Seamless Integration of Artificial Intelligence in Pathological Diagnostics

Artificial intelligence (AI) technologies are increasingly transforming medical diagnostics, yet their full potential remains underutilized across many areas of healthcare. A collaborative effort between Friedrich–Alexander University Erlangen–Nürnberg (FAU) at Universitätsklinikum Erlangen (UKER) and Gravina Hospital in Caltagirone, Italy, is pioneering a process that seamlessly embeds AI into routine pathological workflow.

Every year, over 1.4 million patients in Germany undergo hospital treatment for cancer. During surgical tumor removal, pathological assessment is critical for identifying the cancer type, determining malignancy, and guiding treatment plans, including chemotherapy options. Traditionally, such assessments rely on microscopes to examine tissue samples, but digitization of these samples is still not widespread.

The project focuses on fully digitized pathology departments, where tissue samples are prepared, stained, and then scanned to produce high-resolution digital images accessible through digital systems. This setup paves the way for AI to assist pathologists by highlighting areas of interest, such as malignant regions, enhancing diagnostic precision.

The innovative workflow developed by the researchers involves automatically analyzing these digitized slides. When samples are uploaded, the system transfers the data to a server that runs various AI models tailored to different tissue types and staining methods. Pathologists can also request specific analyses directly from their laboratory information system (LIS). The results, often visualized as heatmaps, overlay critical insights onto the digital images, helping clinicians identify malignant zones more easily.

This approach aims not only to improve diagnostic accuracy but also to facilitate the validation of deep learning models in clinical settings. Continued collaboration and development are expected to refine these algorithms further, promoting their routine use in pathology and potentially transforming diagnostic standards.

By integrating AI efficiently into pathology workflows, this initiative demonstrates promising progress toward smarter, faster, and more reliable cancer diagnosis, ultimately benefiting patient care.

Source: https://medicalxpress.com/news/2025-06-ai-seamlessly-pathological-diagnosis.html