Data Standardization
Interoperability has become a legal requirement (21st century cures act). Interoperability between digital components in the healthcare enterprise, consisting of various data types (EMR, images, specimen processing, LIS information, etc.) depends on standards. Other elements include data sensitivity, the need for “plug and play” as well as integration and best practices in handling real-world data create further arguments for data standardization. Various components of health data are stored differently. At this point these discrepant data silos do not integrate well because of legacy systems that are now confronted with new file formats that have been developed with new system functionality in mind. There is a clear need for data standardization, and we aim to create a database of relevant reference standards to help medical device developers understand what mixed, multi-modal data could look like. We emphasize the need for forward and reverse compatibility – and promoting conformity and standardization across the field.
Key Elements, Next Steps, Timeline
Interoperability (e.g., integration with LIS, scanner agnostic algorithms)
Standardization at different levels and different modalities in the laboratory (e.g., images, reports) and linking/tying information together
Integrate into medical record
Concerns & Problems
Why has standard not yet been adopted?
Consent issues if data is accessible more easily?
Can standard evolve and adapt fast enough to new requirements?
Democratization is long and complicated?
Value Proposition
R&D: Vendor neutral interpretation of data
Patients: sharing data between institutions in a format anyone can open and read, defined content for identification and matching
De-identification but still linked
Implications & Efforts
Circular dependency: manufacturers are waiting for customers and users are waiting for products
Starting somewhere: baseline features/elements that are required but still leaving flexibility (basic functionality versus perfect solution)
Taking action beyond white paper (including academic medical centers and pathologists)
Current Projects
News & Updates
Relevant Publications
DICOM Format and Protocol Standardization—A Core Requirement for Digital Pathology Success.
Date: 2021
Authors: David Clunie
Link: Pubmed
Interoperable slide microscopy viewer and annotation tool for imaging data science and computational pathology.
Date: 2023
Authors: Gorman et al.
Link: PDF Download
Contemporary clinical decision support standards using Health Level Seven International Fast Healthcare Interoperability Resources
Date: June 2021
Authors: Strasberg et al.
Link: PDF Download & Pubmed
Implementing an Integrated Large-Scale Clinical Information System for ISSSTE's Hospital Network in Mexico
Date: January 2021
Authors: Cervantes Trejo et al.
Link: Pubmed
Current State of the Regulatory Trajectory for Whole Slide Imaging Devices in the USA
Date: 2017
Authors: Esther Abels, Liron Pantanowitz
Link: Pubmed
Digital Imaging and Communications in Medicine Whole Slide Imaging Connectathon at Digital Pathology Association Pathology Visions 2017
Date: March 2018
Authors: Clunie et al.
Link: Pubmed
Computational pathology definitions, best practices, and recommendations for regulatory guidance: a white paper from the Digital Pathology Association
Date: November 2019
Authors: Abels et al.
Link: Pubmed
Implementing the DICOM Standard for Digital Pathology
Date: November 2018
Authors: Herrmann et al.
Link: Pubmed
Group Leaders
Markus Herrmann, MD, PhD
Dr. Herrmann serves as Director of Computational Pathology at the Massachusetts General Hospital and Assistant Professor of Pathology at Harvard Medical School. He is an interdisciplinary physician-scientist at the intersection of medicine, biology, and computer science with a specialization in biomedical imaging, computer vision, machine learning, and clinical data science. His research focuses on the development and evaluation of digital imaging technologies and image analysis software as well as their integration into laboratory workflows and translation into safe and effective clinical services. Dr. Herrmann has a strong interest in standard development and has been actively engaged in national and international efforts to advance the standardization and adoption of digital and computational pathology, including the DICOM Working Group 26 Pathology, the IHE Pathology and Laboratory Medicine domain and the CAP Digital and Computational Pathology Committee.
Mike Isaacs
Mike is the Director of Clinical informatics and Business Development in Department of Pathology at Washington University School of Medicine in St. Louis; A computer science background and over 25 years of experience in supporting Information Technology in the laboratory has provided Mike the ability to identify the informatics needs of the Department. For the last two years Mike has focused on building and integrating Digital Pathology Solutions for Internal use and for their Outreach Client business for WU/BJC Hospital; now His primary interest is providing a strategic vision for the implementation of Digital Pathology for the BJC HealthCare system, to better meet their patient’s needs. Mike is currently on the DPA board of Directors and is Co-Charing the Education Committee.