From microscope to pixels: the new age of digital pathology

Key takeaways:

• • • • Digital pathology is becoming the new standard in Asia-Pacific, replacing light microscopes with high-resolution digital slides to streamline workflows and improve diagnostics.
      • Beyond digitisation, full digitalisation reimagines the entire diagnostic pathway – integrating automation, AI, and data connectivity for faster, more accurate results.
      • AI-powered image analysis enhances detection, quantification, and classification, improving reproducibility while freeing pathologists to focus on complex cases.
      • Open, interoperable ecosystems allow integration of best-in-class third-party AI tools, ensuring flexibility, innovation, and long-term adaptability.
      • Early adoption strengthens cancer care pathways, enabling earlier detection, faster turnaround times, and broader access to specialist expertise – even in remote areas.

Across the Asia-Pacific region, pathology laboratories are undergoing a profound transformation. The shift from conventional light microscopy to digital pathology is no longer an emerging trend – it is fast becoming the new standard. Simply put, digital pathology is the conversion of glass slides traditionally viewed under microscopes into high-resolution digital images that can be accessed on computers by pathologists. This evolution is not simply a matter of technology adoption; it is about reshaping diagnostic workflows and addressing some of the most pressing challenges faced by healthcare systems today.

Digital pathology is not just a modern alternative to glass slides. It represents a fundamental shift in how we view, interpret, and manage diagnostic information. The question is no longer if digital pathology will become mainstream – but when and how effectively institutions can make the transition.

The evolution of pathology laboratories: from digitisation to digitalisation

The journey from light microscope to a digital-first model can be seen in two phases: digitisation and digitalisation.

Digitisation began with the scanning of glass microscope slides to create whole slide images (WSIs) for archiving or sharing. While this solved many logistical issues – such as physical storage and remote consultations – it did not transform the pathology workflow itself [1].

Digitalisation, on the other hand, goes much further. It involves rethinking and redesigning the entire diagnostic pathway, where digital images become central to every step of the process – from sample preparation and analysis to collaboration, reporting, and integration with other clinical systems [1].

The benefits of digital pathology are substantial, extending beyond simple digitisation to transform diagnostic workflows and patient care:

• • • Collaboration without boundaries: Remote, simultaneous slide access enables real-time consultations, rapid second opinions, and quicker specialist input for patients in underserved areas.
    • Cost control and resource optimisation: Eliminating slide transport, couriers, and large storage cuts expenses, reduces waste, and saves large networks hundreds of thousands annually.
    • Improved workflow efficiency and turnaround times: Removing manual handling bottlenecks and enabling anytime access shortens diagnosis time by up to 42% and allows review of over 15× more slides per hour [3].
    • Precision and reproducibility in diagnostics: Integrated image analysis tools improve accuracy, grading, and biomarker quantification – essential for consistent, personalised treatment decisions.

Collectively, digital pathology fosters a connected diagnostic ecosystem where collaboration, cost efficiency, operational effectiveness, and advanced analytics work together to improve outcomes for patients and healthcare systems.

Despite these compelling advantages, widespread adoption of digital pathology remains uneven. Many institutions face hurdles related to upfront investment, the need for staff training and workflow redesign, and stringent data privacy or localisation requirements. These challenges, however, are increasingly being addressed through clear digital strategies, strong stakeholder engagement, and an emphasis on long-term clinical and operational value over short-term disruption. It is within this context that the key market drivers and barriers shaping the evolution of digital pathology must be understood.

What’s powering the shift: challenges and technological advancements

1. Challenges accelerating the adoption of digital pathology

Several structural and clinical pressures are pushing laboratories to rethink traditional workflows and move toward digital solutions:

