Related Experiment Video
Updated: Dec 9, 2025
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
Multiview framework using a 3D residual network for pulmonary micronodule malignancy risk classification
Yujie Yang1,2, Qianqian Zhang1,2
1Institute of Computer and Information Engineering, Henan Normal University, Henan Province, Xinxiang, China.
Background:
Pulmonary micronodules account for 80% of all lung nodules. Generally, pulmonary micronodules in the early stages can be detected on thoracic computed tomography (CT) scans. Early diagnosis is crucial for improving the patient's survival rate.
Objective:
This paper aims to estimate the malignancy risk of pulmonary micronodules and potentially improve the survival rate.
Methods:
We extract 3D features of the CT images to obtain richer characteristics. Because superior performance can be achieved by having deep layers, we apply a 3D residual network (3D-ResNet) to classify the pulmonary micronodule. We construct a framework by using three parallel ResNets whose inputs are CT images in different regions of interest, i.e., the multiview of the image. To further evaluate the applicability of the framework, we make a five-category classification and achieve good performance.
Results:
By fusing different characteristics from three views, we achieve the area under the receiver operating characteristic curve (AUC) of 0.9681. Based on the results of the experiments, our 3D-ResNet has a better performance than 3D-VGG and 3D-Inception in terms of precision (the increase rates are 13.7% and 7.4%), AUC (the increase rates are 15.8% and 5.3%), and accuracy (the increase rates are 14.3% and 4.5%). Meanwhile, the recall performance is close to that of the 3D-Inception network.
Conclusion:
Overall, the framework we propose has applicability and feasibility in pulmonary micronodule classification.
Insights
This study developed a 3D-ResNet framework to classify pulmonary micronodules from CT scans, achieving high accuracy. Early detection of these nodules can significantly improve patient survival rates.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Pulmonary micronodules constitute 80% of lung nodules.
- Early detection via thoracic computed tomography (CT) scans is vital for survival.
Purpose of the Study:
- To estimate the malignancy risk of pulmonary micronodules.
- To improve patient survival rates through accurate classification.
Main Methods:
- Extracted 3D features from CT images for enhanced characterization.
- Employed a 3D residual network (3D-ResNet) for pulmonary micronodule classification.
- Utilized a multi-view framework with three parallel ResNets for robust analysis.
Main Results:
- Achieved an area under the receiver operating characteristic curve (AUC) of 0.9681 by fusing multi-view image characteristics.
- Demonstrated superior performance of 3D-ResNet over 3D-VGG and 3D-Inception in precision, AUC, and accuracy.
- Reported recall performance comparable to 3D-Inception.
Conclusions:
- The proposed framework is applicable and feasible for pulmonary micronodule classification.
- This approach holds potential for improving early diagnosis and patient outcomes.
