Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation
Pituitary adenoma is a common neuroendocrine neoplasm, and most of its MR images are characterized by blurred edges, high noise and similar to surrounding normal tissues. Therefore, it is extremely difficult to accurately locate and outline the lesion of pituitary adenoma. To sovle these limitations...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIMS Press
2023-01-01
|
| Series: | Mathematical Biosciences and Engineering |
| Subjects: | |
| Online Access: | https://www.aimspress.com/article/doi/10.3934/mbe.2023003?viewType=HTML |
| _version_ | 1798031095408623616 |
|---|---|
| author | Xiaoliang Jiang Junjian Xiao Qile Zhang Lihui Wang Jinyun Jiang Kun Lan |
| author_facet | Xiaoliang Jiang Junjian Xiao Qile Zhang Lihui Wang Jinyun Jiang Kun Lan |
| author_sort | Xiaoliang Jiang |
| collection | DOAJ |
| description | Pituitary adenoma is a common neuroendocrine neoplasm, and most of its MR images are characterized by blurred edges, high noise and similar to surrounding normal tissues. Therefore, it is extremely difficult to accurately locate and outline the lesion of pituitary adenoma. To sovle these limitations, we design a novel deep learning framework for pituitary adenoma MRI image segmentation. Under the framework of U-Net, a newly cross-layer connection is introduced to capture richer multi-scale features and contextual information. At the same time, full-scale skip structure can reasonably utilize the above information obtained by different layers. In addition, an improved inception-dense block is designed to replace the classical convolution layer, which can enlarge the effectiveness of the receiving field and increase the depth of our network. Finally, a novel loss function based on binary cross-entropy and Jaccard losses is utilized to eliminate the problem of small samples and unbalanced data. The sample data were collected from 30 patients in Quzhou People's Hospital, with a total of 500 lesion images. Experimental results show that although the amount of patient sample is small, the proposed method has better performance in pituitary adenoma image compared with existing algorithms, and its Dice, Intersection over Union (IoU), Matthews correlation coefficient (Mcc) and precision reach 88.87, 80.67, 88.91 and 97.63%, respectively. |
| first_indexed | 2024-04-11T19:50:47Z |
| format | Article |
| id | doaj.art-d399256452014a4c9dee4fa9da384f25 |
| institution | Directory Open Access Journal |
| issn | 1551-0018 |
| language | English |
| last_indexed | 2024-04-11T19:50:47Z |
| publishDate | 2023-01-01 |
| publisher | AIMS Press |
| record_format | Article |
| series | Mathematical Biosciences and Engineering |
| spelling | doaj.art-d399256452014a4c9dee4fa9da384f252022-12-22T04:06:19ZengAIMS PressMathematical Biosciences and Engineering1551-00182023-01-01201345110.3934/mbe.2023003Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentationXiaoliang Jiang0Junjian Xiao1Qile Zhang 2Lihui Wang3Jinyun Jiang 4Kun Lan51. College of Mechanical Engineering, Quzhou University, Quzhou 324000, China1. College of Mechanical Engineering, Quzhou University, Quzhou 324000, China2. Department of Rehabilitation, The Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People's Hospital, Quzhou 324000, China3. Department of Science and Education, The Quzhou Affiliated Hospital of Wenzhou Medical University, Quzhou People's Hospital, Quzhou 324000, China1. College of Mechanical Engineering, Quzhou University, Quzhou 324000, China1. College of Mechanical Engineering, Quzhou University, Quzhou 324000, ChinaPituitary adenoma is a common neuroendocrine neoplasm, and most of its MR images are characterized by blurred edges, high noise and similar to surrounding normal tissues. Therefore, it is extremely difficult to accurately locate and outline the lesion of pituitary adenoma. To sovle these limitations, we design a novel deep learning framework for pituitary adenoma MRI image segmentation. Under the framework of U-Net, a newly cross-layer connection is introduced to capture richer multi-scale features and contextual information. At the same time, full-scale skip structure can reasonably utilize the above information obtained by different layers. In addition, an improved inception-dense block is designed to replace the classical convolution layer, which can enlarge the effectiveness of the receiving field and increase the depth of our network. Finally, a novel loss function based on binary cross-entropy and Jaccard losses is utilized to eliminate the problem of small samples and unbalanced data. The sample data were collected from 30 patients in Quzhou People's Hospital, with a total of 500 lesion images. Experimental results show that although the amount of patient sample is small, the proposed method has better performance in pituitary adenoma image compared with existing algorithms, and its Dice, Intersection over Union (IoU), Matthews correlation coefficient (Mcc) and precision reach 88.87, 80.67, 88.91 and 97.63%, respectively.https://www.aimspress.com/article/doi/10.3934/mbe.2023003?viewType=HTMLpituitary adenomau-netcross-layer connectionimage segmentationinception-dense |
| spellingShingle | Xiaoliang Jiang Junjian Xiao Qile Zhang Lihui Wang Jinyun Jiang Kun Lan Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation Mathematical Biosciences and Engineering pituitary adenoma u-net cross-layer connection image segmentation inception-dense |
| title | Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation |
| title_full | Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation |
| title_fullStr | Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation |
| title_full_unstemmed | Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation |
| title_short | Improved U-Net based on cross-layer connection for pituitary adenoma MRI image segmentation |
| title_sort | improved u net based on cross layer connection for pituitary adenoma mri image segmentation |
| topic | pituitary adenoma u-net cross-layer connection image segmentation inception-dense |
| url | https://www.aimspress.com/article/doi/10.3934/mbe.2023003?viewType=HTML |
| work_keys_str_mv | AT xiaoliangjiang improvedunetbasedoncrosslayerconnectionforpituitaryadenomamriimagesegmentation AT junjianxiao improvedunetbasedoncrosslayerconnectionforpituitaryadenomamriimagesegmentation AT qilezhang improvedunetbasedoncrosslayerconnectionforpituitaryadenomamriimagesegmentation AT lihuiwang improvedunetbasedoncrosslayerconnectionforpituitaryadenomamriimagesegmentation AT jinyunjiang improvedunetbasedoncrosslayerconnectionforpituitaryadenomamriimagesegmentation AT kunlan improvedunetbasedoncrosslayerconnectionforpituitaryadenomamriimagesegmentation |