Over the last years, we have been witnessing rapid technological advances in the field of artificial intelligence (AI), which led to the development of machine learning (ML) models capable of solving complex computer vision problems, including the automated analysis of digital microscopy images in pathology. However, despite promising results in research settings, the application of AI in clinical practice remains largely unrealized.
A major impediment for clinical deployment of ML models is the lack of interoperability between ML systems and health information technology systems. This webinar will provide insight into the importance of standards for realizing AI and will introduce a standard-based framework for computational pathology that facilitates programmatic access to digital images for analysis by ML models and integration of image analysis results into clinical workflows for interpretation and evaluation by pathologists.
- Become familiar with ML components and workflows for model development and deployment.
- Appreciate the importance of data standardization and interoperability for ML.
- Recognize differences between research and clinical work streams and gain an understanding of necessary steps for translating ML models from research tools into clinical services.
Markus D. Herrmann, M.D., Ph.D.