To address the low accuracy of traditional deep learning algorithms in medical image segmentation tasks and the unclear boundaries in the dataset of lung nodule images, an improved Unet network, named DM-Unet, was proposed. The model was enhanced through various optimization strategies to improve segmentation performance.First, a multi-scale residual convolution layer was employed to replace standard convolutional layers, enabling the network to capture feature information at different scales more effectively by increasing its depth. Second, in the decoder stage, a multi-level boundary enhancement modules was introduced. This module utilized upper-layer masks and an attention mechanism to enhance boundary recognition capabilities. Finally, during the skip connection stage, a detailed feature cascade module was embedded to strengthen the perception of multi-scale contextual features, and the final result was obtained through upsampling.Experimental results demonstrate that DM-Unet achieved superior performance on public pulmonary nodule image datasets, with IoU of 79.84%, Sensitivity of 87.33%, dice similarity coefficient of 88.62%, and positive predictive value of 90.19%, thereby fully validating the effectiveness of the proposed method.
ZHANG Wen-hao, YUAN Xiao-ping, HE Zhi-jie, WANG Zi-wei.
Lung Nodule Segmentation Algorithm Based on Multilevel Boundary Enhancement[J].
Science Technology and Engineering, 2026, 26(10): 4299-4308 DOI:10.12404/j.issn.1671-1815.2501991
CaoLei, LuLijun, YangRuimeng, et al. Adaptive morphologic segmentation of lung nodules based on regional growth[J]. Journal of Southern Medical University, 2008, 28(12): 2109-2112.
DongTing, WeiLong, NieShengdong. Progress in segmentation of lung nodules on CT images[J]. Chinese Journal of Image Graphics, 2021, 26(4): 751-765.
[5]
赵世轩. 基于视觉认知的医学图像处理算法与应用研究[D]. 成都: 电子科技大学, 2023.
[6]
ZhaoShixuan. Research on medical image processing algorithms and applications based on visual cognition[D]. Chengdu: University of Electronic Science and Technology, 2023.
[7]
LongJ, ShelhamerE, DarrellT. Fully convolutional networks for semantic segmentation[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Boston: IEEE, 2015: 3431-3440.
[8]
RonnebergerO, FischerP, BroxT. U-net: convolutional networks for biomedical image segmentation[C]// International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2015: 234-241.
WangLingling, LiuYuanlin. Research on medical image segmentation algorithm[J]. Information and Computer(Theoretical Edition), 2023, 35(21): 67-69.
[11]
XiaoX, LianS, LuoZ, et al. Weighted Res-Unet for high-quality retina vessel segmentation[C]// 9th International Conference on Information Technology in Medicine and Education (ITME 2018). Hangzhou: IEEE, 2018: 327-331.
YueYuchao, WangYingmei, QinJiachuan. CADCNet: an improved algorithm for retinal vessel segmention[J]. Science Techno-logy and Engineering, 2025, 25(3): 962-968.
[14]
ZhouZ, Rahman SiddiqueeM M, TajbakhshN, et al. Unet++: a nested U-net architecture for medical image segmentation[C]// International Workshop on Deep Learning in Medical Image Analysis. Cham: Springer International Publishing, 2018: 3-11.
[15]
MilletariF, NavabN, AhmadiS A. V-net: fully convolutional neural networks for volumetric medical image segmentation[C]// Fourth International Conference on 3D Vision 3DV. Cham: Springer, 2016: 565-571.
QiJiawen, LiuYi. Current status and prospect of graph neural network application in Chinese medicine[J]. Chinese Journal of Traditional Chinese Medicine, 2025(5): 1-12.
LiuYan, WangHongwu, AnPeng, et al. Exploring the clinical characteristics of different risks of pulmonary nodules in traditional Chinese medicine based on chest CT image-assisted system[J]. World Journal of Integrative Medicine, 2023, 18(12): 2513-2516.
ZhangXiaowei, LouYanmei, ZhouWeixiang, et al. Exploration of a machine learning based model for the identification of Yin-Yang and 25 hand-diagnosed body qualities[J]. Journal of Nanjing University of Traditional Chinese Medicine, 2024, 40(12): 1323-1330.
[22]
BadrinarayananV, KendallA, CipollaR. Segnet: a deep convolutional encoder-decoder architecture for image segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(12): 2481-2495.
[23]
LiangB, TangC, ZhangW, et al. U-Net: an UNet architecture with dual encoder for medical image segmentation[J]. Signal Image and Video Processing, 2023, 17(6): 3073-3081.
[24]
RahmanM M, MarculescuR. GCASCADE: efficient cascaded graph convolutional decoding for 2D medical image segmentation[C]// IEEE/CVF Winter Conference on Applications of Compu-ter Vision. New York: IEEE, 2024: 7713-7722.
[25]
AhmedM R, FahimM A I, IslamA K M M, et al. DOLG-NeXt: convolutional neural network with deep orthogonal fusion of local and global features for biomedical image segmentation[J]. Neurocomputing, 2023, 546.DOI: 10.1016/j.neucom.2023.126362.
[26]
DutandeP, BaidU, TalbarS. LNCDS: a 2D-3D cascaded CNN approach for lung nodule classification, detection and segmentation[J]. Biomedical Signal Processing Control, 2021, 67. DOI: 10.1016/j.bspc.2021.102527.
[27]
AgnesS A, AnithaJ. Efficient multiscale fully convolutional UNet model for segmentation of 3D lung nodule from CT image[J]. Journal of Medical Imaging, 2022, 9(5). DOI: 10.1117/1.JMI.9.5.052402.
[28]
GaldranA, BouchachiaH. Residual networks for pulmonary nodule segmentation and texture characterization[C]// International Conference on Image Analysis and Recognition. Berlin: Springer, 2020: 396-405
[29]
KeethaN V, BabuP, AnnavarapuC. U-net: a modified U-Net architecture with bidirectional feature network for lung nodule segmentation[J]. arXiv Preprint, arXiv: 2006.03762, 2020.
[30]
ChenJ N, LuY Y, YuQ H, et al. TransUNet: transformers make strong encoders for medical image segmentation[J]. arXiv Preprint, arXiv: 2102.04306, 2021.
HuangXiaoming, HeFuyun, TangXiaohu, et al. A review of research on the application of U-Net and its variants in medical image segmentation[J]. Chinese Journal of Biomedical Engineering, 2022, 41(5): 567-576.
YangMei, LuoJian, ZhangXiaoqian, et al. Image segmentation method for new crown pneumonia based on improved U-Net[J]. Journal of Chengdu Information Engineering University, 2023, 38(1): 44-48.
GuYu, LiuJiaqi, YangLidong, et al. Fusion of attention and dynamic convlution for computer aided diagnosis of pulmonary nodules[J]. Science Technology and Engineering, 2023, 23(16): 6834-6844.
KumarS N, BrunthaP M, DanielS I, et al. Lung nodule segmentation using unet[C]// 7th International Conference on Advanced Computing and Communication Systems (ICACCS). New York: IEEE, 2021: 420-424.
[40]
IsenseeF, JaegerP F, KohlS A A, et al. NNU-Net: a self-configuring method for deep learning-based biomedical image segmentation[J]. Nat Methods, 2021, 18(2): 203-211.