ChiPy Data SIG presents Python Powered Healthcare

ChiPy: Chicago's Official Python User Group
ChiPy: Chicago's Official Python User Group
Public group


222 W Merchandise Mart Plaza #1230 · Chicago, IL

How to find us

MATTER is located in the Merchandise Mart. Take the middle elevators to the 12th floor. Ask security for directions if you are lost

Location image of event venue


Chicago Python Data Special Interest Group is partnering with Redox to spotlight how Python is used in the healthcare industry.

Redox is hiring! Check out their open positions:


6:00 - Doors open
6:30 - Talks Start
8:30 - See you all next time!



Healthcare Security for developers
by Anil Saldanha, Cybersecurity Expert

Talk will focus on cybersecurity challenges for developers in healthcare.


Snakes and (Genomic) Ladders
by Stephen Bush, Tempus

A brief overview of how we use python for bioinformatics analyses at Tempus.


Migrating a machine learning pipeline to Kubernetes
by Zach Lipp, Lumere

Lumere uses Python for everything from serving our application to powering our data science and data engineering teams. In this talk, we'll discuss how Lumere uses machine learning models developed in Python to classify hospital purchases, how we've deployed those models, and what we've learned porting this deployment to Kubernetes.


Using Python to Identify Healthcare Practices: SMA Clinical Data Registry
by Adrian Javier Segura, Cure SMA

Python is used to help clean, manage, and analyze EHR data that is being sent to the Spinal Muscular Atrophy (SMA) Clinical Data Registry. The data collected for the SMA Clinical Data Registry is being analyzed to identify care practices that are important for improving SMA care. Python (and its various libraries/packages) are being leveraged to ingest extensive amounts of raw data from these EMRs and extract insights that can aid physicians in providing care for individuals with SMA and their families. Python allows Cure SMA to work with EHR data in a fast, accurate, reproducible, and cost-effective way.


MD vs Machine: AI in Medical Imaging
by Jay Rodge

Recent progress in AI and machine learning is altering the way doctors practice medicine. Can AI in medical imaging improve health care? This talk will discuss how AI assists doctors in Medical Imaging for better prediction. We'll also see a demo of building a Medical Imaging AI model using MedicalTorch in Python.


Health Insurance Data Science with Python
by Sonjia Waxmonsky and Zack Larsen, Health Care Service Corporation (HCSC)

HCSC receives approximately 40,000 calls per day to our call centers. We perform analytics using PySpark, Databricks, and open-source modeling packages to extract insights and model the user experience with this data source.

Our clinical data science team uses Python and statistical modeling packages for causal inference with our observational data to provide scientific estimates of what effect these programs have on our members’ health.


Interactions in the health care provider network
by Vishal Soni, Blue Cross Blue Shield of Illinois

The health care system is largely made up of patients, healthcare providers, and their interactions. These interactions imply a network, and this network is often a more natural way to represent the underlying data. I will talk about how we use python to construct and gain insights from this network, and how such analyses can be used to identify anomalous behaviors.


Prophet for seasonality in claims data
by Bei Ding, Validate Health

Prophet is a library created by Facebook and available both in Python and R to forecast time series data. Validate Health uses Prophet to analyze seasonality in healthcare claims and project financial performance for provider clients.


Practical Excel automation with pandas and xlwings
by Paul Zuradzki, Evolent Health

When building reports for Excel-users, we often receive data files that must be merged or the other way around (merged data that must be split). If you had to split or merge[masked] files, short scripts can save your team hours and allow reproducibility with less human error.

Today we’ll demo XLwings (built on OpenPyXL), which supports more intuitive handling of tabular data compared to most published Excel-Python tutorials.