The benefits of digital pathology for healthcare

Digital pathology has become widespread worldwide in more than just academic labs. Contract research organizations (CROs) and pharmaceutical companies reap the benefits of digitization in managing and analyzing millions of samples while seeking to reduce costs and time in developing viable drug candidates. Clinical labs, on the other hand, find it particularly helpful for producing quantitative results, taking part in remote diagnostic consultations, and ultimately providing better patient care.

This article introduces six digital pathology benefits applicable to general pathology labs across industries.

1. Enhance lab workflows

Labs that use traditional image analysis methods often suffer from bottlenecks and delays. By digitizing their pathology practices, labs can streamline their work, allowing them to take on larger caseloads while reducing turnaround times. Integrating digital pathology solutions with existing laboratory systems is critical, allowing smoother and better-controlled workflows.

For example, a study¹ on digital pathology implementation, conducted in Memorial Sloan Kettering Cancer Center, revealed a 25% reduction (1 day) in turnaround times for surgical resection cases where the patients’ history included prior WSIs. When pathologists have access to the patients’ entire WSI history in the LIS, they can compare prior patient pathology encounters to prospective cases, allowing for more efficient patient care. 

Another study² compared digital and microscopic diagnostic times of 400 cases (1396 slides) in histology, nongynecological, and fine needle aspiration cytology. The total diagnostic time was 115 minutes shorter in digital diagnosis (1841 min vs. 1956 min), amounting to an additional 250 cases a year for the reporting pathologist. The authors listed the following factors as contributors to shortened digital diagnostic time: Ergonomics (less tiring reporting sessions), larger viewing field (fewer viewing fields), and absence of physical slide handling (faster switching between slides/magnifications and reduced checking slides against requisitions). 

2. Collaborate efficiently

Digital pathology breaks geographical barriers. The electronic transfer of slides from the laboratory to the pathologist makes it possible for work to be shared across sites. This allows for real-time workload distribution, ensuring that pathologists across the organization are evenly utilized. It also helps in the local shortages of pathologists or when reallocating cases is needed due to sick leaves. Remote secondary consultation, collaboration, and even remote education are seamless with digitized images, particularly when viewed and shared via cloud-based software.^3-5

3. Standardize training for new staff

Digital pathology ameliorates another logistical problem: the time it takes to train new pathologists or specialists. Digitalization not only enables the use of digital images but also makes it possible to use AI assistance for image analysis, increasing consistency.

Before digitalizing their work, Mikko Airavaara, a Researcher at the University of Helsinki, described the challenges: “The current method is very limited by the time required to use the microscope, and it is limited in training a student in this and depends on how well and who trains the student. The variability between the persons to count the neurons is an issue.” Read the full case study to learn how the Aiforia® Platform helped the team save time and reduce variability.

Digitized learning materials and digital training resources remove the need for manual preparation and allow for higher-quality materials to be shared consistently with more than one person at a time. 

4. Save costs

Hospital administrators and departments must weigh the initial costs of implementing digital pathology against the potential cost savings it will bring in the future. A five-year study¹ conducted in a high-volume, academic, tertiary care cancer center showed a noteworthy increase in efficiency and operational utility when implementing digital pathology. 

The comparative costs included in the study included personnel, hardware and software, service agreements, IT infrastructure, digital storage, glass slide physical asset storage, and off-site storage vendor services. The calculations show a $1.3 million savings for a protected 5-year period. 

Another large-scale study³ was conducted in a sizeable academic medical center in Western Pennsylvania, expecting to fully adopt digital pathology upon its regulatory approval for primary diagnostic use. The cost savings estimates were based on two main benefits associated with the use of digital pathology:

1. Potential improvements in workflow/productivity and lab consolidation
2. Avoided treatment costs due to reduced rates of interpretive errors by general, nonsubspecialist pathologists

The 5-year productivity savings (1) were estimated at $12.38 million, mainly due to more efficient staffing. The interpretive errors and their related extra treatment costs (2) were calculated based on melanoma and breast cancer cases and then extrapolated to a broader cohort of cancers. Considering the incremental adoption of digital pathology in five years, the total avoided cost, based on improved diagnostic accuracy and related reduced errors and thereby avoided cancer treatment expenses, was estimated at $5.35 million. Thus, the total savings throughout the 5-year digital pathology system rollout were suggested at $17.73 million.

5. Gain new insights

Digital pathology enables the use of AI-assisted image analysis solutions, unlocking even more benefits for pathologists. A survey⁶ conducted for 127 anatomical pathology laboratories in Europe and Asia revealed that 36% of the respondents rely on digital pathology to enable the use of computer-aided diagnosis algorithms. The most common use case for AI algorithms mentioned in the study was objective scoring and quantification of immunohistochemistry. The overall perception of AI among pathologists was favorable, with a majority of the respondents stating that they believe in a synergy between AI and pathologists and that these tools would complement their current practice.

Whole slide images can be viewed in high quality on a computer screen and with software that allows fluid panning and zooming, enabling pathologists to see better and more. These viewing functionalities, combined with artificial intelligence, can help users to discover patterns or markers they may have missed if using traditional, manual methods. Machine learning can also detect features beyond the assessment of traditional histopathology, such as prognosis⁵. For example, Dr. Anna Laury, a pathologist and a researcher at the University of Helsinki, and her team leveraged Aiforia® Create, Aiforia’s versatile AI development tool, to successfully train a deep learning model to identify morphologic regions that can then help predict patient response to platinum-based therapy in high-grade serous carcinoma. Read the case study here.

Learn more about this topic: Deep learning in digital pathology: what, why, and how?

6. Improve patient care

In the clinical setting, all the points above ultimately affect the most important element: better patient care. As the discussed examples show, digital pathology can have major implications on patient care due to improved diagnostic accuracy and reduced errors. 

Many digital pathology systems and software enable integration with hospital information systems and patient registries. Connecting clinical patient information to the data obtained by the pathology lab enhances the ability of healthcare professionals to provide personalized care.

Furthermore, the assistance from AI-based solutions helps detect pathologic features that are diagnostically challenging, features that pathologists often disagree on. One concrete example is QuantCRC, a prognostic AI model developed by Dr. Rish Pai, MD, PhD, Pathologist at the Mayo Clinic. The model identifies different tissue characteristics in colorectal cancer patient samples, and combined with two other clinical parameters, it produces a colorectal cancer recurrence risk score. This can make it easier to decide between three and six months of chemotherapy, reducing the total number of patients getting extensive chemotherapy, but ensuring it is recommended for high-risk patients who can most benefit from it.

For more information, check our open guide: Digital pathology and AI: your guide to basics and beyond

References