Demos
Automate medical chart reviews to summarize and answer key questions in:
- Validation of Clinical Codes
- Prior Authorization
- Accelerated Underwriting
- Extract Clinical Risk Factors
- Report Social Determinants of Health
See Payer-specific Models in Action
Risk Factor Extraction
Automatically identify risk factors such as Coronary artery disease, Diabetes, Family history, Hyperlipidemia, Hypertension, Medications, Obesity, PHI, Smoking habits in clinical documents using pretrained Medical Language Models.
Entity Extraction for Prior Authorization
Automatically identify and extract entities for downstream prior authorization processes.
Social Determinants of Health
Extract terminology on Social Determinants of Health to reduce emergency visits, identify high-risk members, improve satisfaction, and optimize care coordination.
- Prevent costly ER visits and readmissions
- Identify high-risk members early for effective prevention
- Improve satisfaction and participation
- Optimize resources and care
Proven Success across
RAG on FHIR: Using FHIR with Generative AI to Make Healthcare Less Opaque
Identifying Housing Insecurity and Other Social Determinants of Health from Free-Text Notes
Accelerating clinical risk adjustment through Natural Language Processing
Accelerated Underwriting & Clinical Chart Review
Resolve Clinical Entities to Terminology codes automatically using our pre-trained models
Map healthcare codes between taxonomies
These pretrained pipelines map various codes (e.g., ICD10CM codes to SNOMED codes) without using any text data. (...)
Map laboratory terminology to LOINC taxonomy
This demo shows how laboratory terminology can be automatically mapped to the Logical Observation Identifiers Names and Codes (LOINC) taxonomy.(...)
Map healthcare codes between taxonomies
These pretrained pipelines map various codes (e.g., ICD10CM codes to SNOMED codes) without using any text data. (...)
Map laboratory terminology to LOINC taxonomy
This demo shows how laboratory terminology can be automatically mapped to the Logical Observation Identifiers Names and Codes (LOINC) taxonomy.(...)
Resolve Clinical Health Information using the HPO taxonomy
This demo shows how clinical health information can be mapped to codes using the HPO taxonomy. (...)
Resolve Clinical Health Information using the MeSH taxonomy
This demo shows how clinical health information can be mapped to codes using the MeSH taxonomy.(...)
Resolve Clinical Findings using the UMLS CUI taxonomy
This demo shows how clinical findings can be mapped to codes using the UMLS CUI taxonomy.(...)
Resolve Clinical Health Information using the NDC taxonomy
This demo shows how clinical health information and concepts (like drugs/ingredients) can be mapped to codes using the NDC taxonomy.(...)
Resolve Clinical Procedures using CPT taxonomy
This demo shows how to map clinical procedures to codes using CPT taxonomy.(...)
Mapping ICD-10-CM Codes with Corresponding Medicare Severity-Diagnosis Related Group
This pretrained model maps ICD-10-CM codes with their corresponding Medicare Severity-Diagnosis Related Group (MS-DRG).(...)
Mapping RxNorm Codes with Corresponding Drug Brand Names, Actions and Treatments
Mapping RxNorm and RxNorm Extension Codes with Corresponding Drug Brand Names, Actions and Treatments.(...)
Extraction of SNOMED Terms
This NER model is designed to identify SNOMED terms within clinical documents, utilizing the embeddings_clinical embeddings model.(...)
Frequently Asked Questions
Yes, we do offer professional services where our Healthcare NLP Data Scientists and Subject Matter Experts can come in and build an end-to-end solution under a Statement of Work.
You may fine-tune these models completely within your environment and no data will be shared with John Snow Labs. You will retain the IP for the fine-tuning but for the base models to continue running you will require a license.
These models are infrastructure agnostic, you may run them on any cloud VM, Databricks, Snowflake, or even on your local machine.
