Imagine having the ability to uncover latent insights on patient information, Azure Cognitive search helps you create searchable content on patient notes, lab results or other medical information, using the power of integrated AI to surface the most relevant search results based on search intent rather than just keywords.
You can find about the architectural design and a step by step guide to implement this solution in this GitHub repository.
Whether the search is geared to individual patients or to multiple patients for research purposes, you can take advantage of Azure AI Cognitive services, in particular Azure Text Analytics for Healthcare features that include named entity recognition, relation extraction, entity linking, and assertion detection.
Named entity recognition is used to perform a semantic extraction of words and phrases mentioned from unstructured text that are associated with any of the supported entity types, such as diagnosis, medication name, symptom/sign, or age.
Relation extraction is used to identify meaningful connections between concepts mentioned in text that are associated with any of the supported relations, such as the “time of condition” relation, which connects a condition name with a time.
Entity linking is used to disambiguate the extracted entities by associating them with preferred names and codes from the biomedical vocabularies supported by the Unified Medical Language System (UMLS) Metathesaurus.
Assertion detection is used to preserve the meaning of medical content by adding contextual modifiers to the extracted entities using these categories:
- Certainty
- Conditionality
- Association
Check the Doctor Notes Search Application implemented using fake doctor notes to see the power of Azure Cognitive Search and Azure Text Analytics for Healthcare together to help you find insights on patient information.
Keeping in mind patient privacy regulation and security you can choose to mask PII information from unauthorized users implementing user authentication.
Serendipituos Code 