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 Concept Videos

Clinical Trials: Overview01:11

Clinical Trials: Overview

4.5K
Clinical development focuses on how the drug will interact with the human body and encompasses four key phases of clinical trials, each serving a specific purpose in assessing the safety and effectiveness of new drugs. These phases overlap and build upon one another. Phase I involves a small group of healthy volunteers (typically 20-80 individuals) or, in cases where significant toxicity is expected, patients with the targeted disease, such as cancer or AIDS. The volunteers are tested for...
4.5K
Improving Translational Accuracy02:07

Improving Translational Accuracy

3.5K
3.5K
Clinical Trials01:16

Clinical Trials

10.1K
Clinical trials are prospective experimental studies conducted on humans to determine the safety and efficacy of treatments, drugs, diet methods, and medical devices. Using statistics in clinical trials enables researchers to derive reasonable and accurate conclusions from the collected data, allowing them to make wise decisions in uncertain situations. In medical research, statistical methods are crucial for preventing errors and bias.
There are four phases in a clinical trial. A phase one...
10.1K
Types of Reports III: Telephone and Verbal Reports01:26

Types of Reports III: Telephone and Verbal Reports

972
Telephone and Verbal Reports in healthcare settings are two communication methods for conveying therapeutic instructions from healthcare providers to nurses or other healthcare staff.
Here's an overview of each type:
Telephone Orders
972

You might also read

Related Articles

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

Sort by
Same author

FLAG-Ida-venetoclax in acute myeloid leukemia: real-world efficacy and the impact of venetoclax duration on outcomes.

Leukemia & lymphoma·2026
Same author

Predictive Modeling of Clonal Hematopoiesis Across Diverse Cohorts.

Blood advances·2026
Same author

Image Quality Assessment of Diffusion-Weighted Imaging (DWI) and Its Impact on Apparent Diffusion Coefficient (ADC) as a Quantitative Imaging Biomarker for Predicting Response to Neoadjuvant Chemotherapy in High-Risk Early Breast Cancer.

Tomography (Ann Arbor, Mich.)·2026
Same author

Albuminuria Changes as a surrogate endpoint in <i>Apolipoprotein L1</i> Mediated Kidney Disease in Vanderbilt BioVU and the Million Veteran Program.

medRxiv : the preprint server for health sciences·2026
Same author

Deep Learning-Based Synthetic Contrast-Enhanced Breast MRI for Monitoring Response to Neoadjuvant Therapy.

Cancers·2026
Same author

Mutation-dependent responses to sleep and exercise in clonal haematopoiesis.

Nature·2026

Related Experiment Video

Updated: Jan 10, 2026

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts
07:50

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

16.4K

Human level information extraction from clinical reports with finetuned language models.

Longchao Liu1, Long Lian1, Yiyan Hao2

  • 1Electrical Engineering and Computer Sciences, UC Berkeley, 387 Soda Hall, Berkeley, CA, 94720, USA.

Scientific Reports
|November 25, 2025
PubMed
Summary

Open-source large language models (LLMs) can extract structured data from clinical notes with human-level accuracy using minimal resources. The Strata library and fine-tuned LLMs like Llama-3.1 demonstrate efficient, accessible clinical research database creation.

Keywords:
Deep learningInformaticsPathology

More Related Videos

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

1000
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.3K

Related Experiment Videos

Last Updated: Jan 10, 2026

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts
07:50

A Metadata Extraction Approach for Clinical Case Reports to Enable Advanced Understanding of Biomedical Concepts

Published on: September 20, 2018

16.4K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

1000
Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
05:47

Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems

Published on: June 13, 2025

1.3K

Area of Science:

  • Biomedical Informatics
  • Natural Language Processing
  • Clinical Research Data Management

Background:

  • Extracting structured data from unstructured clinical notes is a significant challenge in clinical research.
  • Existing methods often require substantial computational and annotation resources.
  • There is a need for accessible, efficient tools for clinical data extraction.

Purpose of the Study:

  • To evaluate the efficacy of open-source large language models (LLMs) for creating high-quality research databases from clinical reports with minimal resources.
  • To introduce Strata, a low-code library designed to facilitate LLM-based data extraction from clinical notes.
  • To compare the performance of various open-source LLMs against GPT-4 and human annotators.

Main Methods:

  • Development of Strata, a low-code library for LLM-driven data extraction.
  • Annotation of four diverse clinical datasets (prostate MRI, breast pathology, kidney pathology, MDS pathology) by trained researchers.
  • Evaluation of multiple open-source LLMs (instruction-tuned, medicine-specific, reasoning-based, LoRA-finetuned) using Strata.
  • Comparison of LLM performance against zero-shot GPT-4 and a second human annotator based on exact match accuracy.

Main Results:

  • LoRa-finetuned Llama-3.1 8B achieved non-inferior performance to a human annotator, with an average exact match accuracy of 90.0%.
  • Fine-tuned Llama-3.1 significantly outperformed other open-source models, including DeepSeekR1-Distill-Llama (56.8% accuracy) and Llama-3-8B-UltraMedical (39.1% accuracy).
  • GPT-4 demonstrated non-inferior performance across most datasets, while small, open-source LLMs achieved human-level accuracy using limited training data and desktop hardware.

Conclusions:

  • Small, open-source LLMs, when leveraged with Strata, provide an accessible and effective solution for curating local research databases.
  • These models offer human-level accuracy for clinical data extraction, utilizing minimal computational resources (desktop-grade hardware, <100 training reports).
  • Open-source LLMs enable local hosting and version control, offering advantages over commercial alternatives for clinical research data management.