A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma
Abstract Background and Aims Endoscopic ultrasonography‐guided fine‐needle aspiration/biopsy (EUS‐FNA/B) is considered to be a first‐line procedure for the pathological diagnosis of pancreatic cancer owing to its high accuracy and low complication rate. The number of new cases of pancreatic ductal a...
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-08-01
|
| Series: | Cancer Medicine |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/cam4.6335 |
| _version_ | 1797242669191135232 |
|---|---|
| author | Xianzheng Qin Minmin Zhang Chunhua Zhou Taojing Ran Yundi Pan Yingjiao Deng Xingran Xie Yao Zhang Tingting Gong Benyan Zhang Ling Zhang Yan Wang Qingli Li Dong Wang Lili Gao Duowu Zou |
| author_facet | Xianzheng Qin Minmin Zhang Chunhua Zhou Taojing Ran Yundi Pan Yingjiao Deng Xingran Xie Yao Zhang Tingting Gong Benyan Zhang Ling Zhang Yan Wang Qingli Li Dong Wang Lili Gao Duowu Zou |
| author_sort | Xianzheng Qin |
| collection | DOAJ |
| description | Abstract Background and Aims Endoscopic ultrasonography‐guided fine‐needle aspiration/biopsy (EUS‐FNA/B) is considered to be a first‐line procedure for the pathological diagnosis of pancreatic cancer owing to its high accuracy and low complication rate. The number of new cases of pancreatic ductal adenocarcinoma (PDAC) is increasing, and its accurate pathological diagnosis poses a challenge for cytopathologists. Our aim was to develop a hyperspectral imaging (HSI)‐based convolution neural network (CNN) algorithm to aid in the diagnosis of pancreatic EUS‐FNA cytology specimens. Methods HSI images were captured of pancreatic EUS‐FNA cytological specimens from benign pancreatic tissues (n = 33) and PDAC (n = 39) prepared using a liquid‐based cytology method. A CNN was established to test the diagnostic performance, and Attribution Guided Factorization Visualization (AGF‐Visualization) was used to visualize the regions of important classification features identified by the model. Results A total of 1913 HSI images were obtained. Our ResNet18‐SimSiam model achieved an accuracy of 0.9204, sensitivity of 0.9310 and specificity of 0.9123 (area under the curve of 0.9625) when trained on HSI images for the differentiation of PDAC cytological specimens from benign pancreatic cells. AGF‐Visualization confirmed that the diagnoses were based on the features of tumor cell nuclei. Conclusions An HSI‐based model was developed to diagnose cytological PDAC specimens obtained using EUS‐guided sampling. Under the supervision of experienced cytopathologists, we performed multi‐staged consecutive in‐depth learning of the model. Its superior diagnostic performance could be of value for cytologists when diagnosing PDAC. |
| first_indexed | 2024-03-12T00:54:53Z |
| format | Article |
| id | doaj.art-0ee9c830c6f1417c9b96704fb8654fac |
| institution | Directory Open Access Journal |
| issn | 2045-7634 |
| language | English |
| last_indexed | 2024-04-24T18:42:53Z |
| publishDate | 2023-08-01 |
| publisher | Wiley |
| record_format | Article |
| series | Cancer Medicine |
| spelling | doaj.art-0ee9c830c6f1417c9b96704fb8654fac2024-03-27T09:11:00ZengWileyCancer Medicine2045-76342023-08-011216170051701710.1002/cam4.6335A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinomaXianzheng Qin0Minmin Zhang1Chunhua Zhou2Taojing Ran3Yundi Pan4Yingjiao Deng5Xingran Xie6Yao Zhang7Tingting Gong8Benyan Zhang9Ling Zhang10Yan Wang11Qingli Li12Dong Wang13Lili Gao14Duowu Zou15Department of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaShanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai ChinaShanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Pathology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaShanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai ChinaShanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Pathology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaDepartment of Gastroenterology Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University Shanghai ChinaAbstract Background and Aims Endoscopic ultrasonography‐guided fine‐needle aspiration/biopsy (EUS‐FNA/B) is considered to be a first‐line procedure for the pathological diagnosis of pancreatic cancer owing to its high accuracy and low complication rate. The number of new cases of pancreatic ductal adenocarcinoma (PDAC) is increasing, and its accurate pathological diagnosis poses a challenge for cytopathologists. Our aim was to develop a hyperspectral imaging (HSI)‐based convolution neural network (CNN) algorithm to aid in the diagnosis of pancreatic EUS‐FNA cytology specimens. Methods HSI images were captured of pancreatic EUS‐FNA cytological specimens from benign pancreatic tissues (n = 33) and PDAC (n = 39) prepared using a liquid‐based cytology method. A CNN was established to test the diagnostic performance, and Attribution Guided Factorization Visualization (AGF‐Visualization) was used to visualize the regions of important classification features identified by the model. Results A total of 1913 HSI images were obtained. Our ResNet18‐SimSiam model achieved an accuracy of 0.9204, sensitivity of 0.9310 and specificity of 0.9123 (area under the curve of 0.9625) when trained on HSI images for the differentiation of PDAC cytological specimens from benign pancreatic cells. AGF‐Visualization confirmed that the diagnoses were based on the features of tumor cell nuclei. Conclusions An HSI‐based model was developed to diagnose cytological PDAC specimens obtained using EUS‐guided sampling. Under the supervision of experienced cytopathologists, we performed multi‐staged consecutive in‐depth learning of the model. Its superior diagnostic performance could be of value for cytologists when diagnosing PDAC.https://doi.org/10.1002/cam4.6335artificial intelligencedeep learningendoscopic ultrasound‐guided fine‐needle aspirationneural network modelspancreatic ductal carcinoma |
| spellingShingle | Xianzheng Qin Minmin Zhang Chunhua Zhou Taojing Ran Yundi Pan Yingjiao Deng Xingran Xie Yao Zhang Tingting Gong Benyan Zhang Ling Zhang Yan Wang Qingli Li Dong Wang Lili Gao Duowu Zou A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma Cancer Medicine artificial intelligence deep learning endoscopic ultrasound‐guided fine‐needle aspiration neural network models pancreatic ductal carcinoma |
| title | A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma |
| title_full | A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma |
| title_fullStr | A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma |
| title_full_unstemmed | A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma |
| title_short | A deep learning model using hyperspectral image for EUS‐FNA cytology diagnosis in pancreatic ductal adenocarcinoma |
| title_sort | deep learning model using hyperspectral image for eus fna cytology diagnosis in pancreatic ductal adenocarcinoma |
| topic | artificial intelligence deep learning endoscopic ultrasound‐guided fine‐needle aspiration neural network models pancreatic ductal carcinoma |
| url | https://doi.org/10.1002/cam4.6335 |
| work_keys_str_mv | AT xianzhengqin adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT minminzhang adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT chunhuazhou adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT taojingran adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yundipan adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yingjiaodeng adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT xingranxie adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yaozhang adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT tingtinggong adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT benyanzhang adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT lingzhang adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yanwang adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT qinglili adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT dongwang adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT liligao adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT duowuzou adeeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT xianzhengqin deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT minminzhang deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT chunhuazhou deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT taojingran deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yundipan deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yingjiaodeng deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT xingranxie deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yaozhang deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT tingtinggong deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT benyanzhang deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT lingzhang deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT yanwang deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT qinglili deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT dongwang deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT liligao deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma AT duowuzou deeplearningmodelusinghyperspectralimageforeusfnacytologydiagnosisinpancreaticductaladenocarcinoma |