De-Identification Framework

On-Ramp to AI and Machine Learning Innovation

de-identificationBy 2020, imaging studies in the U.S. alone will account for 2.4 exabytes of data (source: IDC), presenting a unique opportunity for biomedical researchers to uncover the next healthcare breakthrough. Safe Harbor methodology requires 18 PHI identifiers to be masked or removed—making data preparation a complex undertaking. To combat these vulnerabilities, biomedical studies must be de-identified in such a way that it can still be of value to researchers without revealing patient identity. Dicom Systems offers a proven and scalable de-identification toolset that unlocks valuable imaging studies for areas such as research, policy assessment, and comparative effectiveness studies.

Consumption of highquality data by deep learning applications is an essential contribution to better machine learning algorithms, unleashing tremendous potential for AI solutions that benefit patient care.

Benefits of Unifier De-Identification

  • Proprietary framework takes HIPAA Privacy Rule, Safe Harbor methodology compliance to a new level
  • Supports full DICOM and HL7 interoperability with all compliant devices
  • Best price-to-performance technology trusted by top healthcare enterprises, government agencies, and imaging partners
  • When deployed in conjunction with Dicom Systems Enterprise Imaging Unifier VNA, leverages robust framework for imaging lifecycle management and archiving




Features of Unifier De-identification Framework

  • Capacity to implement complete de-identification framework from data preparation and migration to building a data lake
  • Bi-directional dynamic tag morphing makes changes on input and output
  • Advanced pixel-level de-identification while avoiding accidental corruption or truncation of the image file
  • Complex DICOM tag substitutions, removals or morphing are automated by designing transformations into LUA script
  • Full customization of de-identification processes and output












Download product brief


De-identification for AI/Machine Learning Webinar

Anonymize PHI in medical images and follow HIPAA Safe Harbor guidelines by ensuring you have a complete de-identification solution. Cross-disciplinary research needs data sharing to assure its clinical validation. Can healthcare organizations avoid breaching HIPAA privacy rules and remove complex DICOM data objects? How can sensitive data like social security numbers, zip codes and the like be obscured, leaving raw data for research purposes? This timely webinar details the requirements for meeting the HIPAA Privacy Act and how to construct a clean data pool for research consumption. Each data set will be impossible to use to identify an individual subject of the information. Dicom Systems also shares a recent customer’s successful experience with big data management. Beginning with de-identification and migration the journey proceeds on to creating an effective data pool.


De-Identification Customer Success Story

Heath IT leader Dicom Systems completes large-scale, advanced de-identification of medical imaging data for Hospital for Special Surgery, a global leader in musculoskeletal sciences.

Campbell, CA, August 22, 2018 – Dicom Systems, Inc. (www.dcmsys.com), a leader in Enterprise Imaging IT, announced the completion of a large-scale de-identification project for Hospital for Special Surgery (“HSS”) in New York, NY, the global leader in musculoskeletal health and the leading orthopedic hospital in the U.S. Learn more >