An attentive and adaptive 3D CNN for automatic pulmonary nodule detection in CT image

作者：

Highlights：

• An adaptive and attentive 3D CNN is proposed for pulmonary nodule detection.

• High resolution features are fused to improve the performance of nodule detection.

• An attentive module is designed to scratch the global and spatial information.

• Adaptive conv-kernels are proposed to encode the multilevel contextual information.

摘要

•An adaptive and attentive 3D CNN is proposed for pulmonary nodule detection.•High resolution features are fused to improve the performance of nodule detection.•An attentive module is designed to scratch the global and spatial information.•Adaptive conv-kernels are proposed to encode the multilevel contextual information.

论文关键词：Pulmonary nodule detection,3D Convolutional Neural Network (3D CNN),False positive reduction,Computer-aided diagnosis

论文评审过程：Received 10 April 2022, Revised 1 August 2022, Accepted 21 August 2022, Available online 28 August 2022, Version of Record 2 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118672