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 Videos

An Explainable Transformer-Based Framework for Lung Cancer Classification and Automated Radiology Report Generation

Oguzhan Katar1,2, Tulin Akbalik3, Ozal Yildirim2

  • 1Graduate School of Natural and Applied Sciences, Firat University, Elazig 23119, Turkey.

Biomedicines
|May 27, 2026
PubMed
Summary

Related Concept Videos

Computed Tomography01:10

Computed Tomography

Tomography refers to imaging by sections. Computed tomography (CT) is a non-invasive imaging technique that uses computers to analyze several cross-sectional X-rays to reveal minute details about structures in the body.
The technique was invented in the 1970s and is based on the principle that as X-rays pass through the body, they are absorbed or reflected at different levels. In the technique, a patient lies on a motorized platform while a computerized axial tomography (CAT) scanner rotates...

You might also read

Related Articles

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

Sort by
Same author

Early detection of colorectal cancer using a hybrid model with enhanced image quality and optimized classification.

Physical and engineering sciences in medicine·2025
Same author

A Novel Hybrid Model for Automatic Non-Small Cell Lung Cancer Classification Using Histopathological Images.

Diagnostics (Basel, Switzerland)·2024
Same author

Comparison of Explainable Artificial Intelligence Model and Radiologist Review Performances to Detect Breast Cancer in 752 Patients.

Journal of ultrasound in medicine : official journal of the American Institute of Ultrasound in Medicine·2024
Same author

An Explainable Vision Transformer Model Based White Blood Cells Classification and Localization.

Diagnostics (Basel, Switzerland)·2023
Same author

An Explainable Deep Learning Model to Prediction Dental Caries Using Panoramic Radiograph Images.

Diagnostics (Basel, Switzerland)·2023
Same author

An Automated Wavelet-Based Sleep Scoring Model Using EEG, EMG, and EOG Signals with More Than 8000 Subjects.

International journal of environmental research and public health·2022
This summary is machine-generated.

This study presents an AI framework for lung cancer detection and reporting from CT scans, achieving high accuracy and aiding clinical decisions. The explainable model supports early lung cancer assessment.

Area of Science:

  • Artificial Intelligence in Medical Imaging
  • Radiology and Diagnostic Imaging
  • Computational Pathology

Background:

  • Lung cancer is a leading cause of mortality, with early detection crucial but challenging.
  • Manual interpretation of CT scans for lung cancer suffers from workload and inter-observer variability.
  • Automated analysis of CT scans is needed to improve diagnostic accuracy and efficiency.

Purpose of the Study:

  • To develop an explainable AI framework for classifying lung cancer subtypes (small-cell, non-small-cell, normal) from CT images.
  • To enable automated radiology report generation from CT scans.
  • To improve the transparency and reliability of AI in lung cancer diagnosis.

Main Methods:

  • A multi-slice transformer-based framework utilizing a ViT encoder and GPT-2 decoder was developed.
Keywords:
explainabilitylung cancermulti-slice CTradiology report generationvision transformer

Related Experiment Videos

  • A Learnable Query Attention Pooling (LQAP) mechanism was used for patient-level representation.
  • Explainability was achieved through slice-wise Grad-CAM maps highlighting decision-making cues.

    • Main Results:

    • The model achieved 97.40% accuracy in Turkish and 94.81% in English for lung cancer classification.
    • Generated automated reports showed strong alignment with human-written reports based on BLEU, ROUGE, METEOR, and CIDEr metrics.
    • The LungCA dataset, comprising 767 patients, was used for model validation.

    Conclusions:

    • The proposed multi-slice transformer framework demonstrates robust performance in lung cancer classification and report generation.
    • The AI solution enhances transparency and effectively supports clinical workflows for lung cancer assessment.
    • This explainable AI offers a promising tool for early and accurate lung cancer detection.