1. 1. 1. Rising diagnostic demand – Ageing populations, a growing cancer burden4, coupled with an emphasis on early detection, are placing unprecedented strain on diagnostic services [5].
      2. Need for external consultation and specialised expertise – Increasingly complex cases require input from subspecialty pathologists. This has led to the emergence of a hub-and-spoke model, where centralised experts review cases from multiple sites [6].
      3. Workforce limitations – A shortage of qualified pathologists – particularly in rural or underserved areas – demands scalable, decentralised solutions. AI-assisted image analysis can help bridge the gap by screening potential abnormalities, with final confirmation from a human pathologist [7].
      4. Laboratory workflow bottlenecks – Rising specimen volumes, fragmented processes, and dependence on manual slide handling slow turnaround times and increase error risk [8].
      5. Pathologist workload and burnout – Administrative burdens and mounting caseloads create the need for workflow tools that support triage, prioritisation, and rapid review [8].
      6. Cost and operational inefficiencies – Physical storage, shipping of slides, and duplication add unnecessary expense and variability in reporting.
      7. IT complexity – Integrating pathology data with hospital EHRs and other diagnostic systems remains a significant hurdle in many settings.

2. Technological advancements driving digitalisation

At the same time, a wave of new innovations is enabling laboratories to overcome these challenges and fully embrace digital workflows:

1. 1. 1. Technology maturity – High-speed whole-slide scanners, cloud-based storage, and secure network infrastructure now make large-scale digitisation feasible for routine use.
      2. Data and digitisation – Moving from analogue microscopy to digital images opens the door to computational pathology and large-scale analytics.
      3. Advanced image analysis algorithms – AI tools can improve the reproducibility and confidence of pathology assessments by standardising the interpretation of subtle but clinically important features, such as HER2-Low expression. Other applications can extend further, detecting subtle features beyond the limits of the human eye, for example, MSI status directly from H&E slides.
      4. AI as quality control – Beyond primary diagnosis, AI can serve as a second reviewer, flagging potential discrepancies or missed findings to enhance consistency.
      5. Seamless data integration – Linking pathology images with other diagnostic modalities (radiology, genomics) creates a more holistic patient view.
      6. Efficiency and scalability – Digital workflows support central review across multiple sites, enable rapid second opinions, and reduce turnaround times.
      7. Pandemic-driven innovation – COVID-19 highlighted the value of remote pathology and telepathology for continuity of care.

Emerging trends and opportunities in digital pathology

Advances in automation, artificial intelligence, and open innovation are not only streamlining laboratory operations but also enabling deeper clinical insights and broader collaboration. These developments are opening new pathways for improved disease detection, treatment planning, and healthcare delivery – positioning digital pathology as a central driver of diagnostic transformation.

1. End-to-end automation and workflow integration

One of the most significant shifts underway is the movement toward full automation. Laboratories are increasingly looking to automate not just imaging but the entire pathology process – from sample tracking and barcode management to case assignment, slide scanning, image management, and digital reporting.

When integrated with laboratory information system (LIS) and electronic health record (EHR) systems, digital pathology platforms can [6]:

• • • Enhance traceability of each specimen from collection to reporting.
    • Reduce errors associated with manual data entry or slide mislabelling.
    • Support remote collaboration and second opinions without delays.
    • Strengthen diagnostic confidence with enhanced image clarity, AI-assisted interpretation, and real-time collaboration.
    • Enable round-the-clock processing through automation, allowing more slides to be prepared, scanned, and queued for review – helping laboratories deliver results faster, even beyond standard working hours.

This integration is particularly vital in environments managing high case volumes or experiencing staff shortages. Automation allows labs to scale operations while maintaining accuracy and throughput, allowing patients to get potentially life-changing results faster.

2. AI-powered algorithms

As digitisation generates vast volumes of high-quality image data, pathologists now have the opportunity to collaborate closely with data scientists, engineers, and AI specialists to extract insights that extend far beyond the limits of conventional visual assessment.

At their core, AI-powered algorithms are computer models – often based on deep learning and other advanced machine learning techniques – that are trained to recognise specific features in pathology images. By learning from large, expertly annotated datasets, these algorithms can detect, segment, quantify, and classify microscopic structures with a high degree of consistency. They can be designed for a wide range of tasks [9, 10, 11, 12]:

• • • Detection algorithms identify the presence or absence of disease features, such as malignant cells or metastases.
    • Segmentation algorithms outline regions of interest, for example, delineating tumour boundaries or highlighting stromal tissue.
    • Quantification algorithms measure expression levels of biomarkers in immunohistochemistry (IHC) stains, such as HER2, PD-L1, ER, PR, or Ki-67.
    • Classification algorithms assign diagnostic categories based on morphology or biomarker patterns, such as differentiating between carcinoma subtypes.

