Hematopathology Diagnostics Software: Can Data Alone Guarantee Diagnosis?

Key Takeaways

• Hematopathology diagnostics software now supports blood smear analysis, bone marrow evaluation, and flow cytometry – but data output alone cannot substitute for expert clinical insight.
• AI-driven cell classification accelerates workflows, but needs to go hand in hand with expert interpretation and contextualization.
• Variability across laboratory environments limits how well digital hematology tools can be adopted at scale.
• Integrating hematology software into clinical workflows enhances efficiency while keeping clinicians actively involved.
• The future of hematology diagnostics depends on effective collaboration between clinicians and AI systems.

How Hematopathology Diagnostics Software Is Changing Clinical Workflows

Hematopathology diagnostics software is reimagining how hematological disorders are identified and managed. However, the output data from these software systems is not a guarantee of clinical accuracy. Traditional microscopy methods such as blood smears, bone marrow aspiration, and genetic testing rely heavily on clinician expertise but are limited in speed, accuracy, and scalability, particularly when it comes to working with complex or large datasets. Increasingly, laboratory workflows are being transformed by digitalisation and advances in hematology diagnostics software.

Hematopathology software now analyses blood smears, bone marrow aspirates, and flow cytometry data to detect and classify disorders, while in the future it may also support outcome prediction, treatment response assessment, and therapeutic decision-making. In cytomorphology, some models trained on peripheral blood images can detect dysmorphic features suggestive of malignancy and assess red blood cell morphology in challenging conditions such as thrombotic microangiopathies.

AI is also an emerging promising tool in rare hematologic diseases, where diagnostic complexity, limited data, and lack of standardised pathways currently exist. Tools such as natural language processing, pattern recognition, and federated learning can help identify hidden patterns in fragmented datasets and support earlier diagnosis of these rare conditions.

Why Hematology Software Data Alone Doesn’t Equal Clinical Insight

AI applications – particularly machine and deep learning – can analyze large, multi-source datasets, including imaging, electronic health records, genomics, and laboratory results, automating complex, error-prone tasks. However, their value in hematology needs to extend beyond classification to diagnosis, and translating outputs into clinical insight. Automation cannot provide the answer to these more complex needs.

Trust in AI depends on its role and ability to be incorporated within clinical workflows, with higher patient confidence when it supports clinical decision-making, rather than acts autonomously without a clinician at the end of the technology. AI-based cell morphology differentiation systems are at their most effective when they enable hematology specialists to focus on cases requiring human expertise, rather than operating without oversight or rules restrictions.

The limitations of AI without human integration are well recognised. AI cannot act alone or independently but it can act as the clinicians “right hand” assisting in rapidly and accurately identifying similar cases to improve them. Human oversight is and will always be essential for training, validating, and interpreting outputs, ensuring a continuous “human-in-the-loop” approach.

A key challenge is limited explainability. Algorithms cannot clearly justify their outputs or the features driving decisions. In cytomorphology, where diagnosis relies on subtle cellular features, this lack of transparency is a significant concern, reinforcing why the clinician needs to be at the helm.

Challenges Limiting Hematopathology Software in Clinical Practice

Several structural challenges limit the reliable performance of hematopathology diagnostics software in practice. Diagnostic hematology workflows are complex and data-intensive, typically starting with complete blood counts and progressing to detailed morphological assessment of peripheral blood and bone marrow, with each step informing the next.

• Variability across laboratory settings
  AI models trained in one laboratory often underperform in another laboratory setting due to bias, overfitting, and “data drift,” all of which require clear definition and evaluation metrics. Data drift refers to differences between the data used in training a machine learning (ML) model and that applied to the model in real-world operation. Data quality, volume, and variability block translation from controlled settings to real-world clinical use.
• Regulatory and evaluation standards
  Regulatory frameworks, including evolving FDA guidance, struggle to keep pace with rapid AI advances. A key challenge is defining the evidence required to demonstrate that the model performance is generalisable to the intended population.
• Adoption barriers
  Adoption of hematology diagnostic systems may be limited by clinician unfamiliarity, difficulty with change and concerns about workflow disruption, highlighting the need for training and ongoing support. Other barriers include high costs of such solutions, limited standardisation, and data security concerns which prevent scalable implementation.

Peripheral Blood Smear analysis remains essential for diagnosing a range of hematological conditions, including anaemia, infections, as well as haematological cancers such as leukaemia. However, there are well-known limitations including it being time-consuming, operator-dependent, and prone to error in complex cases. While deep learning can rapidly analyse smear images and detect subtle abnormalities, diagnostic interpretation still and will always require expert clinical judgment.

Integrating Hematopathology Diagnostics Software into Clinical Decisions

AI is transforming blood smear analysis, bone marrow assessment, and genomic profiling by automating cell classification, detecting abnormalities, and identifying clinically relevant genetic changes. It can:

• classify cells in peripheral blood smears
• detect blasts in bone marrow aspirates
• quantify reticulocytes
• track changes in cell populations over time

Integrated with complete blood count data, patient history, imaging, and genetic testing, these AI tools provide a more comprehensive diagnostic picture.

In bone marrow analysis, AI demonstrates clear benefits, including:

• detecting blasts in acute leukaemias
• identifying dysplasia in myelodysplastic syndromes
• assessing cellularity in aplastic anaemia
• flagging monoclonal populations in myeloma or lymphoma
• predicting relapse risk in leukaemia based on bone marrow features

AI is also advancing the evaluation of coagulation disorders, where diagnosis and management are often complex and reliant on traditional testing and clinical judgment. Algorithms that monitor coagulation profiles, predict bleeding risk in haemophilia, and support decisions in thrombophilia offer a more dynamic and data-driven approach.

