Digital Pathology

Digital pathology refers to the process of converting tissue samples on glass microscope slides into digital images and examining these images in a computer environment. This transformation creates a digital work environment in pathology laboratories by moving traditional light microscopy-based diagnostic methods into a virtual space. Similar to the transition from film to digital in radiology, digitizing tissues enhances practicality and opens up new possibilities. For instance, thanks to high-resolution scanners, an entire tissue section can be recorded as a single “digital slide,” allowing pathologists to zoom in and out on their computer screens as if they were looking through a microscope. The concept of digital pathology encompasses not only image scanning but also the integration of digital tools such as telepathology (remote pathology consultation) and artificial intelligence-assisted analyses into pathological practice. As a result, the digital image of a pathology case can be instantly shared with experts in different geographic locations, reducing waiting times for second opinions and enabling the use of objective data provided by automated image analysis in diagnostic processes.

The most fundamental component of digital pathology is whole slide imaging (WSI) technology. In this technique, a specimen on a glass slide is fully scanned using a specialized scanner to produce a high-resolution digital image file. The resulting digital images are stored on secure servers and made available to pathologists through software on computers. Digital pathology has created a dynamic, image-based work environment that enables the acquisition, management, sharing, and interpretation of pathology information in digital form. This field is rapidly expanding to include numerous applications ranging from routine diagnostic reporting and consultations to education and research.

History

The foundations of digital pathology extend back to the second half of the 20th century, at the intersection of pathology and communication technologies. The idea of remote pathological examination was first proposed in the 1960s; indeed, in 1968, microscopic images from a hospital station at an airport in Boston were transmitted via analog video to pathologists at Massachusetts General Hospital, an experience now regarded as a precursor to digital pathology. These early initiatives were experimental applications considered under the umbrella of telemedicine at the time. The term “telepathology” entered the literature in the 1980s, describing the real-time transmission of slide images through remotely controlled robotic microscopes. Telepathology systems officially introduced by Dr. Ronald S. Weinstein in 1986 were applied in specific situations such as frozen section examinations, consultations, and transplant pathology. Although digital image transmission was still limited to niche applications during this period, the success of the concept laid the groundwork for the broader digitization of pathology practice in the future.

The late 1990s and early 2000s marked turning points in digital pathology. Thanks to technical advances, whole slide scanners (WSI) became commercially available, making it possible for the first time to digitize entire specimens without gaps. These high-speed scanners, introduced approximately 20 years ago, opened the door to using digital images not only for research and education but also in routine diagnostic workflows. However, for a long time, digital pathology remained largely confined to pilot projects and limited applications; laboratories were slow to fully transition to digital systems due to concerns and technical barriers. From the 2010s onward, increased scanning speeds, improved storage solutions, and enhanced software led to a significant rise in the clinical role of digital pathology. A major milestone was reached in 2017, when the U.S. Food and Drug Administration (FDA) granted its first approval for a digital pathology system (Philips IntelliSite) for primary diagnosis in surgical pathology. This was followed in 2020 by approval of a second system (Leica/Aperio). These developments signified the official recognition that digital pathology could be used directly in patient diagnosis, not merely as a research or consultation tool.

The acceleration of digital pathology in recent history was significantly influenced by the COVID-19 pandemic. In 2020, many laboratories faced urgent needs for remote work and digital access; in response, regulatory agencies implemented temporary measures to facilitate the use of digital pathology. For example, during the pandemic, the FDA issued guidelines temporarily relaxing restrictions on performing pathology diagnoses remotely via digital platforms. Similarly, some health authorities rapidly updated standards for digital reporting. The pandemic accelerated the transition to digital pathology by several years, prompting many centers to invest in digital infrastructure. Today, digital pathology, built upon the cumulative progress of these earlier steps, is increasingly becoming part of routine workflows in more laboratories.

The Digitization Process

The digitization process at the heart of digital pathology can be summarized simply as the scanning of a pathology specimen and its integration into a digital archive. This process begins after conventional histopathological preparation steps: following biopsy collection and tissue sectioning and staining, the sample on a glass slide is placed into a digital scanner. The scanner automatically focuses across the entire slide, captures high-resolution images, and stitches them together to create a massive image file. The resulting digital image — often containing gigapixel-level detail — is saved to the laboratory’s network and stored on a powerful server. During this process, the image file is typically in a proprietary format and linked to image management software. The pathologist then accesses the image via a digital viewing software. This software allows the user to pan, zoom in and out, and examine different areas of the image. In this respect, examining a digital slide is analogous to zooming into details on Google Maps: you can view the panoramic image of the entire tissue and then zoom in to examine cellular-level details. Pathologists evaluate digital slides using standard computer monitors or specially calibrated medical displays and can annotate important areas or perform measurements as needed.

Effective operation of the digitization process requires a robust information technology (IT) infrastructure. First, because scanned images can be extremely large, high-capacity data storage units and backup systems are essential. The images of hundreds of slides scanned daily in a pathology laboratory can quickly occupy terabytes of storage, making secure storage and rapid retrieval critically important. Additionally, if images are to be shared over a network, especially for remote access, high bandwidth and low latency are required. For example, it is generally recommended that hospital infrastructure support at least a 10 Mb/s connection to allow a pathologist working from home to smoothly view images stored on a central server. Otherwise, loading delays or interruptions can slow down the diagnostic process.

