Pathology Goes Digital and Takes Diagnosis to a New Level, Powered by Data and AI

Clinical diagnostics are enhanced with new imaging technologies to help physicians make decisions with a higher degree of precision.

By Michael Valante, Chief Technology Officer Digital Pathology, Dell Technologies

Pathologists are sometimes referred to as the doctor’s doctor. Why? Because they’re physicians with special training in tissue diagnosis and laboratory medicine, who help other physicians make or confirm a diagnosis, by studying tissue and fluid samples. Not only are they experts in interpreting cellular structures, they also determine the type and severity of a patient’s cancer and other diseases. The pathologist then works with other members of a patient’s care team to recommend a treatment.

In the past, pathologists used manual methods to examine thin slices of tissue mounted on glass slides. This approach wielded results, but with limitations. Now, new tools and insights are transforming previous ways of working. Instead of examining tissue samples through a microscope, new imaging technologies are being used so pathologists can review high-resolution digital versions, called whole slide images, and contextualize various data and decisions.

Time for Digital Pathology

As innovative tools change the way pathologists diagnose diseases, they’ll also enable them to collaborate more effectively remotely and eliminate unnecessary delays. The technology won’t just help the pathologist do his or her job more efficiently, it will improve communication between care providers.

Unlike glass slides that can take several days to reach distant labs via mail couriers, digital images can be transmitted electronically anywhere, in just a matter of seconds. By electronically sharing information to and from remote locations, and reducing turnaround times, pathologists can deliver crucial, time-sensitive patient care in geographic locations where there may be gaps in clinical resources or where a particular specialist is required.

The shift to digital is exciting but hardly surprising. Other diagnostic specialties such as radiology went digital years ago supporting the introduction of digital imaging innovations such as Magnetic Resonance Imaging (MRI) and Ultrasound. This growth in digital imaging data has created an opportunity to enhance clinical and diagnostic workflows.

Adding an AI Touch

As significant as the conversion from slides to digital images is (allowing for deeper and quicker analysis of disease), digital pathology is also an entry point for artificial intelligence (AI).

With digital imaging, pathologists can work with (AI) tools to identify specific areas of interest that warrant further investigation and even set them on a path towards precision medicine with treatments targeted at a specific individual. Combined with new gene sequencing techniques, AI algorithms are now learning to integrate information from various sources to predict personalized regimens for each patient. For example, cancer treatment depends on the delicate balance between removing the disease and saving organs. When an oncologist discovers a tumor and sends a sample for analysis, the pathologist can help the oncologist tailor a personalized treatment plan for the patient by sequencing their genome and tumor.

In the future, pathologists may use AI algorithms to achieve greater granularity in how they categorize a disease and provide a direct prognostic prediction of disease outcomes to help minimize the need for chemotherapy or radiotherapy while preserving parts of the anatomy.

These algorithms open the door to other possibilities and innovations. The next step in personalized medicine will be to use insights from digital images and build a virtual representation of a patient and their organs—called digital twins—to test treatments on the virtual counterpart of a patient’s physical state before adapting them for the individual.

Modern Medical Imaging Solutions

An increase in the prevalence of cancer and other diagnostic-intensive diseases, combined with the continued growth of imaging innovations, is leading to the explosive growth of imaging data. Many healthcare organizations are managing multiple Picture Archiving and Communication Systems (PACS) and image repository systems, which can create ongoing islands of data that lack interoperability and burden systems and staff to keep synchronized. In this evolving imaging environment, healthcare providers are looking for methodologies to manage medical imaging data more effectively and cohesively. To do this, they need to develop image management strategies with an enterprise view that define the right storage in the right place, supporting the right clinical workload at the right cost.

Digital pathology images, however, are significantly larger than radiology and cardiology images, making them harder to efficiently manage, share and store. Laboratories can generate petabytes of data annually when digitizing their pathology slides. The data must then be retained for many years. At the same time, the practical challenges of going digital need to be addressed given that the demand for digital pathology has skyrocketed recently, spurred by the coronavirus pandemic and the increased necessity to provide clinical services remotely.

The National Institutes of Health recently launched a four-year, $1.15 billion initiative to try to understand the causes and consequences of the lingering brain fog, breathing problems, and malaise reported by many recovering COVID-19 patients. Several labs are looking at ways to speed up their digital efforts to support the huge number of biopsies and pathology investigations researchers will require to gain a greater understanding of the biological causes of these prolonged symptoms.

The Doctor’s Doctor through a Digital Lens

Digital pathology can be transformative, but it’s only one stop in a healthcare provider’s journey to modernization. Digital scanners and systems to store images will start the journey but they’ll require new workflows that align with the enterprise strategies used by other departments, such as radiology, to support multidisciplinary care.

The convergence of different islands of care, to a fully integrated diagnostic system, will further enable pathologists’ collaboration with other clinicians to support improved patient outcomes and increased quality of care.

Arming pathologists with digital data and AI tools takes collaboration to new heights so physicians, surgeons, oncologists, and other specialists on a patient’s care team can more efficiently treat patients with complex diseases, like cancer.

When you consider its many benefits, it’s clear that digital pathology, powered by data and AI, can help healthcare providers achieve their quadruple aim of improving health outcomes, enhancing the patient and clinician experience, and reducing the cost of care.