Denver Health

Ricky Lopez

Artificial intelligence (AI) has become a popular technology, causing digital pathologists to consider how they may employ it in their daily practice.

Digital pathology, also known as computational pathology, is an emerging subject and sub-specialty based on artificial intelligence (AI) that claims to improve the accuracy and availability of high-quality healthcare to patients in various medical fields. While AI-based Digital Pathology has a bright future, the shift from the current state to that future one poses both obstacles and opportunities.

The State of AI in Digital Pathology Currently

Standardization across formats, protocols, workflows and systemic quality control is currently lacking in AI-based Digital Pathology. This causes challenges with interoperability, raises the risk of error, and can even stymie regulatory progress. Automated AI systems can combine massive amounts of complicated data, but most hospitals' outdated Laboratory Information Systems and Picture Archiving and Communication Systems lack the technological ability to process AI data. Because the regulatory path for AI-based Digital Pathology is still being established, and ethics are still being examined, policies and regulations are unclear.

The Future of Artificial Intelligence-based Digital Pathology

Although numerous technological and ethical issues must be resolved, allowing AI-based Digital Pathology to function as a synergistic system would improve workflows and allow clinical teams to share and analyze picture data across a larger platform. To shift to the future state of AI-based Digital Pathology organizations must:

Make it clear that artificial intelligence (AI) is a medical product, not merely an algorithm. This distinction is important when it comes to the decisions that must be made in order to implement a legal framework and how we think about AI in this context.

Address the issue of health disparities and make sure that AI models and algorithms take the entire population into account. AI-based Digital Pathology should benefit all patients, regardless of their location or socioeconomic level. This technology has the ability to improve insight access and distribution.

Share whatever is learned about implementing and supporting AI in digital pathology. Information silos obstruct the sharing of information, standardization of best practices, and consensus needed to drive adoption.

Through consistent risk assessment, businesses may help speed up regulatory certification of AI. Having a standardized approach for analyzing the risks of AI-based digital pathology would help the products or treatments go to market more quickly and enhance patient outcomes.