John Snow Labs
- Website
- johnsnowlabs.com · all feeds from this site
- Feed URL
https://www.johnsnowlabs.com/feed/
Read it in your FeedListing reader
Or add to: Feedly Inoreader Raw feed
Latest posts
-
De-identification Doesn’t End at Deployment: How to Know What Is Safe to Release
Every de-identification program reaches the same moment. A new batch arrives — a million notes from a new hospital, a new note type, a year of fresh records. Someone asks:... The post De-identification Doesn’t End at Deployment: How to Know What Is Safe to Release appeared first…
-
Why Mapping Clinical Notes to ICD-10, CPT, LOINC, and SNOMED Codes Takes More than Entity Recognition
Automated ICD-10, CPT, LOINC, and SNOMED mapping breaks on negation, timing, and family context. See the technical framework and production evidence behind it. The post Why Mapping Clinical Notes to ICD-10, CPT, LOINC, and SNOMED Codes Takes More than Entity Recognition appeared…
-
Clinical Knowledge Graphs in an LLM-Dominated World
LLMs write fluent clinical text but don't reliably normalize or verify it. See the evidence for why knowledge graphs remain essential to clinical AI. The post Clinical Knowledge Graphs in an LLM-Dominated World appeared first on John Snow Labs .
-
dQMs and ECDS: Why Real-Time NLP Pipelines Must Replace Manual Chart Abstraction
HEDIS isn't being retired, but its hybrid data-collection method is. See what digital quality measures and ECDS reporting require from clinical data infrastructure through 2030. The post dQMs and ECDS: Why Real-Time NLP Pipelines Must Replace Manual Chart Abstraction appeared…
-
MLOps for Clinical NLP: What Scaling from Prototype to Production Actually Requires
Clinical NLP prototypes that score well in validation often stall in production. See what infrastructure, governance, and MLOps investment actually close that gap. The post MLOps for Clinical NLP: What Scaling from Prototype to Production Actually Requires appeared first on John…
-
John Snow Labs Announces 2026 Applied AI Summit Program Focused on Proving AI Works
John Snow Labs Announces 2026 Applied AI Summit Program Focused on Proving AI Works More than 50 free sessions will explore how practitioners test, validate, and monitor AI systems, including three open source contributions to the MedHELM project The post John Snow Labs…
-
Why General-Purpose NLP Frameworks Fail
Benchmark accuracy doesn’t predict production success. See the throughput, cost, and audit data that separate purpose-built healthcare NLP from general-purpose frameworks. The post Why General-Purpose NLP Frameworks Fail appeared first on John Snow Labs .
-
Real-Time Evidence Extraction for Risk Adjustment: What the NLP Infrastructure Requires
Risk adjustment models miss complexity that notes document but codes don't capture. See what real-time NLP extraction infrastructure requires. The post Real-Time Evidence Extraction for Risk Adjustment: What the NLP Infrastructure Requires appeared first on John Snow Labs .
-
The Data Engineering of RAG: Why Retrieval Quality Depends on What Happens Before Retrieval
RAG quality is decided before a query ever runs. See why chunking, terminology normalization, and de-identification determine whether clinical RAG retrieval is reliable. The post The Data Engineering of RAG: Why Retrieval Quality Depends on What Happens Before Retrieval appeared…
-
John Snow Labs and Data4Healthcare Announce Partnership to Bring Medical Language Models to Health Plan Risk Adjustment and Quality Programs
John Snow Labs and Data4Healthcare partner to bring medical language models and Patient Journey Intelligence to health plan risk adjustment, quality, and care programs. The post John Snow Labs and Data4Healthcare Announce Partnership to Bring Medical Language Models to Health…
-
AI and Personalized Medicine
Personalized treatment pathways require data engineering more than better models. See what Roche, MiBA, and Guideline Central built to get there in production. The post AI and Personalized Medicine appeared first on John Snow Labs .
-
De-identifying Pathology Slides at Scale: A Serverless AWS Batch Pipeline for SVS Whole-Slide Images
De-identify SVS pathology slides at scale with AWS Batch and John Snow Labs Visual NLP. Learn how to build an automated, HIPAA-ready PHI redaction pipeline. The post De-identifying Pathology Slides at Scale: A Serverless AWS Batch Pipeline for SVS Whole-Slide Images appeared…
-
Closing HEDIS and Stars Quality Gaps: A Step-By-Step Evidence-Extraction Blueprint
A step-by-step blueprint for automating HEDIS and Star Ratings evidence extraction from clinical notes, grounded in production deployments and benchmarks. The post Closing HEDIS and Stars Quality Gaps: A Step-By-Step Evidence-Extraction Blueprint appeared first on John Snow Labs…
-
Multilingual Clinical NLP
English-only clinical NLP creates a two-tier standard of AI-assisted care. Evidence on what equitable, regulatory-grade multilingual deployment actually requires. The post Multilingual Clinical NLP appeared first on John Snow Labs .
-
High-Accuracy Clinical Term Mapping to Standard Medical Terminologies (ICD-10, RxNorm, SNOMED, and 90+ vocabularies) with John Snow Labs’ Medical Language Models
Map clinical terms to ICD-10, RxNorm, SNOMED CT, and 90+ medical vocabularies with high-accuracy entity resolution models, contextual embeddings, and Terminology Server. The post High-Accuracy Clinical Term Mapping to Standard Medical Terminologies (ICD-10, RxNorm, SNOMED, and…
-
Beyond Masking: How Expert Determination Unlocks More Value from Your De-Identified Data
A guide for healthcare data leaders navigating the gap between de-identification and provable compliance. The post Beyond Masking: How Expert Determination Unlocks More Value from Your De-Identified Data appeared first on John Snow Labs .
-
Benchmarking Databricks ai_mask on clinical de-identification: 0.71 PHI F1
Benchmark Databricks ai_mask() for clinical de-identification: 0.71 PHI F1 vs. 0.96 for John Snow Labs Healthcare NLP, with key accuracy and compliance gaps The post Benchmarking Databricks ai_mask on clinical de-identification: 0.71 PHI F1 appeared first on John Snow Labs .
-
Clinical de-identification benchmarks 2026: John Snow Labs against OpenAI, Databricks, Presidio, and LLM APIs
Compare 2026 clinical de-identification benchmarks across John Snow Labs, OpenAI, Claude, Gemini, Databricks, Presidio, and healthcare APIs. The post Clinical de-identification benchmarks 2026: John Snow Labs against OpenAI, Databricks, Presidio, and LLM APIs appeared first on…
-
Generate DICOM de-identification pipelines with the Visual NLP DICOM Skill
Generate grounded DICOM de-identification pipelines with the Visual NLP DICOM Skill, supporting metadata, pixel, free-text, PDF, and zero-shot PHI de-identification workflows. The post Generate DICOM de-identification pipelines with the Visual NLP DICOM Skill appeared first on…
-
PII coverage is not HIPAA or GDPR coverage: what clinical de-identification requires
PII detection and clinical de-identification aren't the same task. See why HIPAA Safe Harbor's 18 identifier categories and GDPR Article 9 require entity coverage that Presidio, AWS, and Google DLP don't ship out of the box. The post PII coverage is not HIPAA or GDPR coverage:…