In resource-limited settings, AI-enabled peripheral blood smear analysis supports remote diagnostics via telemedicine. Digitized samples can be analysed and shared with off-site experts, improving access to specialist input, particularly in rural or emergency contexts.

Platforms like Scopio Labs’ Full-Field digital morphology system are already enabling this shift – capturing all clinically relevant areas of the sample at 100x resolution and delivering AI-powered decision support through a browser-based interface accessible from wherever the expert is, without returning to the microscope.

From Software Output to Clinical Diagnosis: Bridging the Gap

A key challenge in integrating digital health tools in hematology is interoperability. Electronic health records (EHRs), laboratory information systems (LIS), and diagnostic platforms must communicate seamlessly to provide a combined full picture of patient data. Ensuring data security is equally critical, requiring robust safeguards, regulatory compliance (e.g. HIPAA and GDPR compliance), and continuous updates to anonymisation and encryption to address evolving cyber threats.

Digital image management and cloud-based platforms are advancing the sharing of hematopathology data across institutions. The financial case is also strengthening, with Reddy et al, citing documented savings of $267,000 annually, or $1.3 million over five years, from reduced reliance on glass slides.

To meet rising demand with limited resources, laboratories are increasingly digitising workflows and leveraging computational tools to reduce hematopathologist and laboratory staff workload and prioritize expert input. This shift supports remote consultation and telepathology, enabling collaboration on patient cases and research across distances.

However, challenges remain in evaluating and integrating AI into clinical hematology practice. While tools influencing patient care must meet high standards, these standards are not yet clearly defined. As AI adoption grows, hematologists will need to work alongside these technologies to enhance diagnosis, improve efficiency, and deliver more personalised care, while ensuring implementation remains both patient- and provider-centred.

About Scopio Labs

Scopio Labs has developed Full-Field digital morphology platforms for peripheral blood smear and bone marrow aspirate analysis, enabling scalable, high-throughput remote review. Unlike cell-snapshot systems that capture individual cells in isolation, Scopio’s Full-Field platform captures all clinically relevant areas of the peripheral blood smear and provides Whole Slide Imaging for bone marrow aspirate – all at 100x resolution – without compromising field of view, supporting seamless navigation, AI-assisted cell pre-classification, and improved workflow efficiency.

The platform enables automated white blood cell pre-classification, platelet estimation, and comprehensive morphology assessment, with published data demonstrating up to 60% reduction in hands-on review time for peripheral blood smear analysis. In bone marrow aspirate evaluation, it provides high-resolution imaging with AI-enabled quantification across hematopoietic lineages, delivering outputs such as nucleated differential counts and M:E ratios with strong agreement to manual microscopy. By enabling browser-based remote review, standardized reporting, and integration with laboratory information systems, Scopio’s platforms demonstrate the practical impact of AI-driven hematology workflows in clinical practice.

Disclaimers:

*Scopio Labs’ Full-Field Peripheral Blood Smear application is CE marked and FDA-cleared, and it’s commercially available across the U.S., UK and Europe and other territories.
*Scopio Labs’ Full-Field Bone Marrow Aspirate Application is CE-marked and cleared for sale in additional regions. Not commercially available in the US for in vitro diagnostic procedures.
*Remote capabilities are available through the secure hospital network.

References:

1. Waggiallah HA, Al-Garni A, Ghazwani AAM, et al. The future insights of AI applications in hematology diseases diagnosis and prognosis: review article. Salud Cienc Tecnol. 2025;5:1430. doi: 10.56294/saludcyt20251430
2. Nazha A, Elemento O, Ahuja S, et al. Artificial intelligence in hematology. Blood. 2025;146(19):2283–2292. doi: 10.1182/blood.2025029876
3. Reddy A, Williams DKA, Graifman G, et al. Exploring the business aspects of digital pathology, deep learning in cancers. Intelligence-Based Medicine. 2024;10:100172. doi: 10.1016/j.ibmed.2024.100172
4. Nicholls M. Bridging the gap between pathologist and algorithm. Healthcare in Europe. 25 April 2021. Available from: https://healthcare-in-europe.com/en/news/bridging-the-gap-between-pathologist-algorithm.html [Accessed 12 April 2026]
5. Sahiner B, Chen W, Samala RK, Petrick N. Data drift in medical machine learning: implications and potential remedies. Br J Radiol. 2023;96(1150):20220878. doi: 10.1259/bjr.20220878.
6. Byrne A. Revolutionizing cancer diagnosis: the game-changing role of digital pathology and artificial intelligence. CAP Today. May 2025. Available from: https://www.captodayonline.com [Accessed 12 April 2026]
7. Advancements in Hematology. The Role of Digital Health Tools in Hematology. Available from: https://advancementsinhematology.com/technology/the-role-of-digital-health-tools-in-hematology/ [Accessed 12 April 2026]
8. Bagg A, Raess PW, Rund D, Bhattacharyya S, Wiszniewska J, Horowitz A, et al. Performance Evaluation of a Novel Artificial Intelligence-Assisted Digital Microscopy System for the Routine Analysis of Bone Marrow Aspirates. Mod Pathol. 2024;37(9):100542. doi: 10.1016/j.modpat.2024.100542