GIT-CXR: End-to-End Transformer for Chest X-Ray Report Generation
Open Access
- 23 June 2025
- journal article
- research article
- Published by MDPI AG in Information
- Vol. 16 (7) , 524
Abstract
Medical imaging is crucial for diagnosing, monitoring, and treating medical conditions. The medical reports of radiology images are the primary medium through which medical professionals can attest to their findings, but their writing is time-consuming and requires specialized clinical expertise. Therefore, the automated generation of radiography reports has the potential to improve and standardize patient care and significantly reduce the workload of clinicians. Through our work, we have designed and evaluated an end-to-end transformer-based method to generate accurate and factually complete radiology reports for X-ray images. Additionally, we are the first to introduce curriculum learning for end-to-end transformers in medical imaging and demonstrate its impact in obtaining improved performance. The experiments were conducted using the MIMIC-CXR-JPG database, the largest available chest X-ray dataset. The results obtained are comparable with the current state of the art on the natural language generation (NLG) metrics BLEU and ROUGE-L, while setting new state-of-the-art results on F1 examples-averaged F1-macro and F1-micro metrics for clinical accuracy and on the METEOR metric widely used for NLG.Keywords
All Related Versions
Funding Information
- “Romanian Hub for Artificial Intelligence-HRIA”, Smart Growth, Digitization and Financial Instruments Program (334906)
This publication has 31 references indexed in Scilit:
- Radiology report generation with medical knowledge and multilevel image-report alignment: A new method and its verificationArtificial Intelligence in Medicine, 2023
- Improving chest X-ray report generation by leveraging warm startingArtificial Intelligence in Medicine, 2023
- GRIT: Faster and Better Image Captioning Transformer Using Dual Visual Features2022
- Automated Radiographic Report Generation Purely on Transformer: A Multicriteria Supervised ApproachIEEE Transactions on Medical Imaging, 2022
- Curriculum learning for improved femur fracture classification: Scheduling data with prior knowledge and uncertaintyMedical Image Analysis, 2022
- MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reportsScientific Data, 2019
- Automatic CNN-based detection of cardiac MR motion artefacts using k-space data augmentation and curriculum learningMedical Image Analysis, 2019
- Preparing a collection of radiology examinations for distribution and retrievalJournal of the American Medical Informatics Association, 2015
- Long Short-Term MemoryNeural Computation, 1997
- A STATISTICAL INTERPRETATION OF TERM SPECIFICITY AND ITS APPLICATION IN RETRIEVALJournal of Documentation, 1972