Based on Convolutional Neural Network Study of CT Images of Ovarian Cystadenoma
DOI:
https://doi.org/10.6911/WSRJ.202504_11(4).0010Keywords:
Medical image segmentation, U-net, Deep learning, ovarian cystadenoma.Abstract
Cystadenoma of the ovary is a disease that occurs in the ovary. In order to realise automatic segmentation of ovarian CT images, an improved U-net model based lesion segmentation method of ovarian cystadenoma CT images was proposed. VGG16 was used as the encoder to further simplify the structure of U-net model, and combined with CT image features for data enhancement. An ovarian CT image dataset was constructed based on clinical diagnosis data, and the model was trained and evaluated. The model achieved an Intersection over Union (IoU) of 88.85% on the test set, and an Area Under the Curve (AUC) of 99.72%, indicating the feasibility and accuracy of the improved U-net model for the segmentation of ovarian cystadenoma lesions. In comparison with the original U-net model, the proposed method can reduce the size of the model without compromising the accuracy of segmentation, and is more suitable for auxiliary clinical diagnosis.
Downloads
References
[1] Liu Tingting, Qian Yong, Wang Xuemei, et al. The clinical value of CT and MRI in the diagnosis of ovarian cystadenoma [J]. Current Medicine, 2019,26(30):74-75. (in Chinese
[2] Qiang Jinwei, Zhou Kangrong, Liao Zhihe, et al. CT diagnosis of cystadenoma of ovary [J]. Journal of Practical Radiology,2004,20(3):253-256. (in Chinese)
[3] Wu Guoliang, Wang Shuping, Fu Guangyin, et al. Diagnostic value of CT and MRI for cystadenoma of ovary [J]. Chinese Research on Maternal and Child Health,2017,28(7):893-896.
[4] Sun L Z. Clinical analysis of CT and MRI diagnosis of cystadenoma of ovary [J]. Imaging Research and Medical Applications, 21,5(3):135-136.
[5] Chen Yidong, Zhang Qin, LAN LAN, et al. Application of deep convolutional neural networks in medical image segmentation [J]. Chinese Journal of Health Information Management,202118(2): 278-284.
[6] Ronneberger O,Fischer P, Brox T. U-Net: Convolutionalnetworks for biomedical image segmentation [ C ]//International Conference on Medical lmage Computing and Computer-Assisted Intervention ,2015 :234 -241.
[7] Long J, Shelhamer E, Darrell T. Fully convolutional net-works for semantic segmentation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence ,2015,39(4):640-651.
[8] Simonyan K, Zisserman A. Very deep convolutional net-works for large-scale image recognition[EB]. arXiv: 14091556v1 ,2014.
[9] Glorot X, Bordes A, Bengio YS. Deep sparse rectifier neu-ral networks [J]. Journal of Machine Learning Research, 2011.15 :315 -323.
[10] Crum W R,Camara O, Hill DLG. Generalized overlapmeasures for evaluation and validation in sMedical image a-nalysis[ J]. EEE Transactions on Medical Imaging, 2006,25 :1451 -1461
[11] Salehi S, Erdogmus D, Gholipour A. Tversky loss function[19]for image segmentation using 3D fully convolutional deep net-works[C]//International Workshop on Machine Learning in Medical lmaging,2017:379 -387r
Downloads
Published
Issue
Section
License
Copyright (c) 2025 World Scientific Research Journal
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
