Picnic’s mission is to structure the world’s medical data to make it useful. We work directly with patients to collect, digitize, and manage their complete medical records, giving them with control over their care through a personal health timeline
. We also partner with biotech, genomics, and pharma companies who sponsor PicnicHealth accounts for research volunteers. Through this work we’re building the data sets that power some of today’s most cutting edge medical research.
Clinical Data at Picnic
Machine learning is at the heart of our work. While the healthcare industry is waiting for standards and APIs to take hold, we’re able to create meaningful medical data today through our process for extracting and structuring medical information from records in whatever format they exist in the real world. We ingest medical records (faxes, scans, etc of printouts from EHRs, electronic data from EHRs) and convert them into structured, labeled medical data. Think of what we do as a multi-stage, human-in-the-loop machine learning system. We (1) run OCR to transform the image into text, (2) identify and label regions of the record, and (3) classify medical concepts with codes from standard ontologies (SNOMED CT, LOINC, RxNorm, ICD, etc) and structure relevant attributes. Ultimately, structuring data is just the starting point. We’ll use machine learning to pull insights from medical data and surface the most relevant information to make better decisions around care.
We’ve sourced and trained a scalable team of nurses in the Philippines to identify and label data and have already labeled records from more than 15,000 different doctors. We’re going through a crazy period of 20x growth this year, coming off our recent Grand Prize finish in Google Cloud’s Machine Learning Startup Competition
As the first biomedical informatician on our team, you’ll take ownership over the data we’re producing. You’ll help us ensure that we’re pulling the right information out of records, that we’re doing it accurately, and that we’re making that data intuitive to analyze. Today Picnic is the best way create a complete, longitudinal clinical dataset across any set of people. As we grow, we’re making Picnic the best way for people to analyze and learn from any health data. We want you to help shape that future.
- Refine Picnic’s clinical data model (based on OMOP CDM)
- Develop, train, and deploy machine learning models to improve PicnicHealth’s product and processing efficiency
- Build data pipelines
- Help shape design and development of research-facing products (interfaces, tools, and data products) to make them as useful for researchers as possible
- Work with product, clinical, operations, and sales teams to help define use cases, and develop methodology and benchmarks to evaluate different machine learning approaches
A candidate must have:
- A MS or PhD in biomedical informatics or a related field, or equivalent experience
- Published papers in biomedical informatics or a related field
- Performed research on EHR data or real-world evidence
- Shipped code for 2+ years
- Experience with machine learning
An ideal candidate has:
- Experience with clinical named-entity recognition or natural language processing in medical records
- Experience with Python, and Python machine learning tooling
- Exposure to Deep Learning, Gradient Boosting Machines, and other supervised classification methods
- Desire to kick ass and save lives