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

Extracorporeal Removal of Drugs: Continuous Renal Replacement Therapy01:26

Extracorporeal Removal of Drugs: Continuous Renal Replacement Therapy

Continuous Renal Replacement Therapy (CRRT) is an essential intervention for patients experiencing severe kidney dysfunction. This therapy offers a continuous mechanism for removing fluids and toxins from the bloodstream, leveraging the patient’s blood pressure to facilitate filtration through a specialized filter. This method contrasts with intermittent dialysis, providing a gentler and more consistent removal of waste products and excess fluid, which is particularly beneficial in critically...
Continuous Renal Replacement Therapy01:30

Continuous Renal Replacement Therapy

Continuous Renal Replacement Therapy, also known as CRRT, is a procedural treatment for acute kidney injury (AKI) that gradually removes uremic toxins and fluids while maintaining acid-base balance and stabilizing electrolytes. It is particularly useful for hemodynamically unstable patients. Unlike intermittent hemodialysis, which is faster, CRRT provides a gentler approach over 24 hours, closely mimicking the function of natural kidneys. However, CRRT is not ideal for patients with...
Extracorporeal Removal of Drugs: Hemoperfusion and Hemofiltration01:25

Extracorporeal Removal of Drugs: Hemoperfusion and Hemofiltration

Hemoperfusion and hemofiltration are critical techniques in medical treatments to eliminate accumulated drugs, metabolites, and electrolytes from the bloodstream. These methods are particularly vital in cases of accidental poisoning and drug overdose.Hemoperfusion involves passing blood through an adsorbent material to remove unwanted substances. The main adsorbents used in hemoperfusion include activated charcoal and Amberlite resins. Activated charcoal can adsorb both polar and nonpolar...

You might also read

Related Articles

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

Sort by
Same author

Correction to: Whole Genome Sequencing in 25 Families with Suspected Inborn Errors of Immunity: Diagnostic Yield and Clinical Relevance of Genome-wide Analysis.

Journal of clinical immunology·2026
Same author

Natural Killer Cell Assays: Clinical Applications and Future Directions.

Annals of laboratory medicine·2026
Same author

Effective resolution of daratumumab-induced false-positive antibody screening with commercial reagents.

Vox sanguinis·2026
Same author

Complementary Roles of Next-Generation Flow and Next-Generation Sequencing for Measurable Residual Disease Assessment in Multiple Myeloma.

Journal of Korean medical science·2026
Same author

Microbiome and metabolite biomarkers of CAR T-cell therapy outcomes in relapsed/refractory diffuse large B-cell lymphoma.

Blood advances·2025
Same author

Integrated network propagation identifies prognostic metabolic signatures in acute myeloid leukemia.

Journal of translational medicine·2025

Related Experiment Video

Updated: May 8, 2026

Generation of Multivirus-specific T Cells to Prevent/treat Viral Infections after Allogeneic Hematopoietic Stem Cell Transplant
08:52

Generation of Multivirus-specific T Cells to Prevent/treat Viral Infections after Allogeneic Hematopoietic Stem Cell Transplant

Published on: May 27, 2011

An automatic consult reply system for therapeutic plasma exchange using retrieval-augmented generation.

Jong Kwon Lee1,2, Sooin Choi3, Sholhui Park4

  • 1Department of Laboratory Medicine, Hallym University Sacred Heart Hospital, Hallym University College of Medicine, Anyang, Republic of Korea.

Vox Sanguinis
|May 6, 2026
PubMed
Summary
This summary is machine-generated.

Retrieval-augmented generation (RAG) significantly improves clinical decision support for therapeutic plasma exchange (TPE) by grounding large language models (LLMs) in guidelines. This RAG-TPE framework enhances accuracy and reliability compared to conventional LLMs.

Keywords:
clinical decision supportlarge language modelretrieval‐augmented generationtherapeutic plasma exchangetransfusion medicine

More Related Videos

Automated Cell Enrichment of Cytomegalovirus-specific T cells for Clinical Applications using the Cytokine-capture System
10:24

Automated Cell Enrichment of Cytomegalovirus-specific T cells for Clinical Applications using the Cytokine-capture System

Published on: October 5, 2015

Platelet-Rich Plasma Lysate for Treatment of Eye Surface Diseases
05:57

Platelet-Rich Plasma Lysate for Treatment of Eye Surface Diseases

Published on: August 2, 2022

Related Experiment Videos

Last Updated: May 8, 2026

Generation of Multivirus-specific T Cells to Prevent/treat Viral Infections after Allogeneic Hematopoietic Stem Cell Transplant
08:52

Generation of Multivirus-specific T Cells to Prevent/treat Viral Infections after Allogeneic Hematopoietic Stem Cell Transplant

Published on: May 27, 2011

Automated Cell Enrichment of Cytomegalovirus-specific T cells for Clinical Applications using the Cytokine-capture System
10:24

Automated Cell Enrichment of Cytomegalovirus-specific T cells for Clinical Applications using the Cytokine-capture System

Published on: October 5, 2015

Platelet-Rich Plasma Lysate for Treatment of Eye Surface Diseases
05:57

Platelet-Rich Plasma Lysate for Treatment of Eye Surface Diseases

Published on: August 2, 2022

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Medicine
  • Transfusion Medicine

Background:

  • Large language models (LLMs) offer potential for clinical decision support but are prone to factual inaccuracies.
  • Retrieval-augmented generation (RAG) enhances LLM reliability by integrating authoritative knowledge.
  • Therapeutic plasma exchange (TPE) necessitates precise, guideline-adherent decision-making based on current recommendations.

Purpose of the Study:

  • To assess if a RAG-based framework improves the accuracy, reliability, and standardization of TPE decision support compared to standard LLMs.
  • To evaluate the performance of a hybrid RAG pipeline incorporating multiple retrieval methods and structured prompts.
  • To quantify improvements in specific TPE decision elements, including diagnosis, ASFA category, insurance applicability, and plasma volume calculations.

Main Methods:

  • A hybrid RAG pipeline was developed using BAAI/bge-base-en-v1.5 embeddings, Chroma, and BM25.
  • Structured prompts integrated ASFA guidelines, HIRA criteria, and plasma volume computation rules.
  • Thirty real-world TPE cases were queried across six RAG and three non-RAG LLM configurations, with each case answered five times for comprehensive evaluation.

Main Results:

  • RAG configurations consistently surpassed non-RAG models in accuracy across all evaluated elements.
  • Significant improvements were observed in plasma-volume calculation and ASFA classification accuracy with RAG.
  • The RAG GPT-4.1-mini configuration demonstrated optimal performance, balancing high accuracy with low response latency and enhanced reproducibility.

Conclusions:

  • A guideline-grounded RAG approach substantially improves the accuracy, stability, and standardization of TPE consultations.
  • The RAG-TPE framework proves the feasibility of reliable, clinically focused decision support in transfusion medicine.
  • Further prospective clinical validation of this RAG-TPE system is warranted for integration into healthcare workflows.