Brain tumor image segmentation based on improved FPN
Abstract Purpose Automatic segmentation of brain tumors by deep learning algorithm is one of the research hotspots in the field of medical image segmentation. An improved FPN network for brain tumor segmentation is proposed to improve the segmentation effect of brain tumor. Materials and methods Aim...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-10-01
|
| Series: | BMC Medical Imaging |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12880-023-01131-1 |
| _version_ | 1797636591004418048 |
|---|---|
| author | Haitao Sun Shuai Yang Lijuan Chen Pingyan Liao Xiangping Liu Ying Liu Ning Wang |
| author_facet | Haitao Sun Shuai Yang Lijuan Chen Pingyan Liao Xiangping Liu Ying Liu Ning Wang |
| author_sort | Haitao Sun |
| collection | DOAJ |
| description | Abstract Purpose Automatic segmentation of brain tumors by deep learning algorithm is one of the research hotspots in the field of medical image segmentation. An improved FPN network for brain tumor segmentation is proposed to improve the segmentation effect of brain tumor. Materials and methods Aiming at the problem that the traditional full convolutional neural network (FCN) has weak processing ability, which leads to the loss of details in tumor segmentation, this paper proposes a brain tumor image segmentation method based on the improved feature pyramid networks (FPN) convolutional neural network. In order to improve the segmentation effect of brain tumors, we improved the model, introduced the FPN structure into the U-Net structure, captured the context multi-scale information by using the different scale information in the U-Net model and the multi receptive field high-level features in the FPN convolutional neural network, and improved the adaptability of the model to different scale features. Results Performance evaluation indicators show that the proposed improved FPN model has 99.1% accuracy, 92% DICE rating and 86% Jaccard index. The performance of the proposed method outperforms other segmentation models in each metric. In addition, the schematic diagram of the segmentation results shows that the segmentation results of our algorithm are closer to the ground truth, showing more brain tumour details, while the segmentation results of other algorithms are smoother. Conclusions The experimental results show that this method can effectively segment brain tumor regions and has certain generalization, and the segmentation effect is better than other networks. It has positive significance for clinical diagnosis of brain tumors. |
| first_indexed | 2024-03-11T12:36:16Z |
| format | Article |
| id | doaj.art-0def319fe2bd438289085be17cb434f6 |
| institution | Directory Open Access Journal |
| issn | 1471-2342 |
| language | English |
| last_indexed | 2024-03-11T12:36:16Z |
| publishDate | 2023-10-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Imaging |
| spelling | doaj.art-0def319fe2bd438289085be17cb434f62023-11-05T12:32:27ZengBMCBMC Medical Imaging1471-23422023-10-0123111010.1186/s12880-023-01131-1Brain tumor image segmentation based on improved FPNHaitao Sun0Shuai Yang1Lijuan Chen2Pingyan Liao3Xiangping Liu4Ying Liu5Ning Wang6Department of Radiotherapy Room, Zhongshan Hospital of Traditional Chinese MedicineDepartment of Radiotherapy and Minimally Invasive Surgery, The Cancer Center of The Fifth Affiliated Hospital of Sun Yat-Sen UniversityDepartment of Radiotherapy Room, Zhongshan Hospital of Traditional Chinese MedicineDepartment of Radiotherapy Room, Zhongshan Hospital of Traditional Chinese MedicineDepartment of Radiotherapy Room, Zhongshan Hospital of Traditional Chinese MedicineDepartment of the Radiotherapy, The Fifth Affiliated Hospital of Guangzhou Medical UniversityDepartment of Radiotherapy Room, Zhongshan Hospital of Traditional Chinese MedicineAbstract Purpose Automatic segmentation of brain tumors by deep learning algorithm is one of the research hotspots in the field of medical image segmentation. An improved FPN network for brain tumor segmentation is proposed to improve the segmentation effect of brain tumor. Materials and methods Aiming at the problem that the traditional full convolutional neural network (FCN) has weak processing ability, which leads to the loss of details in tumor segmentation, this paper proposes a brain tumor image segmentation method based on the improved feature pyramid networks (FPN) convolutional neural network. In order to improve the segmentation effect of brain tumors, we improved the model, introduced the FPN structure into the U-Net structure, captured the context multi-scale information by using the different scale information in the U-Net model and the multi receptive field high-level features in the FPN convolutional neural network, and improved the adaptability of the model to different scale features. Results Performance evaluation indicators show that the proposed improved FPN model has 99.1% accuracy, 92% DICE rating and 86% Jaccard index. The performance of the proposed method outperforms other segmentation models in each metric. In addition, the schematic diagram of the segmentation results shows that the segmentation results of our algorithm are closer to the ground truth, showing more brain tumour details, while the segmentation results of other algorithms are smoother. Conclusions The experimental results show that this method can effectively segment brain tumor regions and has certain generalization, and the segmentation effect is better than other networks. It has positive significance for clinical diagnosis of brain tumors.https://doi.org/10.1186/s12880-023-01131-1Full convolutional neural networkU-Net modelImproved FPN modelBrain tumor segmentation |
| spellingShingle | Haitao Sun Shuai Yang Lijuan Chen Pingyan Liao Xiangping Liu Ying Liu Ning Wang Brain tumor image segmentation based on improved FPN BMC Medical Imaging Full convolutional neural network U-Net model Improved FPN model Brain tumor segmentation |
| title | Brain tumor image segmentation based on improved FPN |
| title_full | Brain tumor image segmentation based on improved FPN |
| title_fullStr | Brain tumor image segmentation based on improved FPN |
| title_full_unstemmed | Brain tumor image segmentation based on improved FPN |
| title_short | Brain tumor image segmentation based on improved FPN |
| title_sort | brain tumor image segmentation based on improved fpn |
| topic | Full convolutional neural network U-Net model Improved FPN model Brain tumor segmentation |
| url | https://doi.org/10.1186/s12880-023-01131-1 |
| work_keys_str_mv | AT haitaosun braintumorimagesegmentationbasedonimprovedfpn AT shuaiyang braintumorimagesegmentationbasedonimprovedfpn AT lijuanchen braintumorimagesegmentationbasedonimprovedfpn AT pingyanliao braintumorimagesegmentationbasedonimprovedfpn AT xiangpingliu braintumorimagesegmentationbasedonimprovedfpn AT yingliu braintumorimagesegmentationbasedonimprovedfpn AT ningwang braintumorimagesegmentationbasedonimprovedfpn |