Quality control and validation steps are also crucial during the digital scanning process. Optical focusing of scanners may not be perfect at every level; certain areas may appear blurry due to thick tissue sections or air bubbles on the slide. In such cases, the slide may need to be rescanned or critical areas verified using a conventional microscope. Experienced digital pathology centers use software that automatically evaluates scanned images for criteria such as focus quality and color accuracy to detect potential errors early. During the archiving phase, it is mandatory that each digital image be correctly labeled with identifying information (patient and block details) to establish a well-organized digital slide library over the long term. When sharing digital images, compliance with patient privacy and data security protocols is required; images are typically sent with patient identifiers removed or via encrypted links for consultation purposes. Although this entire digitization process adds steps compared to manual microscopy, when integrated with appropriate infrastructure and workflows, it significantly enhances the speed and efficiency of pathology services.

Digital Pathology Ecosystem

For digital pathology technology to successfully integrate into daily practice, a comprehensive ecosystem must be established. This ecosystem is a holistic structure composed of hardware, software, human resources, and standards. Academic literature emphasizes three main components of the digital pathology ecosystem: information systems, digital pathology systems, and system tools. Information systems include hospital and laboratory information management systems (e.g., laboratory information system [LIS], hospital information system [HIS]). Digital pathology systems encompass the infrastructure that creates and uses digital images, such as scanners, digital viewers, image management software, and high-resolution monitors. The third component, system tools, refers to applications running on the digital pathology platform; examples include image analysis programs, artificial intelligence-based diagnostic support software, and data-sharing tools. When these three components function together harmoniously, the digital pathology ecosystem supports high-quality and efficient patient care.

Establishing and operating a digital pathology ecosystem requires multidisciplinary collaboration. Pathologists, technicians, IT specialists, and administrators must work together to plan and implement the system. Since each pathology laboratory has different workloads, case types, and needs, the setup of digital systems requires institution-specific planning. For example, the digitalization needs of a large academic center differ significantly from those of a small private laboratory in scale and scope. However, the core elements of the infrastructure are similar, and many institutions can adopt comparable framework components. This framework includes installing scanners and servers, establishing network and storage infrastructure, integrating software, and providing user training. Integration is particularly critical: the digital pathology system must be compatible with existing hospital information systems and other image archive systems (e.g., radiology PACS, electronic health records). In an integrated ecosystem, digital images generated by pathology can be directly linked to patient record systems, accelerating reporting processes and enabling collaboration across disciplines (e.g., radiology).

Compliance with international standards is also vital for the sustainability of the digital pathology ecosystem. In environments where equipment and software from different manufacturers are used, standardized data formats and communication protocols ensure long-term interoperability between systems. For this purpose, standards widely adopted in health informatics — such as DICOM (medical digital imaging standard) and HL7/FHIR (health information exchange standard) — are also applied in digital pathology. For example, leading scanner manufacturers have begun offering the option to save scanned images in DICOM format, enabling digital slides from one device to be opened in another institution’s software. Similarly, standardized messaging protocols (such as HL7) are used to transfer digital pathology reports into hospital systems, facilitating integration between disparate platforms.

Adherence to these standards enables the digital pathology ecosystem to remain flexible and resilient over the long term. As technology evolves or data sharing between institutions becomes necessary, compatibility issues are minimized. The digital pathology ecosystem is a comprehensive environment that brings together devices, software, human factors, and standards with the goal of enhancing patient care quality.

Applications and Future Perspectives

The applications of digital pathology are expanding rapidly, and future expectations are highly positive. Clinical diagnosis is one of the most important applications: digital systems have begun to be used for routine biopsy and surgical material reporting in some hospitals. Particularly through telepathology, pathologists in remote regions can now provide consultations based on digital images. This enables areas with limited access to pathology services to receive rapid second opinions from major centers. Digital pathology has also been successfully tested in frozen section (intraoperative rapid diagnosis) procedures: fresh tissue sections obtained during surgery are scanned and immediately transmitted to a pathologist, allowing remote diagnosis during the operation. In addition, digital pathology has revolutionized education: medical students and pathology residents can access digital slide archives from anywhere in the world to study rare cases and practice at their own pace using virtual microscopy applications. Through seminars and digital “tumor panels,” multiple experts can discuss the same digital image simultaneously. Quality control and archiving are also significant applications; digital storage eliminates the risk of slide loss, allowing even cases from years ago to be instantly retrieved and reviewed. This ensures the integrity of medical records and facilitates retrospective analysis when needed.

Digital pathology has also opened new horizons in the realm of research and development. Especially with advances in artificial intelligence and image analysis techniques, it is now possible to discover novel biomarkers from digital pathology data. For instance, in a research setting, machine learning algorithms applied to digitized pathology images can identify microscopic patterns in tumor tissues that are imperceptible to the human eye but carry prognostic significance. Indeed, recent studies integrating digital pathology data with molecular data have begun to infer genetic mutation profiles of tumors directly from pathological images. Such approaches strengthen the role of pathology in the future of personalized medicine. In summary, digital pathology research holds the potential to generate new scientific knowledge about diseases and may achieve diagnostic performance surpassing human capability through artificial intelligence support.

Looking ahead, the integration of digital pathology with artificial intelligence is an exciting frontier. Currently, AI algorithms are already being used to perform specific tasks on digital pathology images. For example, an AI software can automatically detect and mark metastatic cancer foci in a digital lymph node biopsy. Such clinical decision support algorithms should be viewed not as replacements for pathologists but as tools that enhance their work. AI provides high speed and consistency in repetitive, time-consuming tasks — such as counting hundreds of cells or identifying mitoses across an entire slide — allowing pathologists to focus on more complex interpretive work. In coming years, approved AI applications are expected to become a standard component of pathology practice. Some AI-assisted systems may even detect subtle foci that pathologists might overlook, acting as a safety net in diagnosis. In light of these developments, the role of the pathologist may evolve: a future work model is envisioned in which routine tasks are automated, and pathologists assume a more prominent role as supervisors and decision-makers. Similarly, integrating digital pathology with other medical data — such as radiological images and genomic data — will enable a more holistic approach to diagnosis and treatment planning.