Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network
Abstract Endometrial cancer (EC) is the most common gynecological tumor in developed countries, and preoperative risk stratification is essential for personalized medicine. There have been several radiomics studies for noninvasive risk stratification of EC using MRI. Although tumor segmentation is u...
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Nature Portfolio
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/962db06ca6e14edf99d551abec3fc483 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:962db06ca6e14edf99d551abec3fc483 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:962db06ca6e14edf99d551abec3fc4832021-12-02T15:33:01ZAutomatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network10.1038/s41598-021-93792-72045-2322https://doaj.org/article/962db06ca6e14edf99d551abec3fc4832021-07-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-93792-7https://doaj.org/toc/2045-2322Abstract Endometrial cancer (EC) is the most common gynecological tumor in developed countries, and preoperative risk stratification is essential for personalized medicine. There have been several radiomics studies for noninvasive risk stratification of EC using MRI. Although tumor segmentation is usually necessary for these studies, manual segmentation is not only labor-intensive but may also be subjective. Therefore, our study aimed to perform the automatic segmentation of EC on MRI with a convolutional neural network. The effect of the input image sequence and batch size on the segmentation performance was also investigated. Of 200 patients with EC, 180 patients were used for training the modified U-net model; 20 patients for testing the segmentation performance and the robustness of automatically extracted radiomics features. Using multi-sequence images and larger batch size was effective for improving segmentation accuracy. The mean Dice similarity coefficient, sensitivity, and positive predictive value of our model for the test set were 0.806, 0.816, and 0.834, respectively. The robustness of automatically extracted first-order and shape-based features was high (median ICC = 0.86 and 0.96, respectively). Other high-order features presented moderate-high robustness (median ICC = 0.57–0.93). Our model could automatically segment EC on MRI and extract radiomics features with high reliability.Yasuhisa KurataMizuho NishioYusaku MoribataAki KidoYuki HimotoSatoshi OtaniKoji FujimotoMasahiro YakamiSachiko MinamiguchiMasaki MandaiYuji NakamotoNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-10 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
Medicine R Science Q |
| spellingShingle |
Medicine R Science Q Yasuhisa Kurata Mizuho Nishio Yusaku Moribata Aki Kido Yuki Himoto Satoshi Otani Koji Fujimoto Masahiro Yakami Sachiko Minamiguchi Masaki Mandai Yuji Nakamoto Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network |
| description |
Abstract Endometrial cancer (EC) is the most common gynecological tumor in developed countries, and preoperative risk stratification is essential for personalized medicine. There have been several radiomics studies for noninvasive risk stratification of EC using MRI. Although tumor segmentation is usually necessary for these studies, manual segmentation is not only labor-intensive but may also be subjective. Therefore, our study aimed to perform the automatic segmentation of EC on MRI with a convolutional neural network. The effect of the input image sequence and batch size on the segmentation performance was also investigated. Of 200 patients with EC, 180 patients were used for training the modified U-net model; 20 patients for testing the segmentation performance and the robustness of automatically extracted radiomics features. Using multi-sequence images and larger batch size was effective for improving segmentation accuracy. The mean Dice similarity coefficient, sensitivity, and positive predictive value of our model for the test set were 0.806, 0.816, and 0.834, respectively. The robustness of automatically extracted first-order and shape-based features was high (median ICC = 0.86 and 0.96, respectively). Other high-order features presented moderate-high robustness (median ICC = 0.57–0.93). Our model could automatically segment EC on MRI and extract radiomics features with high reliability. |
| format |
article |
| author |
Yasuhisa Kurata Mizuho Nishio Yusaku Moribata Aki Kido Yuki Himoto Satoshi Otani Koji Fujimoto Masahiro Yakami Sachiko Minamiguchi Masaki Mandai Yuji Nakamoto |
| author_facet |
Yasuhisa Kurata Mizuho Nishio Yusaku Moribata Aki Kido Yuki Himoto Satoshi Otani Koji Fujimoto Masahiro Yakami Sachiko Minamiguchi Masaki Mandai Yuji Nakamoto |
| author_sort |
Yasuhisa Kurata |
| title |
Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network |
| title_short |
Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network |
| title_full |
Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network |
| title_fullStr |
Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network |
| title_full_unstemmed |
Automatic segmentation of uterine endometrial cancer on multi-sequence MRI using a convolutional neural network |
| title_sort |
automatic segmentation of uterine endometrial cancer on multi-sequence mri using a convolutional neural network |
| publisher |
Nature Portfolio |
| publishDate |
2021 |
| url |
https://doaj.org/article/962db06ca6e14edf99d551abec3fc483 |
| work_keys_str_mv |
AT yasuhisakurata automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT mizuhonishio automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT yusakumoribata automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT akikido automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT yukihimoto automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT satoshiotani automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT kojifujimoto automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT masahiroyakami automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT sachikominamiguchi automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT masakimandai automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork AT yujinakamoto automaticsegmentationofuterineendometrialcanceronmultisequencemriusingaconvolutionalneuralnetwork |
| _version_ |
1718387155668041728 |