These tools are capable of operating in two complementary ways. As first-line aids, they can pre-screen slides to flag potentially relevant cases for review, streamlining the prioritisation of workloads in high-volume laboratories. As second-line reviewers, they can act as a quality control measure, offering a consistent, objective comparison against human interpretation and helping to identify subtle features that might be overlooked during manual review. For example, certain AI models can reliably detect HER2-low expression, a marker that can influence therapeutic decisions but is sometimes challenging to identify with complete certainty under a microscope [13]. Algorithms are considered companion diagnostic tools that identify patients who are eligible for targeted therapies.

The range of applications is expanding rapidly. Some algorithms focus on specific tumour types, such as prostate, breast, or colorectal cancer, while others are designed to assess prognostic or predictive biomarkers. Emerging models can even integrate histopathological features with genomic, radiological, and clinical data to forecast disease progression, estimate recurrence risk, or predict a patient’s likely response to a given therapy [9,10].

The greatest value of AI lies in creating a symbiotic diagnostic workflow, where repetitive, time-intensive tasks – such as mitotic figure counting, slide triage, or complex quantitative measurements – are delegated to algorithms. This frees pathologists to devote more time to nuanced interpretation and multidisciplinary case discussions.The result is not only faster turnaround times, but also more reproducible and richly detailed diagnostic insights – helping to advance the promise of precision medicine while preserving the central role of the pathologist in patient care [11].

3. Open ecosystems and collaboration with AI algorithm providers

A third major force shaping this landscape is the rise of open digital pathology ecosystems. Modern digital pathology platforms are increasingly designed to integrate third-party AI algorithms from a wide array of collaborators, creating a modular, interoperable, and innovation-friendly environment. Rather than locking users into a single proprietary toolset, these platforms enable pathologists to deploy best-in-class AI modules from multiple vendors directly within a unified viewer interface, reducing training burdens and supporting seamless workflow adoption. Many of these algorithms provide targeted insights – such as tumour grading, metastasis detection, or biomarker quantification – that complement the expertise of the pathologist rather than replace it. This approach ensures faster, more consistent, and clinically relevant reporting.

This open model has attracted significant global investment, with over USD 1.7 billion directed toward AI-driven digital pathology solutions since 2014, including a surge of 42% in 2021 alone [14]. Despite the rapid pace of innovation, deployment into routine clinical workflows remains challenging, hindered by regulatory requirements and the need for solutions that are hardware-agnostic and adaptable to local IT constraints. Notably, companies such as Qritive in Asia-Pacific exemplify how startups can bring value to this space by offering hardware-agnostic, on-premise AI modules that address regional priorities like prostate, colon, and liver cancer, while ensuring data sovereignty.

Ultimately, this combination of end-to-end pathology lab automation, AI-powered analysis algorithms, and open, AI-driven ecosystems fosters a connected and adaptive network that transforms static diagnostic data into actionable insights. This connectedness – across partners, data types, and geographies – empowers healthcare providers to enhance diagnostic accuracy, streamline workflows, and ensure that the right patient receives the right intervention at the right time.

What to prioritise as digital pathology matures

As digital pathology moves from vision to reality, choosing the right partner becomes a critical step in realising its full potential. When evaluating a digital pathology vendor, laboratories should look for a partner that offers end-to-end support – from high-quality slide scanning through to AI-driven image analysis and data integration – ensuring a seamless workflow rather than a collection of disconnected tools.

Equally important is the vendor’s commitment to openness: the ability to integrate and onboard third-party algorithms within an open ecosystem fosters innovation and future-proofs the platform. In parallel, the chosen vendor – or AI-algorithm provider – should demonstrate proven expertise in service and support. Beyond simply delivering technology, they must be able to provide responsive technical assistance, ongoing training, and proactive maintenance to ensure uninterrupted operations and optimal system performance.

Finally, robust expertise in navigating regulatory standards, data privacy, and cybersecurity safeguards is essential to protect sensitive patient data while maintaining compliance across diverse regions. These considerations ensure that digital transformation delivers not only speed and efficiency, but also resilience and long-term value.

