• AI-assisted chest radiograph i...

Treffer: AI-assisted chest radiograph interpretation enhances diagnostic confidence and standardizes diagnostic accuracy across radiologists: A multi-reader study.

Title:
AI-assisted chest radiograph interpretation enhances diagnostic confidence and standardizes diagnostic accuracy across radiologists: A multi-reader study.
Authors:
Huang HY; Department of Radiology, Tri-Service General Hospital and National Defense Medical University, 325, Section 2, Cheng-Gong Rd., NeiHu, Taipei 114, Taiwan., Huang YH; Department of Radiology, Tri-Service General Hospital and National Defense Medical University, 325, Section 2, Cheng-Gong Rd., NeiHu, Taipei 114, Taiwan., Lin CH; Department of Radiology, Tri-Service General Hospital and National Defense Medical University, 325, Section 2, Cheng-Gong Rd., NeiHu, Taipei 114, Taiwan., Tao WT; Department of Radiology, Tri-Service General Hospital and National Defense Medical University, 325, Section 2, Cheng-Gong Rd., NeiHu, Taipei 114, Taiwan., Liao WC; Department of Radiology, Tri-Service General Hospital and National Defense Medical University, 325, Section 2, Cheng-Gong Rd., NeiHu, Taipei 114, Taiwan., Yu S; AI Lab, Quanta Computer lnc., No. 211, Wen-Hua 2nd Rd., Guishan Dist., Taoyuan City, 333, Taiwan., Mo HC; AI Lab, Quanta Computer lnc., No. 211, Wen-Hua 2nd Rd., Guishan Dist., Taoyuan City, 333, Taiwan., Feng W; AI Lab, Quanta Computer lnc., No. 211, Wen-Hua 2nd Rd., Guishan Dist., Taoyuan City, 333, Taiwan., Hsu YT; AI Lab, Quanta Computer lnc., No. 211, Wen-Hua 2nd Rd., Guishan Dist., Taoyuan City, 333, Taiwan., Wang JC; AI Lab, Quanta Computer lnc., No. 211, Wen-Hua 2nd Rd., Guishan Dist., Taoyuan City, 333, Taiwan., Ko KH; Department of Radiology, Tri-Service General Hospital and National Defense Medical University, 325, Section 2, Cheng-Gong Rd., NeiHu, Taipei 114, Taiwan. Electronic address: m860818@gmail.com.
Source:
Clinical imaging [Clin Imaging] 2026 Feb; Vol. 130, pp. 110694. Date of Electronic Publication: 2025 Dec 07.
Publication Type:
Journal Article
Language:
English
Journal Info:
Publisher: Elsevier Country of Publication: United States NLM ID: 8911831 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1873-4499 (Electronic) Linking ISSN: 08997071 NLM ISO Abbreviation: Clin Imaging Subsets: MEDLINE
Imprint Name(s):
Original Publication: [New York, NY] : Elsevier, [c1989-
MeSH Terms:
Radiography, Thoracic*/methods , Radiography, Thoracic*/standards , Artificial Intelligence* , Radiographic Image Interpretation, Computer-Assisted*/methods , Lung Diseases*/diagnostic imaging, Humans ; Retrospective Studies ; Female ; Sensitivity and Specificity ; Male ; Clinical Competence ; Middle Aged ; Observer Variation ; Radiologists ; Aged ; Reproducibility of Results ; Adult
Contributed Indexing:
Keywords: Artificial intelligence; Chest radiography; Computer-aided detection (CAD); Diagnostic accuracy; Radiologist; Resident
Entry Date(s):
Date Created: 20251212 Date Completed: 20260119 Latest Revision: 20260119
Update Code:
20260120
DOI:
10.1016/j.clinimag.2025.110694
PMID:
41386116
Database:
MEDLINE

Weitere Informationen

Purpose: To evaluate the impact of an artificial intelligence (AI)-assisted computer-aided detection (CAD) system on the diagnostic accuracy and confidence in chest radiograph interpretation among nonthoracic radiologists and radiology residents with varying levels of experience.
Methods: In this retrospective multiple-reader, multiple-case (MRMC) study, 400 chest radiographs (100 each for pulmonary nodules, pleural effusion, pneumothorax, and controls) were independently interpreted by 12 readers (two nonthoracic radiologists, four senior residents, and six junior residents). Readings were conducted under CAD-assisted and unassisted conditions, with a 30-day washout period. Readers assigned confidence scores (0-100) to their diagnosis. Diagnostic performance was evaluated using the area under the curve (AUC), sensitivity, and specificity, while reader confidence was assessed by the proportion of high-confidence ratings among correctly interpreted cases.
Results: The AI-assisted CAD system improved diagnostic performance across all abnormalities, with significant gains for pulmonary nodules (AUC: 0.781 → 0.854; P < 0.001) and pleural effusion (0.896 → 0.948; P < 0.001). The sensitivity increased by 7.2% for effusion, while the specificity for nodules improved markedly by 15.7%. Among all the readers, junior residents showed the greatest gains, especially for nodules, where the CAD closed their baseline AUC gap (originally -7.3%, P = 0.006) relative to nonthoracic radiologists. Reader confidence also increased significantly with the CAD, particularly for nodules (+15.2 %; P < 0.001).
Conclusion: The AI-assisted CAD system significantly enhanced diagnostic accuracy and reader confidence in chest radiograph interpretation, especially for junior radiology residents. This approach may bridge experience-related diagnostic gaps and support clinical decision-making, particularly in institutions lacking thoracic radiologists.
(Copyright © 2025 Elsevier Inc. All rights reserved.)

Declaration of competing interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.