Related Experiment Videos
Dual-domain adversarial learning and feature constraint for unsupervised anomaly detection in chest X-rays
Na Liang1,2, Yehong Tong1, Xing Zhang1
1College of Computer Science, Chongqing University, Chongqing, China.
Frontiers in Medicine
|August 15, 2026
Summary
This study introduces DualA-AD, an unsupervised anomaly detection method for chest X-rays, improving early lung abnormality detection using only normal images. DualA-AD enhances screening accuracy for thoracic abnormalities, aiding early diagnosis and management.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Radiology
Background:
- Chest X-rays are crucial for early pulmonary disease detection.
- Unsupervised anomaly detection using only normal chest X-rays is vital due to scarce abnormal labels.
- Existing reconstruction-based methods struggle with global information, multi-layer analysis, and anomaly scoring.
Purpose of the Study:
- To develop an improved reconstruction-based unsupervised anomaly detection method for chest X-rays.
- To enhance the screening of thoracic abnormalities for earlier detection of lung lesions.
- To address limitations in spatial domain reliance, feature consistency, and anomaly score balancing.
Main Methods:
- Proposed DualA-AD architecture incorporating Dual-domain Adversarial Learning (DAL), Feature Constraint Module (FCM), and Distributed Anomaly Score (DAS).
- DAL integrates spatial and frequency domains for dual-domain adversarial reconstruction, capturing global and detailed information.
- FCM ensures feature consistency via multi-layer and multi-image analysis; DAS quantifies deviations from normal distribution for enhanced discriminability.
Main Results:
- DualA-AD achieved superior performance on three public chest X-ray datasets, outperforming existing methods.
- Significant improvements in Area Under the Curve (AUC), Accuracy (ACC), and F1 scores were observed.
- AUC improvements over state-of-the-art (SOTA) reached 0.93%, 0.77%, and 0.48% on different datasets.
Conclusions:
- DualA-AD effectively improves the accuracy of distinguishing normal from abnormal chest X-rays.
- The method offers an efficient, low-cost solution for intelligent detection of lung abnormalities.
- DualA-AD holds significant potential for computer-aided diagnosis and early management of pulmonary disorders.
Related Concept Videos
01:24X-ray Imaging
German physicist Wilhelm Röntgen (1845–1923) was experimenting with electrical current when he discovered that a mysterious and invisible "ray" would pass through his flesh but leave an outline of his bones on a screen coated with a metal compound. In 1895, Röntgen made the first durable record of the internal parts of a living human: an "X-ray" image (as it came to be called) of his wife’s hand. Scientists worldwide quickly began their own experiments with X-rays, and by 1900, X-ray was widely...
01:30Radiological Investigation I: X-ray and CT
Radiological investigations, including X-rays and computed tomography (CT) scans, are critical for diagnosing and evaluating various medical conditions. These imaging techniques provide valuable insights into the body's internal structures, aiding in the detection of abnormalities, assessment of disease progression, and development of treatment strategies. This article delves into two primary radiological investigations, chest X-rays and CT scans, outlining their purpose, procedures, and the...