AI in digital pathology: facilitating research and monetizing assets

Understanding digital pathology

Advances in digital pathology can be thrilling, promising significant benefits to medical research and collaboration. The advent of digital pathology delivers an image-driven environment where pathologists can better understand medical events, train students, and contribute to research.

Whole slide images (WSIs) serve as valuable assets— information that can be acquired, managed, shared, and interpreted. Globally, the digital pathology market is expected to increase at a compound annual growth rate of 8% through 2030. With this growth comes many benefits. Pathologists anticipate a faster, more precise understanding of diseases affecting individuals and global populations.

Technology solutions, especially those designed to boost efficiency, can serve a dual purpose. Digital pathology platforms will enable effective collaboration across time zones and borders—bringing experts together quickly. Access to whole slide images will reduce time spent waiting on physical slides to support pathology work anytime, anywhere.

Enhanced with AI, digital pathology systems will speed up each step, from image scans to tissue analysis. And access to digital pathology platforms will support comprehensive education and training using rich data sets and annotations from a variety of collaborators.

Already, digital pathology is easing collaboration among pathologists regardless of their physical locations. Digital samples are readily accessible when a specialist or second review is needed. Across geographic regions, pathologists can reach underserved populations and improve diagnostic confidence. These capabilities can positively impact resource shortages and enrich education, especially in research settings and in developing nations.

With digital pathology, the pathologist can take advantage of AI and other technology tools to perform critical measurements, image and historical analyses, and pattern recognition. Research integration with AI machine learning shows significant potential in detecting disease patterns and predicting outcomes. These technologies also reduce cost and risk for providers, labs, and other healthcare organizations via offsite storage and cloud capabilities. Workflow automation and AI algorithms enable substantial time savings as well.

What about generative AI?

Just as generative AI tools—such as large language models and multi-modal models—are impacting other professions and disciplines, so will they affect pathology. Specialized AI and machine learning systems have already matched, and in some cases surpassed, human abilities in areas such as image analysis. However, these AI models require adequate training and massive volumes of accurate data.

Now, a more flexible kind of model for generative AI is gaining traction. Foundation models are big AI systems trained on huge data stores that can be fine-tuned for specific tasks. These models learn from mistakes, require less additional data, and can adjust based on feedback.

A pathology-specific generative AI, based on foundation models, may serve as an expert companion to pathologists. It could add greater efficiency and objectivity to routine laboratory tasks such as quantifying image analysis, or generating reports, diagnosis, and prognosis. Foundation models and generative AI could also help standardize the pathology laboratory workflow, education and training.

Exploring the rise of AI in digital pathology

AI stands to transform how pathologists analyze WSIs, as well as how they approach research, teaching at conferences and virtual workshops and presentations before tumor boards.

Supporting research and education

AI tools enhance the training of future pathologists by offering easily accessible digital slides that can be shared and viewed remotely. Such technologies might include features like automated quizzes to test trainee knowledge or the ability to create synthetic digital slides for educational purposes. In fact, the College of American Pathologists now provides WSIs alongside traditional glass slides for certain proficiency tests. For research, high-speed, high-resolution digital slides can be used in remote consultations and for assessing the consistency of diagnoses among pathologists. The availability of massive volumes of data and AI analysis tools allow disease researchers to grow their knowledge base and find deeper insights.

Maintaining quality standards

It’s challenging for pathologists and radiologists to stay updated on all medical conditions. Regular interactions and feedback, whether through manual review or AI tools, can help researchers and practitioners refine their skills and keep current with new diagnostic methods. AI can assist in routine quality control processes or as part of formal assessments in pathology labs. It also acts as a safety measure by double-checking pathologist assessments using automated algorithms, ultimately improving accuracy.

Expanding research horizons with digital pathology and AI

Advances in technology have excited the pathology discipline. Pathologists are talking about uncovering new biomarkers and disease associations and enabling large- scale population studies. Such excitement and a growing emphasis on precision medicine are driving further development of AI-infused digital pathology methods. emphasis on precision medicine are driving further development of AI-infused digital pathology methods.

One buzzword, translational research, is gaining a lot of attention and is predicted to attract a fair amount of funding in the next several years. Translational research centers on turning research findings into practical treatments for patients. Digital pathology and AI applications can support this research model by improving quantitative accuracy and lending geographic context to data via spatial algorithms.

Translational research, which moves discoveries from the lab to the clinic and back again, is becoming more important in medical research. However, it’s a relatively new area that warrants proper discussion about ethical implications.