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

Targeted Cancer Therapies02:57

Targeted Cancer Therapies

8.6K
The targeted cancer therapies, also known as “molecular targeted therapies,” take advantage of the molecular and genetic differences between the cancer cells and the normal cells. It needs a thorough understanding of the cancer cells to develop drugs that can target specific molecular aspects that drive the growth, progression, and spread of cancer cells without affecting the growth and survival of other normal cells in the body.
There are several types of targeted therapies against...
8.6K
Cancer Survival Analysis01:21

Cancer Survival Analysis

650
Cancer survival analysis focuses on quantifying and interpreting the time from a key starting point, such as diagnosis or the initiation of treatment, to a specific endpoint, such as remission or death. This analysis provides critical insights into treatment effectiveness and factors that influence patient outcomes, helping to shape clinical decisions and guide prognostic evaluations. A cornerstone of oncology research, survival analysis tackles the challenges of skewed, non-normally...
650

You might also read

Related Articles

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

Sort by
Same author

Correction: RAPID-LC: rapid evidence-to-practice uptake of large core thrombectomy across a stroke consortium.

Journal of neurology·2026
Same author

Image-Guided Thermal Ablation of Stage 1 Single and Multiple Primary Lung Carcinoma: Five-Year Outcomes.

Medical sciences (Basel, Switzerland)·2026
Same author

Toward universal dose prediction: A multi-scale, multi-objective framework for sequential boost radiotherapy.

Medical physics·2026
Same author

Toward Noninvasive Magnetic Resonance-Based Target Delineation for Ventricular Tachycardia Stereotactic Radiation Therapy: Development of a 3-Dimensional Wideband Late Gadolinium-Enhanced Magnetic Resonance Imaging Technique.

International journal of radiation oncology, biology, physics·2026
Same author

RAPID-LC: rapid evidence-to-practice uptake of large core thrombectomy across a stroke consortium.

Journal of neurology·2026
Same author

Reply: Why fan therapy does not yet demonstrate enhanced training benefits in people with chronic respiratory disease.

The European respiratory journal·2026

Related Experiment Video

Updated: Jan 17, 2026

Endobronchial Ultrasound-guided Intratumoral Injection of Cisplatin for the Treatment of Isolated Mediastinal Recurrence of Lung Cancer
04:04

Endobronchial Ultrasound-guided Intratumoral Injection of Cisplatin for the Treatment of Isolated Mediastinal Recurrence of Lung Cancer

Published on: February 12, 2017

10.9K

Revisiting LLMs and Lung Cancer Questions: How AI Responds to Common Lung Cancer Questions Two Years Later.

Dan Nguyen1, Amir A Rahsepar2, Hamid Chalian3

  • 1University of Massachusetts Chan Medical School, Worcester, MA (D.N.).

Academic Radiology
|September 17, 2025
PubMed
Summary
This summary is machine-generated.

GPT-o3-mini and Gemini show improved accuracy in answering lung cancer questions, but expert oversight remains crucial due to inconsistent performance. This study evaluated five large language models (LLMs) on their reliability.

Keywords:
Artificial IntelligenceChatGPTLLMsLung cancerRadiology

More Related Videos

Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer
07:59

Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer

Published on: September 8, 2023

1.6K
A Minimally Invasive Method for Generating a Syngeneic Orthotopic Mouse Model of Lung Cancer
07:46

A Minimally Invasive Method for Generating a Syngeneic Orthotopic Mouse Model of Lung Cancer

Published on: August 19, 2025

1.1K

Related Experiment Videos

Last Updated: Jan 17, 2026

Endobronchial Ultrasound-guided Intratumoral Injection of Cisplatin for the Treatment of Isolated Mediastinal Recurrence of Lung Cancer
04:04

Endobronchial Ultrasound-guided Intratumoral Injection of Cisplatin for the Treatment of Isolated Mediastinal Recurrence of Lung Cancer

Published on: February 12, 2017

10.9K
Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer
07:59

Author Spotlight: Advancements in Molecular Biomarker Testing for Non-Squamous Non-Small Cell Lung Cancer

Published on: September 8, 2023

1.6K
A Minimally Invasive Method for Generating a Syngeneic Orthotopic Mouse Model of Lung Cancer
07:46

A Minimally Invasive Method for Generating a Syngeneic Orthotopic Mouse Model of Lung Cancer

Published on: August 19, 2025

1.1K

Area of Science:

  • Artificial Intelligence in Medicine
  • Medical Imaging and Diagnostics
  • Natural Language Processing in Healthcare

Background:

  • Large language models (LLMs) are increasingly used in healthcare, necessitating evaluation of their accuracy and reliability.
  • Prior assessments in 2023 highlighted variability in LLM performance for medical queries.
  • Standardized guidelines like Lung-RADS and Fleischner Society guidelines provide a basis for objective medical question development.

Purpose of the Study:

  • To compare the accuracy, consistency, and reliability of five leading LLMs (GPT-o3-mini, Gemini, DeepSeek R1, Claude, Perplexity) in answering standardized lung cancer questions.
  • To build upon previous evaluations of LLM performance in medical contexts.
  • To assess the suitability of current LLMs for clinical decision support in radiology.

Main Methods:

  • Development of 40 standardized lung cancer-related questions by experienced radiologists based on established guidelines.
  • Evaluation of five LLMs (GPT-o3-mini, Gemini, DeepSeek R1, Claude, Perplexity) on February 16, 2025.
  • Independent grading of responses by three radiology experts using a majority-voting system (correct, partially correct, incorrect, refusal).
  • Statistical analysis including accuracy calculation, logistic regression, and inter-rater agreement (Cohen's Kappa).

Main Results:

  • GPT-o3-mini achieved the highest accuracy (75.83%), followed closely by Gemini (74.17%).
  • Claude had the highest rate of partially correct answers (17.5%), with GPT-o3-mini showing significantly higher odds of fully correct answers (p=0.02).
  • GPT-o3-mini demonstrated better identification of intentionally incorrect Lung-RADS questions, despite notable inter-rater disagreement.

Conclusions:

  • GPT-o3-mini exhibited the best overall accuracy but with significant inter-rater variability.
  • Both GPT-o3-mini and Gemini showed improved accuracy compared to their predecessors, yet inconsistencies persist.
  • The study underscores the continued need for expert radiologist oversight in utilizing LLMs for clinical applications due to ongoing issues like incorrect responses and refusals.