It’s time to break through the noise surrounding the AI-enabled digital pathology platform. As the case for AI adoption strengthens, thanks to growing evidence and regulatory clearances, more laboratories are beginning to deploy computational pathology. The AI-enabled platform has – quite rightfully – been posited as a future-proof means of bringing these applications into the workflow. It’s hardly surprising that we’re seeing a tranche of perspectives on what it is and what it should do. But how should laboratories make sense of these differing views to chart their path to success? Let’s unpack the key considerations.
An AI-enabled platform sits at the center of your digital pathology operations
Whether it's in the form of detection and prognostic solutions, or in applications that automate and eliminate routine tasks such as quality control, computational pathology can only add value when it is integrated into day-to-day operations. An AI-enabled platform must deliver the robust functionality and compelling user experience needed to power your routine practice. Only then can it meaningfully introduce computational applications into the workflow.
An AI-enabled platform powers computational pathology across the end-to-end workflow
Integrating AI into routine operations is a three-step process, which spans the end-to-end workflow and should be driven fully by the platform to empower the pathologist. This means that the AI-enabled platform must first ensure that a computational application runs at the appropriate times. Keep in mind that many applications are not only subspeciality-specific, but also only intended for use on certain patient cases. A true AI-enabled platform helps to manage this mapping, saving time for everyone in the lab.
Next, the platform should execute the AI solution, giving the pathologist the option to run it automatically or manually. The pathologist may want to automatically execute a triaging solution on all relevant cases. However, they will likely want to manually run an application that evaluates a region of interest after they’ve made the necessary annotation. Finally, once the application has been executed, the platform must seamlessly display AI results where and when the pathologist needs them so that they can make a more informed decision. And that is where the rubber meets the road – or, put another way, where the pathologist can realize the promise of the AI.
An AI-enabled platform supports a broad portfolio of applications
It’s inevitable that the modern laboratory will leverage a variety of computational solutions given that many of these applications are disease and use case-specific. Though the makeup of this portfolio will differ for each laboratory, some solutions will almost certainly come from a mix of AI companies, while others may be home grown.
The AI-enabled platform must offer native support for all of these applications, or it will fail to meaningfully incorporate computational pathology into the routine workflow. An open platform that integrates any AI solution will also continue to meet your needs as they evolve, and as new applications emerge.
Though many laboratories will understandably want to take small steps at the beginning of their computational pathology journey, it’s important that you think big to set yourself up for long-term success. A true AI-enabled platform will scale with your laboratory, giving the professionals within the opportunity to realize the full promise of digital and computational pathology today and in the future.
Newsletters
Receive the latest analytical scientist news, personalities, education, and career development – weekly to your inbox.
