Anatomic therapeutic chemical resolver

This advanced pipeline extracts DRUG entities from clinical texts and utilizes the sbiobert_base_cased_mli Sentence Bert Embeddings to map these entities to their corresponding Anatomic Therapeutic Chemical (ATC) codes.

Just pass in your clinical text document and get back:

• Detected NER chunk and position

• Named Entity Recognition (NER) chunk label and its confidence score

• Resolution code of the NER chunk.

• Resolution of the NER chunk.

• Cosine distance score of the resolution.

• All the other possible resolutions of the NER chunk.

• All the concept class IDs for the all resolutions.

• Codes of the resolutions

• All the cosine distance scores of the all resolutions.

Use Cases:

1. Medication Normalization for Clinical Data Pipelines

Clinical narratives often contain medication references in free text, with variations in spelling, brand names, and dosage descriptions.

The model can automatically extract drug mentions and map them to standardized ATC codes, enabling:

• Clean, interoperable medication data for downstream analytics

• Consistent drug categorization across EHR systems

• Improved accuracy in clinical decision-support systems

This significantly reduces manual coding effort and enhances data quality across clinical workflows.

2. Pharmacovigilance & Drug Utilization Analysis

Healthcare organizations and research teams need structured, comparable drug information to detect trends, measure drug usage, and identify potential safety signals.

By converting raw text into ATC-classified drug data, your model supports:

• Automated detection of drug patterns in real-world clinical documents

• Aggregation of medication usage by therapeutic class

• More accurate monitoring of adverse drug events and outcomes

This enables faster, more reliable insights for safety monitoring and population-level studies.

IMPORTANT USAGE INFORMATION: After subscribing to this product and creating a SageMaker endpoint, billing occurs on an HOURLY BASIS for as long as the endpoint is running. -Charges apply even if the endpoint is idle and not actively processing requests. -To stop charges, you MUST DELETE the endpoint in your SageMaker console. -Simply stopping requests will NOT stop billing. This ensures you are only billed for the time you actively use the service.