Skip to content
ushealthcloud

Data Science

Healthcare data science on infrastructure built for PHI

We turn claims, EHR, lab and research data into models and analyses that clinical, actuarial and compliance teams can question, reproduce and sign off on.

The pipeline

One governed pipeline, four parts

Every engagement runs on the same backbone. Choose a part to see what it does to the data.

patient_record → de-identifiedExample · synthetic
Name
Alex Sampleremoved
Date of birth
1961-04-121961
ZIP
10027100**
MRN
MRN-DEMO-4471token 7f3a…
Diagnosis
E11.9E11.9

Safe Harbor identifiers handled before analysts see the data

Inputs

The data we work with

  • Claims (X12 837/835)
  • EHR clinical notes
  • Lab results
  • Imaging metadata
  • Pharmacy claims
  • Registry data
  • Remote monitoring
  • Prior-auth packets

How it works

How an engagement runs

  1. 01 Assess

    Map the question, the data sources and where PHI lives. Agree what must be de-identified and who may see what.

  2. 02 Build

    Stand up governed pipelines and features, then develop models or analyses against a documented baseline.

  3. 03 Validate

    Test on held-out data and across patient subgroups. Clinical and compliance stakeholders review before use.

  4. 04 Operate

    Monitor drift and data quality, retrain on schedule and keep documentation current.

FAQ

Questions we hear from compliance and data teams

  • Data accessDo your analysts work with identifiable PHI?

    We design for minimum necessary access. Analysis runs on de-identified data wherever the question allows, and identifiable data stays restricted to named roles for the specific tasks that need it.

  • StandardsWhich healthcare data standards do you work with?

    Claims in X12 837/835, clinical data in HL7 v2 and FHIR, research data in the OMOP common data model, and DICOM metadata for imaging.

  • ValidationHow do you validate a model before it is used?

    Held-out testing, performance checks across patient subgroups to surface bias, calibration review, and a documented model card that clinical and compliance owners sign off on.

  • EngagementCan we start with a single question rather than a platform?

    Yes. Most engagements start with one decision a team needs to make, such as which members to prioritize for outreach, and grow from there.

Next step

Bring us the data problem your compliance team worries about most.

We'll map where protected health information flows today, and how to model, automate and secure it.

What the first call gives you

  1. 1A map of where PHI flows today
  2. 2One workflow worth starting with
  3. 3The controls that workflow needs

A short scoping call, then a written proposal. Please don't send patient information.