Penn Medicine has been working with researchers and clinicians on a solution to access and derive meaning from unstructured data, reducing time spent on abstraction to facilitate greater focus on research.

What Is It?

Watch a 3-minute video introducing PennSeek   (accessible while connected to the Penn Medicine network)

PennSeek is a tool to search unstructured or semi-structured medical documents currently residing in Penn Medicine’s EMRs and diagnostic applications (Radiology, Pathology, Cardiology, Ambulatory, etc.), to analyze and mine this data for identifying trends and input for patient care and research. Currently, PennSeek is being piloted by the Anatomical Pathology, Dermatology, and Rheumatology departments at the Hospital of the University of Pennsylvania (HUP), as well as several research areas.

Pennseek: A Closer Look   (accessible while connected to the Penn Medicine network)

Our Goal

To bring together data from all our major clinical applications, allowing immediate search and information discovery capability across those applications from one place.

What We Do

The PennSeek team works with clinicians and researchers to develop stories of workflows that could be solved by access to data. These stories are then transformed into prototypes that allow visualization of both the problem and solution. These prototypes are then further developed into applications that tap into clinical data sources to bring data together in a way that resolves each story into an end-to-end solution.


PennSeek servers sit inside the UPHS firewall, and are accessible only to individuals logged in to the UPHS network. Currently, login to PennSeek is secured via per-user assigned credentials, specific to the application. Each user is granted access to data they are authorized to see depending on their HIPAA level—4 & 5 permits access to PHI, all others are restricted from viewing patient data.

Supporting Research – How can PennSeek projects help you Now

PennSeek Preparatory Research
  • Provides a count of potential patients that may be eligible to participate in a proposed study. Access does not require IRB approval.
  • Although it provides access to all patients with HUP MRNs defined, discrete fields with PHI data are redacted and export functionality is restricted.
PennSeek Recruitment
  • Designed to allow both refinement of search criteria and immediate review of identified cohort data. It is ideal for narrowing your cohort, requesting research study IDs, and accessing patient data across categories for research purposes.
  • PennSeek Recruitment provides access to all patients with HUP MRNs defined, and requires IRB approval.
˜ PennSeek Research
  • Intended for review of specific patient cohorts. It provides the ability to export any and all data related to predefined cohorts for additional analysis or pre-publication de-identification tasks, and requires IRB approval.
  • PennSeek research only allows access to patient cohorts defined by research study IDs.


To obtain access to PennSeek, please fill out the PennSeek Access form. The form is available for download here (must be on the UPHS network to download). In your submission, be sure to include the IRB protocol document listing:

  1. Your name as a member of the research staff
  2. The protocol number
  3. The most recent protocol start date

And please submit request via IS Service Portal to Data Analytics Services with form attached.

Once you have access, we're also happy to meet with you to demo the tool and answer any questions.

For additional information, please submit request via IS Service Portal to Data Analytics Services.