Efficient Perineural Invasion Detection of Histopathological Images Using U-Net
Perineural invasion (PNI), a sign of poor diagnosis and tumor metastasis, is common in a variety of malignant tumors. The infiltrating patterns and morphologies of tumors vary by organ and histological diversity, making PNI detection difficult in biopsy, which must be performed manually by pathologi...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/10/1649 |
| _version_ | 1797500237820985344 |
|---|---|
| author | Youngjae Park Jinhee Park Gil-Jin Jang |
| author_facet | Youngjae Park Jinhee Park Gil-Jin Jang |
| author_sort | Youngjae Park |
| collection | DOAJ |
| description | Perineural invasion (PNI), a sign of poor diagnosis and tumor metastasis, is common in a variety of malignant tumors. The infiltrating patterns and morphologies of tumors vary by organ and histological diversity, making PNI detection difficult in biopsy, which must be performed manually by pathologists. As the diameters of PNI nerves are measured on a millimeter scale, the PNI region is extremely small compared to the whole pathological image. In this study, an efficient deep learning-based method is proposed for detecting PNI regions in multiple types of cancers using only PNI annotations without detailed segmentation maps for each nerve and tumor cells obtained by pathologists. The key idea of the proposed method is to train the adopted deep learning model, U-Net, to capture the boundary regions where two features coexist. A boundary dilation method and a loss combination technique are proposed to improve the detection performance of PNI without requiring full segmentation maps. Experiments were conducted with various combinations of boundary dilation widths and loss functions. It is confirmed that the proposed method effectively improves PNI detection performance from 0.188 to 0.275. Additional experiments were also performed on normal nerve detection to validate the applicability of the proposed method to the general boundary detection tasks. The experimental results demonstrate that the proposed method is also effective for general tasks, and it improved nerve detection performance from 0.511 to 0.693. |
| first_indexed | 2024-03-10T03:59:03Z |
| format | Article |
| id | doaj.art-74ae1e1943e5498aa802ede2c2c978a0 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T03:59:03Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-74ae1e1943e5498aa802ede2c2c978a02023-11-23T10:48:17ZengMDPI AGElectronics2079-92922022-05-011110164910.3390/electronics11101649Efficient Perineural Invasion Detection of Histopathological Images Using U-NetYoungjae Park0Jinhee Park1Gil-Jin Jang2School of Electronic and Electrical Engineering, Kyungpook National University, Daegu 41566, KoreaSchool of Electronic and Electrical Engineering, Kyungpook National University, Daegu 41566, KoreaSchool of Electronic and Electrical Engineering, Kyungpook National University, Daegu 41566, KoreaPerineural invasion (PNI), a sign of poor diagnosis and tumor metastasis, is common in a variety of malignant tumors. The infiltrating patterns and morphologies of tumors vary by organ and histological diversity, making PNI detection difficult in biopsy, which must be performed manually by pathologists. As the diameters of PNI nerves are measured on a millimeter scale, the PNI region is extremely small compared to the whole pathological image. In this study, an efficient deep learning-based method is proposed for detecting PNI regions in multiple types of cancers using only PNI annotations without detailed segmentation maps for each nerve and tumor cells obtained by pathologists. The key idea of the proposed method is to train the adopted deep learning model, U-Net, to capture the boundary regions where two features coexist. A boundary dilation method and a loss combination technique are proposed to improve the detection performance of PNI without requiring full segmentation maps. Experiments were conducted with various combinations of boundary dilation widths and loss functions. It is confirmed that the proposed method effectively improves PNI detection performance from 0.188 to 0.275. Additional experiments were also performed on normal nerve detection to validate the applicability of the proposed method to the general boundary detection tasks. The experimental results demonstrate that the proposed method is also effective for general tasks, and it improved nerve detection performance from 0.511 to 0.693.https://www.mdpi.com/2079-9292/11/10/1649deep learningU-Netboundary detectionperineural invasion detectionhistopathological image |
| spellingShingle | Youngjae Park Jinhee Park Gil-Jin Jang Efficient Perineural Invasion Detection of Histopathological Images Using U-Net Electronics deep learning U-Net boundary detection perineural invasion detection histopathological image |
| title | Efficient Perineural Invasion Detection of Histopathological Images Using U-Net |
| title_full | Efficient Perineural Invasion Detection of Histopathological Images Using U-Net |
| title_fullStr | Efficient Perineural Invasion Detection of Histopathological Images Using U-Net |
| title_full_unstemmed | Efficient Perineural Invasion Detection of Histopathological Images Using U-Net |
| title_short | Efficient Perineural Invasion Detection of Histopathological Images Using U-Net |
| title_sort | efficient perineural invasion detection of histopathological images using u net |
| topic | deep learning U-Net boundary detection perineural invasion detection histopathological image |
| url | https://www.mdpi.com/2079-9292/11/10/1649 |
| work_keys_str_mv | AT youngjaepark efficientperineuralinvasiondetectionofhistopathologicalimagesusingunet AT jinheepark efficientperineuralinvasiondetectionofhistopathologicalimagesusingunet AT giljinjang efficientperineuralinvasiondetectionofhistopathologicalimagesusingunet |