As we approach the juncture when each patient will possess a universal electronic health record (EHR), the true importance of digital pathology is emerging. Regardless of your laboratory information system (LIS), or the make, model, and number of your whole slide scanners, system-wide integration is the key to successful implementation of a cohesive digital pathology service. Establishing common ground where disparate systems can communicate with one another is only the first step. The next step is to ensure that information is shared between systems in a way that enhances the anatomic pathology (AP) clinical workflow.
Digital pathology will require LIS integration to receive widespread adoption. This integration will, in effect, create what will become recognized as “digital pathology systems” (DPS).
Then and now
A dozen years ago, scanners were in their infancy and emphasis was placed on creating hardware that could reliably, reproducibly scan slides and convert them to whole slide images (WSIs). The level of software integration that would be required to store, view, share, annotate, and report pathology opinions on these images was unknown. The scanners were going to do their part, convert analog slides to digital images, and present them to a pathologist. The pathologist would then obtain the associated clinical data and metadata for the image(s) from his/her LIS. This process is exactly what surgical pathologists do every day with glass slides.
In recent years, labs have begun taking advantage of technological advances to streamline work processes and improve accuracy of reporting. Access to history and other ancillary testing through a patient’s EHR provides critical information about that patient’s disease process. Radiology results, molecular testing, and clinician observations provide valuable information when interpreting a specimen.
In the lab, barcodes are increasingly being used to correctly identify the patient’s specimen and track each step of the process from specimen collection and accessioning the case to grossing, embedding, making the slide, labeling the slide, and case distribution. At the time they sign out the case, pathologists merely scan the slide’s barcode at their desktop to “pull the case” up in the LIS. From there, they can review the patient’s history and gross description of the current specimen. The patient’s diagnosis is finalized in the LIS and then uploaded to the EHR for review by the clinician. The integrity of the patient’s identification is maintained throughout the process through the use of barcodes and scanning devices. And, with today’s specimen tracking systems, pathologists know exactly where their case is at any time in the process.
Digital pathology in the workflow
This same degree of positive patient identification and specimen tracking must be required when using digital pathology for clinical workflows to ensure the appropriate glass slides are scanned on the correct patient and then presented to the pathologist with the associated information in the LIS. The workflow then becomes a worklist-driven process, with the patient-centric LIS driving the images being viewed.
The process of patient identification highlights a sharp and distinct contrast between digital radiology and digital pathology. In digital radiology, there is positive patient identification when the images are acquired, becoming the substrate by which the case is “accessioned” and then reported in the radiology information system (RIS). This process is separate and distinct from the devices that capture the images, much like our whole slide scanners.
Unlike RIS systems, however, the LIS is required to further manage case information and additional workups on the specimen such as additional slide levels, special stains, immunohistochemistry, molecular tests, and send-out testing requests.
Already, pathology labs have begun to integrate their digital pathology systems with their LIS. Whether using WSIs for quality assurance, collaboration, teaching, research, or other applications, the pathologists do not want to work in disparate systems. The Department of Pathology at University of Pittsburgh Medical Center integrated its digital pathology platform with its LIS, thereby streamlining its digital pathology workflow. Its next steps include establishing a bi-directional data flow between the LIS and the digital pathology system and the integration of the clinical information and radiology images from the patient’s EHR.1 Other labs have also begun the process of integrating systems.
Return on investment
These efforts are not without cost in terms of time and expertise. Integration of these systems requires careful coordination among the LIS team, the healthcare facility’s information services team, vendors supporting the digital pathology service, and the lab staff. The payoff, however, can be significant in terms of productivity and lab consolidation savings and savings as a result of improvements in accuracy of diagnosis. If costs for a digital pathology system do not exceed these savings, not only is there a significant return on investment and improvement of patient care, but the pathology laboratory is positioned to take advantage of new tools such as computer-aided diagnoses.2
As the use of digital pathology expands, its applications are becoming more sophisticated. Quantified image analysis and other image-based diagnostic tests are already becoming standard for predicting disease progression and patient outcomes. Newer studies have shown that digital pathology may also serve as a means of aligning in vivo radiographic imaging and ex vivo histopathology to better correlate histomorphometric changes.3 This exciting new technology will require careful integration with the RIS.
These efforts to create integrated digital workflows come at a time when we are seeing a shortage in pathologists worldwide, while seeing an increase in the number and complexity of specimens. In an effort to control costs, healthcare facilities in the United States are banding together to create health systems across geographically dispersed locations. Providing immediate access to a patient’s complete record, including the digital images of their specimens, affords a new level of expertise and care.
Emerging opportunities
Within digital pathology systems, new technologies will create unique opportunities for pathology in near real-time that do not exist in an analog world. In doing so, the treasure troves of graphical content can be further incorporated into the bounty of textual data within the LIS. This technology will allow for analyses at a much higher throughput than can be achieved with analog slides alone. As DPS, LIS, RIS, and EHR systems become truly integrated, a comprehensive view of the patient’s disease process will be obtained, and a multi-disciplinary team can better manage the patient’s care, resulting in the best possible outcome.
REFERENCES
- Guo H, Birsa J, Farahani N, et al. Digital pathology and anatomic pathology laboratory information system integration to support digital pathology sign-out. J Pathol
Inform. 2016;May 4:7:23. - Ho J, Ahlers SM, Stratman C, et al. Can digital pathology result in cost savings? A financial projection for digital pathology implementation at a large integrated health care organization. J Pathol Inform. 2014;5:33.
- Madabhushi A, Lee G. Image analysis and machine learning in digital pathology: Challenges and opportunities. Medical Image Analysis. 2016;33:170-175.
Keith Kaplan, MD, serves as Chief Medical Officer of Corista, a pathology solutions provider. While at Walter Reed, in conjunction with the Armed Forces Institute of Pathology (AFIP), Dr. Kaplan founded and directed the Army Telepathology Program. Dr. Kaplan is the publisher of tissuepathology.com.
Robin Weisburger, MS, HTL(ASCP), serves as manager of Corista’s Client Support Services team.
