Wenju Wang, Shuya Yin, Fang Ye, Yinan Chen, Lin Zhu, Hong Yu
{"title":"GC-WIR : 3D global coordinate attention wide inverted ResNet network for pulmonary nodules classification.","authors":"Wenju Wang, Shuya Yin, Fang Ye, Yinan Chen, Lin Zhu, Hong Yu","doi":"10.1186/s12890-024-03272-7","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Currently, deep learning methods for the classification of benign and malignant lung nodules encounter challenges encompassing intricate and unstable algorithmic models, limited data adaptability, and an abundance of model parameters.To tackle these concerns, this investigation introduces a novel approach: the 3D Global Coordinated Attention Wide Inverted ResNet Network (GC-WIR). This network aims to achieve precise classification of benign and malignant pulmonary nodules, leveraging its merits of heightened efficiency, parsimonious parameterization, and robust stability.</p><p><strong>Methods: </strong>Within this framework, a 3D Global Coordinate Attention Mechanism (3D GCA) is designed to compute the features of the input images by converting 3D channel information and multi-dimensional positional cues. By encompassing both global channel details and spatial positional cues, this approach maintains a judicious balance between flexibility and computational efficiency. Furthermore, the GC-WIR architecture incorporates a 3D Wide Inverted Residual Network (3D WIRN), which augments feature computation by expanding input channels. This augmentation mitigates information loss during feature extraction, expedites model convergence, and concurrently enhances performance. The utilization of the inverted residual structure imbues the model with heightened stability.</p><p><strong>Results: </strong>Empirical validation of the GC-WIR method is performed on the LUNA 16 dataset, yielding predictions that surpass those generated by previous models. This novel approach achieves an impressive accuracy rate of 94.32%, coupled with a specificity of 93.69%. Notably, the model's parameter count remains modest at 5.76M, affording optimal classification accuracy.</p><p><strong>Conclusion: </strong>Furthermore, experimental results unequivocally demonstrate that, even under stringent computational constraints, GC-WIR outperforms alternative deep learning methodologies, establishing a new benchmark in performance.</p>","PeriodicalId":9148,"journal":{"name":"BMC Pulmonary Medicine","volume":null,"pages":null},"PeriodicalIF":2.6000,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11414275/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"BMC Pulmonary Medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s12890-024-03272-7","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"RESPIRATORY SYSTEM","Score":null,"Total":0}
引用次数: 0
Abstract
Purpose: Currently, deep learning methods for the classification of benign and malignant lung nodules encounter challenges encompassing intricate and unstable algorithmic models, limited data adaptability, and an abundance of model parameters.To tackle these concerns, this investigation introduces a novel approach: the 3D Global Coordinated Attention Wide Inverted ResNet Network (GC-WIR). This network aims to achieve precise classification of benign and malignant pulmonary nodules, leveraging its merits of heightened efficiency, parsimonious parameterization, and robust stability.
Methods: Within this framework, a 3D Global Coordinate Attention Mechanism (3D GCA) is designed to compute the features of the input images by converting 3D channel information and multi-dimensional positional cues. By encompassing both global channel details and spatial positional cues, this approach maintains a judicious balance between flexibility and computational efficiency. Furthermore, the GC-WIR architecture incorporates a 3D Wide Inverted Residual Network (3D WIRN), which augments feature computation by expanding input channels. This augmentation mitigates information loss during feature extraction, expedites model convergence, and concurrently enhances performance. The utilization of the inverted residual structure imbues the model with heightened stability.
Results: Empirical validation of the GC-WIR method is performed on the LUNA 16 dataset, yielding predictions that surpass those generated by previous models. This novel approach achieves an impressive accuracy rate of 94.32%, coupled with a specificity of 93.69%. Notably, the model's parameter count remains modest at 5.76M, affording optimal classification accuracy.
Conclusion: Furthermore, experimental results unequivocally demonstrate that, even under stringent computational constraints, GC-WIR outperforms alternative deep learning methodologies, establishing a new benchmark in performance.
期刊介绍:
BMC Pulmonary Medicine is an open access, peer-reviewed journal that considers articles on all aspects of the prevention, diagnosis and management of pulmonary and associated disorders, as well as related molecular genetics, pathophysiology, and epidemiology.