CAP TODAY Pathology/Laboratory Medicine/Laboratory Management
Dr. Schmidt (Cortex): The underlying driver is customers needing functionality but wanting to minimize the costs of obtaining it, such as data center and IT administration costs. We are implementing technologies for customers to move in this direction when they choose to do so.
How is artificial intelligence affecting anatomic pathology labs and their future plans?
Johnson (Orchard): AI is being talked about a lot right now; it has potential. But it’s still far enough off that it is not affecting plans today. We believe the first area in which AI will make the biggest impact is in augmenting pathologists’ diagnoses. Computers will be able to take in more information than a pathologist can, particularly when it comes to difficult, specialized, one-off, uncommon cases. The other place AI will make a big difference is in workflow efficiency. Having the ability to augment rules-based technology standards with artificial intelligence ramps up efficiencies and the workflow in the lab to a whole new level. But it’s still a few years away.
Nollar (Xifin): Artificial intelligence in the world of pathology, in some respects, has been around for a long time. Whole slide imaging and image analysis technologies have been available but have seen limited adoption because of inability to gain FDA approval for primary diagnosis. The recent progress digital pathology providers have made should change the marketplace rapidly. Integrating digital pathology AI into existing workflows should allow pathologists to screen out negative cases and focus their time on critical cases. Xifin has been involved in integrating digital pathology and image analysis solutions since the early days of Chromavision, Aperio, and BioImagene and continues to integrate the latest digital pathology solutions into its LIS workflow for review and reporting. Beyond AP, AI is poised to play a major role in clinical decision support, bioinformatics, and big data analysis solutions across many other testing modalities, including molecular diagnostics and genomics.
Soufi (NovoPath): There are many ways in which AI may soon be ready to augment patient care. These include scanning the EHR for previously performed test results, analyzing genetic data, ordering additional tests, suggesting treatment options, and supporting a pathologist’s diagnosis with computational diagnostic information. However, in the core anatomic pathology laboratory today, AI is still considered an early-stage technology. As AI matures, NovoPath will evolve with it to meet client needs.
Meyers (Sunquest): AI may help laboratories and pathologists become more efficient and scalable, as well as more creative and data driven, to enable faster, more accurate diagnoses. Digitization of AP data could save a significant amount of time for physicians and laboratorians because using digital information is effective and timely and supports rapid analytics and image retrieval. Most important, digitization reduces the barriers between sites and facilitates consultations and referrals without boundaries. This will ensure that pathologists can have an impact on more patients. Algorithms can be used in decision support for a wide range of disease classification and may enable pathologists to spend more time on the most complex and challenging cases. Finally, AI could integrate incredibly complex, varied data from numerous sources to help pathologists make diagnoses with higher accuracy and confidence, leading to more precise, targeted treatment options for patients, all while simultaneously reducing human error.
Trentadue (Epic): AI is certainly a popular area right now and it is important to differentiate the marketing hype from the demonstrated execution. At Epic we have incorporated AI into our software development process in two ways. First, we formed a data science team of experts in math and statistics who work to understand patterns in data to create predictive models that are ultimately embedded into the final software. Second, we have a team of software engineers who focus on the technical underpinnings of platforms as well as how to visually surface machine learning insights to users. To date, we have over 20 machine learning models available, and more than 250 customers are running at least one model in their production system.
Dr. Schmidt (Cortex): The industry seems to be in the early days of AI integration as AI capabilities are being developed. Many groups are watching carefully and readying themselves for adoption as specific AI functions mature. AP-LIS vendors need to be ready to include new features as the marketplace recognizes value. Whether this is through in-house development or through partnership with AI companies depends on the cost and complexity of including AI solutions in the AP-LIS products.
How will the trends of practice consolidation and practice complexity affect your company and the questions and needs you want to answer in the marketplace?
Johnson (Orchard): At Orchard, we listen to our clients in order to really understand their work and work environment, evaluate what changes are taking place, respond to thought leaders, and step up to become thought leaders with regard to the AP-LIS. It is only then that we can provide the best path forward for customers through this turbulent, changing time. AP-LIS vendors need to focus on labs’ understanding of how data needs to be collected from a complex analyzer, and how to integrate digital pathology and all the new tools available to pathologists in a cost-effective and efficient way.
Nollar (Xifin): One of our largest growth areas has been with labs that have robust acquisition strategies and want all their labs on the same cloud-based LIS, as well as administrative functionality that enables them to “spin-up” new labs and manage all labs under a single LIS build.
We are also seeing the trend toward more complex practices reflected in an increased demand for lab information systems that perform next-generation sequencing. Many labs have developed next-generation sequencing assays for research and want to take them commercial and/or supply diagnostic testing for clinical trials. They need a system that can adapt to the requirements of the commercial marketplace and, in the case of clinical trials, can manage and report aggregate data by project. These labs are asking for flexible, configurable tool sets that can adapt to any type of testing, and they want user-configurable tools to handle their specialized and clinical trials testing.
Overall, labs recognize that the LIS is a critical, strategic component to their overall revenue growth strategies.
Soufi (NovoPath): In our experience, when laboratories are given an opportunity, they favor remaining with us. Otherwise, there is usually a transition and transfer time lag since the lab is usually one of the last departments to be consolidated. NovoPath hasn’t been affected much by this trend. I would attribute this to the strength of our specialized offering and high level of service. On the positive side, we are noticing that new labs continue to form because they recognize there is a gap in the service and/or value being offered. These labs are looking at their LIS investment as a competitive advantage.
Meyers (Sunquest): The consolidation dynamic is a new reality that has an impact on the entire marketplace. It leads to more integrated pathology groups and laboratories, requiring greater IT flexibility and functionality. The spectrum of patient care interaction points continues to expand, shifting lab operation models from homogeneous ones to more heterogeneous ones. Sunquest must continue to support customers with high-quality diagnostic information to enable quality care decisions.
Trentadue (Epic): Consolidation in the health care industry as a whole has affected the laboratory as well. We see pathology practices getting larger and able to consolidate operations and create centers of excellence to drive better diagnosis and leaner processing practices. The laboratory is an integral part of the patient care team and is gaining more exposure as practices grow and specialize.
Dr. Schmidt (Cortex): The trend toward practice consolidation is continuing and there is pressure on independent pathology labs to merge or join larger enterprises. Surviving AP-LIS products will handle practice complexity—multiple labs, complex reporting requirements, efficient workflows, complex interfacing capabilities, workload and management tools, whole slide imaging integration and partnerships, and advanced molecular testing integration and partnerships.n