Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 13, 2026

Industrialized, Artificial Intelligence-guided Laser Microdissection for Microscaled Proteomic Analysis of the Tumor Microenvironment
13:01

Industrialized, Artificial Intelligence-guided Laser Microdissection for Microscaled Proteomic Analysis of the Tumor Microenvironment

Published on: June 3, 2022

Digital Registrar: A Schema-First Framework for Multi-Cancer Privacy-Preserving Pathology Abstraction via Local LLMs.

Nan-Haw Chow1,2,3, Han Chang2, Hung-Kai Chen2

  • 1Center for Precision Medicine, China Medical University Hospital, Taichung 404327, Taiwan.

Diagnostics (Basel, Switzerland)
|June 12, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

PEPRVision: a visualization framework for Pharmacoepidemiology research.

American journal of epidemiology·2026
Same author

Effectiveness of THC-containing cannabis for inflammatory bowel disease: a systematic review.

Journal of cannabis research·2026
Same author

Effectiveness of Chemotherapy With/Without Radiotherapy for Locally Advanced Esophageal Squamous Cell Carcinoma: A Population-based Target Trial Emulation Study.

Anticancer research·2026
Same author

NF‑κB‑driven LUZP1 promotes metastasis and chemoresistance in head and neck squamous cell carcinoma.

Oncology reports·2026
Same author

Incorporation of lipids improves cryo-tolerance of vitrified gorgonian coral oocytes.

PloS one·2026
Same author

Chronic liver disease is associated with greater risk of hospital readmission in patients with heart failure: a nationwide database study in the US.

European heart journal open·2026
Same journal

Correction: Luca et al. Global and Regional Diagnostic Results of Progress Toward Cervical Cancer Elimination, According to the WHO Strategy: A Systematic Literature Review with Narrative Synthesis. <i>Diagnostics</i> 2026, <i>16</i>, 1224.

Diagnostics (Basel, Switzerland)·2026
Same journal

Association Between Systemic Inflammatory Response Biomarkers and Disease Activity in Systemic Lupus Erythematosus: A Multi-Center Retrospective Study.

Diagnostics (Basel, Switzerland)·2026
Same journal

Vertebrogenic Low Back Pain and Basivertebral Nerve Ablation: A Review of Mechanisms, Imaging-Driven Selection, and Clinical Outcomes.

Diagnostics (Basel, Switzerland)·2026
Same journal

Multivalvular Carcinoid Heart Disease: The Role of Echocardiography in Diagnosis and Selection for Heterotopic Bicaval Valve Implantation.

Diagnostics (Basel, Switzerland)·2026
Same journal

Data-Efficient and Explainable Multimodal Survival Prediction in NSCLC Using Deep Image Embeddings, Clinical Variables, and Gradient-Boosted Trees.

Diagnostics (Basel, Switzerland)·2026
Same journal

Anomalous Left Coronary Artery from the Pulmonary Artery: Cinematic Volume Rendering Technique for Enhanced Anatomic Visualization.

Diagnostics (Basel, Switzerland)·2026
See all related articles
This summary is machine-generated.

A novel Large Language Model (LLM) pipeline transforms pathology reports into structured data for cancer registries. This privacy-preserving system enhances data accuracy and accessibility for research and surveillance.

Area of Science:

  • * Computational pathology
  • * Medical informatics
  • * Natural Language Processing (NLP)

Background:

  • * Free-text surgical pathology reports present challenges for automated cancer registry data entry and secondary analysis.
  • * Existing systems lack interoperability, hindering efficient data utilization.
  • * Need for a standardized, machine-readable format for pathology data.

Purpose of the Study:

  • * To develop and validate a Large Language Model (LLM) pipeline for extracting structured data from surgical pathology reports.
  • * To create a clinically governed schema layer for interoperability and registry-grade extraction.
  • * To assess the feasibility of on-premises deployment for privacy-preserving data handling.

Main Methods:

  • * Development of a College of American Pathologists (CAP)-aligned clinical ontology for 10 cancer types and 192 fields.
Keywords:
CAP protocolDSPycancer registryclinical ontologyinteroperabilitylarge language modelprivacy-preserving AIstructured data extraction

More Related Videos

A Multimodal Imaging Framework to Advance Phenotyping of Living Label-free Breast Cancer Cells
10:37

A Multimodal Imaging Framework to Advance Phenotyping of Living Label-free Breast Cancer Cells

Published on: August 22, 2025

Related Experiment Videos

Last Updated: Jun 13, 2026

Industrialized, Artificial Intelligence-guided Laser Microdissection for Microscaled Proteomic Analysis of the Tumor Microenvironment
13:01

Industrialized, Artificial Intelligence-guided Laser Microdissection for Microscaled Proteomic Analysis of the Tumor Microenvironment

Published on: June 3, 2022

A Multimodal Imaging Framework to Advance Phenotyping of Living Label-free Breast Cancer Cells
10:37

A Multimodal Imaging Framework to Advance Phenotyping of Living Label-free Breast Cancer Cells

Published on: August 22, 2025

  • * Implementation of a model-agnostic LLM pipeline using Declarative Self-improving Python (DSPy) with grammar-constrained decoding.
  • * Benchmarking on 893 internal reports and external validation using 242 The Cancer Genome Atlas (TCGA) reports.
  • * Hardware feasibility confirmed on a single 48 GB Graphics Processing Unit (GPU).

    • Main Results:

    • * Achieved 92.0% macro-mean exact-match accuracy on internal data using the gpt-oss-20b model.
    • * High fidelity for critical indicators: breast ER/PR (98.7%) and margin positivity (>93%).
    • * External validation on TCGA data yielded 77.5% accuracy, improving to 88.0% after refinement.
    • * Processing time ranged from 40-70 seconds per report, balancing speed and accuracy.

    Conclusions:

    • * A schema-first abstraction layer effectively decouples clinical logic from specific AI models.
    • * Reliable transformation of narrative reports into machine-readable structures is achievable.
    • * Establishes a portable, privacy-preserving foundation for automated cancer surveillance and data reuse.