How AI Is Accelerating the Digital Transformation of Pathology

Digital pathology adoption for routine clinical diagnostics globally is relatively low but growing quickly. The development of AI tools requires large datasets to train the AI models. As more laboratories adopt digital pathology, this creates a 'flywheel effect'; as more data becomes available, we create better AI tools. In turn, more laboratories are compelled to utilize digital pathology, which enhances workflow efficiency.

There is also a learning curve for pathologists transitioning to digital workflows. Navigating images on a screen differs from using a traditional microscope and some tissue features on digital slides can look slightly different to their physical counterparts. To allow pathologists time to acclimatize to new systems and ensure clinical safety, laboratories are typically taking a gradual approach to adoption, starting with specific subsets of slides or tests.

Another, sometimes overlooked, challenge laboratories face when adopting AI-driven digital pathology systems is the need to optimize their tissue processing workflows to ensure they produce 'digital-ready slides' that meet the needs of AI and digital systems. Collaboration between laboratories and companies with expertise in the entire diagnostic ecosystem is crucial.

The advancement of digital pathology for computational and AI-driven enablement offers our health systems, providers and patients a lot of promise. With that comes significant consideration for the regulatory compliance approval and ethical use of AI-driven digital pathology systems.

For AI tools to be deployed in clinical practice, they must meet stringent regulatory standards that prove their safety, efficacy and reliability. This includes demonstrating that AI systems are unbiased and capable of delivering consistent and accurate results across diverse patient populations. Regulators often require rigorous validation studies to ensure that AI tools meet these standards, which can be resource-intensive for laboratories and developers. As these solutions grow, the FDA and other regulatory bodies continue to actively work to provide guidance to the many stakeholders of these AI-enabled advancements to ensure the utmost safety and effectiveness.

For laboratories, whether a test, device or diagnostic aid is approved by a regulatory body or not, both Clinical Laboratory Improvement Amendments (CLIA) and College of American Pathologists (CAP) require that it must be validated in the laboratory by the appropriate medical staff prior to being used as an assistive solution for clinical decision support.

Data privacy and security are equally important, as digital pathology systems and AI-enabled solutions need to safeguard protected health information and how it is used. Furthermore, with the rise in cybersecurity risk, considerations for robust risk management programs, personnel and governance will be needed. 

Looking to the future, several advancements in digital pathology are anticipated to further revolutionize the diagnostic landscape. One major development is the increasing integration of AI into diagnostic workflows. AI tools are expected to become more sophisticated and capable of handling complex tasks. Whether that is assisting in the analysis of multiplexed biomarkers, quantifying points of interest and/or data integration from vast sets across multiple sources all coming together to provide powerful tools for the care of those served. 

As more laboratories embrace digital pathology, the availability of large datasets for AI training will increase, leading to the development of more advanced and reliable AI tools. These tools, in turn, will drive further adoption by offering new diagnostic capabilities and improving the efficiency of workflows. This cycle of growth and innovation will likely further the global use of digital pathology.

By analyzing complex biomarker profiles, digital and computational pathology systems play a central role in tailoring treatments to individual patients. This approach has immense potential to not only improve patient outcomes but also reduce healthcare costs by ensuring that treatments are more targeted and effective.

Finally, advancements in automation and data integration are expected to streamline diagnostic workflows further. Fully automated end-to-end systems, from tissue processing to diagnosis, could drastically reduce human intervention and error while enabling faster turnaround times.

As digital pathology becomes more prevalent, it has the ability to revolutionize global health and its reach. Notably, for those in more rural or resource-limited settings, they will be able to adopt these technologies at scale, democratizing access to advanced diagnostics and improving healthcare equity worldwide.

This content includes text that has been created with the assistance of generative AI and has undergone editorial review before publishing. Technology Networks' AI policy can be found here.