The new age of digital pathology

The transition from light microscopes to full digitalisation is transforming pathology from a niche innovation into a clinical necessity. Across Asia-Pacific – especially emerging in Southeast Asia – digital pathology offers a pivotal opportunity to strengthen cancer care pathways, for example through earlier detection, reduced delays, and more personalised treatment.

When combined with AI-powered image analysis, these platforms improve diagnostic accuracy, enable multidisciplinary collaboration, and provide timely access to specialist expertise – even in remote settings. This is particularly vital for high-burden cancers such as lung and breast cancer, where earlier diagnosis can greatly improve survival. Open, interoperable ecosystems further ensure adaptability, integration of validated third-party tools, and regulatory compliance, helping to future-proof diagnostic infrastructure.

Clinically validated solutions, shaped by pathologists and tested on representative datasets, drive consistency, while robust cybersecurity and scalable hardware enable secure, seamless adoption across the care continuum. In a healthcare era defined by precision, speed, and connectivity, digital pathology is not just a game-changer – it’s the new backbone of modern diagnostics.

References:

[1] Eloy, C., et al. 2025. “Digital transformation of pathology – the European Society of Pathology expert opinion paper.” Virchows Archiv. doi:10.1007/s00428-025-04090-w.

[2] Zia, S., et al. 2025. “An update on applications of digital pathology: primary diagnosis; telepathology, education and research.” Diagnostic Pathology 20 (17). doi:10.1186/s13000-025-01610-9.

[3] Menter, T., et al. 2020. “Intraoperative frozen section consultation by remote whole-slide imaging analysis –validation and comparison to robotic remote microscopy.” Journal of Clinical Pathology 73 (6): 350–352.

[4] International Agency for Research on Cancer. 2022. Global Cancer Observatory. Accessed August 10, 2025. https://gco.iarc.fr/.

[5] Zehra, T., et al. 2023. “A suggested way forward for adoption of AI-Enabled digital pathology in low resource organizations in the developing world.” Diagnostic Pathology 18 (1): 68. doi:10.1186/s13000-023-01352-6.

[6] Mosquera-Zamudio, A., M. Gomez-Suarez, J. Sprockel, J.C. Riaño-Moreno, E.A.M. Janssen, L. Pantanowitz, and R. Parra-Medina. 2024. “Globalization of a telepathology network with artificial intelligence applications in Colombia: The GLORIA program study protocol.” Journal of Pathology Informatics 15: 100394. doi:10.1016/j.jpi.2024.100394.

[7] Walsh, E., and N.M. Orsi. 2024. “The current troubled state of the global pathology workforce: a concise review.” Diagnostic Pathology 19 (163). doi:10.1186/s13000-024-01590-2.

[8] Edayan, J.M., et al. 2024. “Informatics in Medicine Unlocked” 50: 101566. doi:10.1016/j.imu.2024.101566.

[9] Go, H. 2022. “Digital Pathology and Artificial Intelligence Applications in Pathology.” Brain Tumor Research and Treatment 10 (2): 76–82.

[10] Shafi, S., and A.V. Parwani. 2023. “Artificial intelligence in diagnostic pathology.” Diagnostic Pathology 18 (109).

[11] Jarrahi, M.H., et al. 2022. “The key to an effective AI-powered digital pathology: Establishing a symbiotic workflow between pathologists and machine.” Journal of Pathology Informatics 13: 100156.

[12] Gaffney, H., and K.M. Mirza. 2025. “Pathology in the artificial intelligence era: Guiding innovation and implementation to preserve human insight.” Academic Pathology 12: 100166.

[13] Mulder, D., et al. 2025. “Use of artificial intelligence–assistance software for HER2-low and HER2-ultralow IHC interpretation training to improve diagnostic accuracy of pathologists and expand patients’ eligibility for HER2-targeted treatment.” Abstract presented at the 2025 ASCO Annual Meeting.

[14] Signify Research. 2023. “Digital Pathology Investment Matures, VCs Get Selective.” Accessed September 4, 2025. https://www.signifyresearch.net/insights/complimentary-digital-pathology-vc-investment-analysis/.