data, as well as creating refined datasets from unstructured notes labeled by clinicians. This saves time and effort that can go toward other initiatives. 

Q: What advancements do you see in your field in the next 5 years?

A: Over the last year or so, the development and utilization of natural language processing and large language models has significantly shaped clinical analytics, which has the potential to substantially transform healthcare technology capabilities over the next five years. Automation is also likely to be a major focal point for advancements, as there’s an increasing need to minimize the number of repetitive manual tasks.

In his free time, Rohan enjoys traveling (Pictured in New York City with the Brooklyn Bridge in the background).

Verana Health Logo

Let's Accelerate Research Together

To learn more about Verana Health, please fill out the information below and our team will follow up with you as soon as possible.