For decades, AI has been a goal pursued by researchers and developers across many industries. While tangible benefits were limited for a long time, the situation has changed significantly in recent years. More powerful computing resources, the availability of large data volumes, and close collaboration between computer scientists, medical professionals, and hospital management have contributed to AI being applied more broadly and delivering practical value in healthcare contexts.

The term AI is interpreted in different ways. A common and practical approach is to distinguish between three technological levels.

1. Pattern recognition is already in use, particularly in medical imaging. In this context, success means that clinical studies show detection rates comparable to established reference standards. Pattern recognition is based on algorithms—predefined computational procedures—and large volumes of quality-assured data. This approach reflects expert knowledge that has been encoded in advance.
    
2. The next level is machine learning. Here, an algorithm learns to perform a task autonomously through repeated exposure to data. The learning process is guided by predefined quality criteria and the informational content of the data. Unlike conventional algorithms, machine learning does not require an explicitly modeled solution path.
    
3. The third level is inspired by the interconnected neurons of the human brain: neural networks. This artificial analogy consists of multiple layers of data nodes connected by weighted links. Learning algorithms within neural networks are continuously trained with new data. Deep neural networks can comprise a large number of layers, enabling the application of deep learning methods to complex problem spaces.

Artificial intelligence (AI) is increasingly used in healthcare to support data-driven processes, streamline workflows, and enable new forms of analysis. When applied appropriately, AI can support medical professionals in managing information and administrative tasks, while providing structured insights derived from large volumes of data.

• Automated structuring: Many medical documents are unstructured and therefore difficult to use without significant manual effort. AI can extract and structure health data and make it available for clinical workflows and documentation processes.
   
• Data analysis at scale: AI-based systems can process large volumes of health and imaging data and identify patterns relevant for further review.
   
• Individualized data views: By analyzing patient-specific data, AI can support the compilation of structured, patient-centered information views for clinical use.
   
• Administrative efficiency: AI can support administrative processes such as documentation and scheduling by automating routine tasks.
   
• Predictive analysis: Data-driven methods can be used to analyze trends and patterns in health data for planning and monitoring purposes.
   
• Medical research: AI supports research activities by enabling the analysis of large and heterogeneous datasets.

A wide range of AI tools are now used in healthcare, including applications for initial information exchange or the analysis of medical images. Some solutions focus specifically on structuring health data and making it usable across systems. One such solution is the AI platform Health Discovery.

With Health Discovery, artificial intelligence can be integrated into existing systems and applications. The platform structures health data in real time and processes a wide range of unstructured data formats, including PDFs, Word documents, free-text database entries, and speech-based input. The software extracts more than 50 medical entities—such as diagnoses, medications, laboratory results, and vital parameters—and translates them into established standards including FHIR, SNOMED CT, ICD-10, and LOINC. The platform provides 150 pre-trained AI models and supports more than 15 terminologies.